Fall-related Hip Fracture

Transcript

Digital Comprehensive Summaries of Uppsala Dissertations from the Faculty of Medicine 1082 Fall-Related Hip Fracture Predisposing and Precipitating Factors BREIFFNI LEAVY ACTA UNIVERSITATIS UPSALIENSIS UPPSALA 2015 ISSN 1651-6206 ISBN 978-91-554-9202-1 urn:nbn:se:uu:diva-247286 Dissertation presented at Uppsala University to be publicly examined in Rosénsalen, Ingång 95/96, Akademiska Sjukhuset, Uppsala, Thursday, 7 May 2015 at 09:15 for the degree of Doctor of Philosophy (Faculty of Medicine). The examination will be conducted in English. Faculty examiner: Docent Ann Bremander (Department of clinical sciences/ Orthopedics, Lund University: Spenshults FoU-centrum, Halmstad). Abstract Leavy, B. 2015. Fall-Related Hip Fracture. Predisposing and Precipitating Factors. Digital Comprehensive Summaries of Uppsala Dissertations from the Faculty of Medicine 1082. 84 pp. Uppsala: Acta Universitatis Upsaliensis. ISBN 978-91-554-9202-1. A physically inactive lifestyle is a predisposing risk factor for fall-related hip fracture. The circumstances, or precipitating factors, surrounding hip fractures are, however, not well understood, a factor of relevance for Swedish adults who have one of the highest hip fracture risks in the world. The aims of this thesis, therefore, were: to explore perceptions of physical activity (PA) among older adults, to describe the circumstances surrounding hip fracture events and the health characteristics of those who experience them. Four observational studies were conducted involving qualitative, epidemiological and mixed method designs. Participants in study I were recruited from community settings in Stockholm and Dublin (n=30). Studies II-IV (sample sizes, n=484, n=125, n=477) were based on a population-based sample of people admitted to Uppsala University hospital due to hip fracture. Study IV also incorporated the background population of Uppsala county in 2010 (n=117 494). Analysis of PA perceptions in study I revealed that PA which is functional nature is perceived as most meaningful among certain participants. The uptake of PA in later years was a means of creating a new self-identify and being active in outdoor environments was an important culture-specific motivator to PA among Swedish participants. Analysis of hip fractures patterns in studies II-III showed that: hip fractures among psychotropic drug users were twice as likely to occur during night-time hours compared to those occurring among people not receiving these drugs. Additionally, the fall-related hip fractures of community dwellers with poorest health and function tended to occur indoors during positional changes. In study IV, all categories of disease (according to the International Classification of Diseases, 10th Revision) were seen to be positively associated with hip fracture. Cardiovascular disease and previous injury (including previous fracture) posed the highest relative and absolute fracture risks. Detailed investigation of hip fracture circumstances reveal patterns in health and functional characteristics, which provide information regarding predisposing and precipitating factors for these events. This knowledge, in combination with findings regarding PA perceptions, can be used when identifying individuals at high risk for hip fracture and when tailoring fracture prevention at an individual level to those at risk. Keywords: Co-existing disease, Epidemiology, Fall circumstances, Hip fracture, In-depth interviews, Mixed methods, Physical activity perceptions, Psychotropic drugs, Qualitative Breiffni Leavy, Department of Surgical Sciences, Akademiska sjukhuset, Uppsala University, SE-75185 Uppsala, Sweden. © Breiffni Leavy 2015 ISSN 1651-6206 ISBN 978-91-554-9202-1 urn:nbn:se:uu:diva-247286 (http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-247286) We have listened long enough to catastrophic demographic projections of how society will collapse under the expense of providing for all of us old people, it’s like nobody can think in terms of dynamic effects. Bodil Jönsson, Professor Emerita, Lund University List of Papers This thesis is based on the following papers, which are referred to in the text by their Roman numerals. I Leavy B, Åberg AC. ‘Not ready to throw in the towel’: Perceptions of Physical Activity held by Older Adults in Stockholm and Dublin. J Aging Phys Act, 2010 Apr; 18 (2):219-236. II Leavy B, Åberg AC, Melhus H, Mallmin H, Michaëlsson K, Byberg L. When and where do hip fractures occur? A population-based study Osteoporos Int. 2013 March; 24:2387–2396. III Leavy B, Byberg L, Michaëlsson K, Melhus H, Åberg AC. The Fall Descriptions and Health Characteristics of Older Adults with Hip Fracture: a Mixed Methods Study. BMC Geriatrics, Article in Press. IV Leavy B, Michaëlsson K, Åberg AC, Melhus H, Byberg L. The Impact of Disease on Hip Fracture Risk. Manuscript. Reprints were made with permission from the respective publishers. Contents Introduction ................................................................................................... 13  Physical activity in later life ..................................................................... 14  Epidemiology of hip fracture ................................................................... 15  Risk factors for hip fracture...................................................................... 16  Factors related to falling ...................................................................... 16  Factors related to the bone or impact on the bone ............................... 17  Environmental factors .......................................................................... 18  The importance of investigating different perspectives ........................... 19  Rationale for the current thesis ................................................................. 20  Aims .............................................................................................................. 21  Methods ........................................................................................................ 22  Naturalism using interview technique ...................................................... 22  Postpositivism and epidemiology ............................................................. 23  Mixed methods ......................................................................................... 23  Theoretical perspectives ........................................................................... 24  International Classification of Function, Disability and Health .......... 24  The fall risk triangle............................................................................. 24  Ethical considerations .............................................................................. 25  Participants ............................................................................................... 26  Study I.................................................................................................. 26  Study II and IV .................................................................................... 26  Study III ............................................................................................... 27  Data collection and analysis ..................................................................... 29  Study I − A qualitative study ............................................................... 29  Study II −A population-based cohort study ........................................ 29  Study III − A concurrent triangulation mixed method study .............. 31  Study IV − A population-based cohort study ...................................... 33  Findings ........................................................................................................ 35  Perceptions of physical activity (Study I) ................................................ 35  When and where hip fractures occur (Study II) ....................................... 36  Circumstances surrounding fall-related hip fractures (Study III)............. 38  Fall circumstances ............................................................................... 38  Patterns between fall circumstances and health characteristics ........... 39  The impact of disease on hip fracture risk (Study IV) ............................. 44  Methodological Considerations................................................................ 48  Mixed methods .................................................................................... 48  Relying on self-reports of fall circumstances ...................................... 48  Trustworthiness ................................................................................... 49  The use of single item questions to capture health data....................... 50  Separating the effects of disease and medication ................................ 51  General Discussion ....................................................................................... 52  Summary of results................................................................................... 52  The role played by outdoor environments for physical activity beliefs and hip fracture patterns ........................................................................... 52  Being outdoors as a motivating factor for physical activity ................ 52  Outdoor environments and the risk for hip fracture ............................ 53  Developing walking programs for fall prevention ............................... 54  Patterns in fall circumstances reflect the health and functional status of people with hip fracture............................................................................ 55  Night-time falls and the use of psychotropic medication .................... 55  Activity at the time of the fall and health status .................................. 56  Falls in outdoor environments and good perceived health .................. 57  Gender-related differences................................................................... 57  The impact of co-existing disease on hip fracture and physical activity beliefs ....................................................................................................... 58  Disease as a predisposing risk factor for hip fracture .......................... 58  Previous injury as a risk factor for hip fracture ................................... 59  Perceptions of the relevance of previous falls ..................................... 59  The impact of cardiovascular and other diseases................................. 60  Perceptions of disease in relation to falls and physical activity........... 61  Main Conclusions ......................................................................................... 63  Clinical implications ..................................................................................... 64  Acknowledgments......................................................................................... 66  References ..................................................................................................... 68  Abbreviations ATC BMD DXA FRID ICD-10 ICF IRR MMSE OR PA PPAR RR RD RCF WHO Anatomical Therapeutic Chemical Classification Bone mineral density Dual-energy X-ray absorptiometry Fall-risk-increasing drugs International Classification of Diseases, 10th Revision International Classification of Functioning, Disability and Health Incidence rate ratio Mini-mental state examination Odds ratio Physical activity Population proportional attributable risk Risk ratio Risk difference Residential care facility World Health Organisation The definition of concepts used in relation to health in the current thesis, are based on the nomenclature outlined in the International Classification of Functioning Disability and Health (ICF). A depiction of this health model can be seen in figure 1. Health Condition (disorder or disease) Body Functions and Structures Environmental Factors Activities Participation Personal Factors Figure 1. Schematic portrayal of the interactional components of the ICF model Definition of concepts Activity The execution of a task or an action, such as walking. Fall An unexpected event in which the person comes to rest on the ground, floor or a lower level.1 Fall-risk-increasing drugs A sub-group of drugs strongly associated with increased fall risk which includes drugs acting on the central nervous system (psychotropic drugs) as well as cardiovascular drugs with hypotensive effects. Geriatric syndrome A clinical condition (such as falling) in older people which does not fit into a distinct disease category, and is the result of the accumulated effects of multiple impairments.2 Health In studies II-IV we use ‘health’ when referring to the levels of co-existing disease and medication use. We use ‘overall health’ when referring to disease, medication status and functional capacity or performance (mobility, personal activities of daily living). Older adults There is no standard definition for older adult, but in westernized countries the most common chronological cut-off point is 65 years. In the current thesis, we refer to ‘older adults‘ primarily in relation to participants in study I. Physical activity Any bodily movement produced by the skeletal muscles that require energy expenditure.3 (Exercise is subset of physical activity involving a planned and structured body movement, performed with an aim of improving or maintaining fitness.) Physical inactivity Can be defined as not meeting any of the following criteria: 30 minutes of moderate intensity aerobic activity, at least 5 days/week or 20 minutes of vigorous activity on at least 3 days of the week.4 Recommended levels Moderate intensity aerobic activity for 30 minutes, 5 of physical activity days/week, plus (for adults ≥ 65 years) Moderate intensity muscle strengthening exercise 2−3 days/week, plus Flexibility exercise for 10 minute bouts 2 days/week.5 Disability An umbrella term which covers impairments, activity limitations and participation, which reflects the interaction between bodily and societal features. Participation Involvement in a life situation. Polypharmacy Defined in the current thesis as the use of 5 or more medications from any category. Psychotropic medications A group of drugs, such as benzodiazepines, antidepressants, anticonvulsants and narcotics, which by their effects on the central nervous system can impair alertness and neuromuscular function and increase the risk for falls and fractures.6 Predisposing risk factors Underlying risk factors which make an individual susceptible to disease or health outcomes, such as hip fracture. Precipitating risk factors Factors which are responsible for triggering the acute onset of a health outcome, such as a fall, as is the case in this thesis. Osteoporosis A systemic skeletal disease in which the density and quality of bone are reduced. Diagnosed in women on the basis of bone mineral density levels which lies 2.5 standard deviations below the mean value for a young white female (BMD T score ≤ 2.5). Introduction Physical activity is an effective way to improve functional capacity in later life, despite the presence of co-existing disease.7 Conversely, a physically inactive lifestyle, in combination with aging, can lead to muscle weakness and bone loss,8 commencing a negative health spiral for the older person. One such negative effect of muscle weakness in later life is an increased risk for falls, and physically inactive older people appear especially predisposed to falls and hip fractures.9-11 Hip fractures are among the most serious outcomes of falls among older people as they often incur detrimental effects on the functional capacity,12 independence13 and mortality14 of those affected. Previously, hip fracture prevention has focused mainly on the pharmaceutical treatment of bone,15 primarily using bisphosphonates, and reductions in fracture risk have been demonstrated in women who are younger than 80 years, with diagnosed osteoporosis, or in those with a history of fracture.16, 17 However, the majority of people with hip fracture do not have bone density levels which fall within the range defined as osteoporosis,18 and firm evidence is lacking for the efficacy of using bisphosphonates,19, 20 or other bone antiresorptive drugs,21 to prevent fracture in this group. In over 90 % of cases however, a hip fracture is preceded by a fall, but details surrounding these falls have scarcely been examined. Physical activity is the most effective single fall prevention strategy for older adults.22 It appears that up to one third of all falls in community settings can be prevented by exercise interventions23 and recent reports indicate that fall prevention can reduce fracture risk at levels comparable to that achievable through pharmaceutical intervention.24 In consideration of the established benefits of physical activity in later life, the promotion of physical activity among older people should be a public health priority. In order for such promotion to be effective however, an understanding of how older people perceive physical activity is required. This thesis focuses on physical beliefs and fall-related hip fracture occurrence, two domains which strongly influence health and wellbeing in the lives of older people. 13 Physical activity in later life That physical activity (PA) benefits health and the management of chronic disease among the general population is widely accepted in the literature.25, 26 Evidence from observational and intervention studies shows that PA can reduce the risk of: cardiovascular disease,27 type-2 diabetes,28 depression29 and mortality from cancer.30 The specific physical and psychological benefits of PA among older adults are also well documented.31 It is also of note that increased exercise uptake after the age of 50 years can lead to lower mortality rates.32 In consideration of the age-related declines in musculoskeletal strength8 and postural stability,33 maintaining an active lifestyle in later life may be especially motivated as evidence suggests that PA can offset these age-related declines in physical capacity.34 Neither does advanced age contraindicate participation in activities such as strength training,35 on the contrary, older adults with the lowest strength capacities prior to training have demonstrated the greatest improvements in trial situations.36 Additionally, weight-bearing exercise appears to have positive, although modest, effects on bone mass in postmenopausal women,37 making PA a recommended component in the avoidance and management of osteoporosis. These combined effects may therefore explain the reported associations between exercise interventions and reductions in the rate and risk of falls22, 23 and fractures.24 Despite the multitude of established health benefits, levels of PA are decreasing and older people are the least physically active members of society.4, 38 According to self-report, 60% of Swedish adults, aged 65–84 years meet the PA recommendations of 30 minutes of moderate intensity daily exercise.39 When more vigorous PA (such as swimming or tennis) is reported however, only 11% of 65–84 year olds are physically active at this level.40 When PA levels have been objectively assessed, on the other hand, even smaller proportions of older people are reported to meet the recommended PA levels.41, 42 The most common barriers to PA reported by older people include: health problems, fear of falling, inertia, as well as aspects of the physical environment, such as poor walkability and access to amenities.43-45 Culture-specific motivators and barriers to PA have also been highlighted among older populations.46, 47 A challenge facing PA promotion among older people is that uptake levels of PA programs are low and intervention studies aimed at improving physical function48 or reducing falls49 report high drop-out rates among older participants. 14 Epidemiology of hip fracture Hip fractures are classified by x-ray confirmation according to anatomical location. Fractures situated proximal to the femoral neck region are referred to as cervical (50 %) and the more distally located fracture types are defined as intertrochanteric (45 %) and sub-trochanteric (5%). Subsequent labelling of hip fractures occurs in reference to the degree of dislocation for cervical fractures, or instability for fractures of the trochanteric region. Hip fracture predominately affects older people, as the majority of fractures occur in people over the age of 70 years.50 Given the projected growth of the proportion of older people in the population, global hip fracture incidence is projected to rise from a rate of 1.6 million in 2000, to 6 million annually by the year 2050.51 The risk of incurring a hip fracture differs in relation to geographical location and ethnic identity, with 10-fold variations in hip fracture risk being reported for different countries across the world.52 Women have a lifetime risk of fracture which is approximately twice that of men53 and a 50-year-old Swedish woman has a 28 % lifetime risk for hip fracture, placing her amongst those at highest risk in the world.54 It also appears that falling incidence, bone mineral density and vitamin D levels, can only marginally account for the relatively higher occurrence of hip fracture among the older Swedish population.55-57 Additionally, older people living in residential care have up to twice the risk of incurring a hip fracture than elderly people who live in the community.58 In consideration that hip fracture primarily affects older people, many of whom already suffer from pre-existing health conditions,59 the impact of hip fracture on morbidity and mortality is often severe.60 Mortality rates of up to 33% are reported within the year following hip fracture, with variations dependant on age, gender and health status.59, 61, 62 Higher excess mortality is reported in men whose rate of death increases by 7-fold in the year following fracture, in comparison to women who have a 4fold increased excess mortality risk.14 Among those who survive the consequences of fracture, a substantial proportion of them experience reductions in functional capacity and independence.63 For instance, for every 100 people who experience a hip fracture, 26 of those who could walk independently before the event, cannot do so a year after the fracture.12 As a result of these reductions in mobility and functional independence, many older people are unable to return to independent living in the community, which makes hip fracture a frequent reason for admission into residential care facilities.61 There are also reports that mobility limitation and institutionalisation are more frequently occurring among men who fracture their hip than among women.64 15 The societal burden imposed by hip fracture is higher than all other fractures, as the total cost of caring for one female patient with hip fracture varies between 12 000−18 000 Euro annually, dependant on age at fracture.65 Estimations of the physical and financial burden imposed by hip fracture however, do not adequately account for the psychological effects incurred by these events. Fear of falling and depression are such two commonly occurring examples,66 which although not always immediately apparent, are known to influence both functional recovery and quality of life following the fracture.67, 68 Risk factors for hip fracture The risk factors for hip fracture have been studied extensively by prospective cohort studies,69-72 but categorization of risk factors is inconsistent in the literature. The most commonly referred to categories are, ‘intrinsic’ (referring to individual risk factors), and ‘extrinsic’ (referring to environmental risk factors). Risk factors for hip fractures can also be thought of as being predisposing− underlying factors which increase susceptibility to fracture (such as age, or physical activity level) or precipitating factors – factors which are responsible for the acute onset of the fall or fracture (such as acute physical symptoms). Regardless of the method of categorization, and owing to the fact that the vast majority of hip fractures occur in combination with a fall,73 the majority of established risk factors for hip fracture are also proven risk factors for falls.74 Falls are nevertheless much more frequently occurring than hip fractures in elderly populations as only an approximate 1−2 % of all falls are reported to result in fracture of the hip.75, 76 The interaction of other factors are therefore necessary in explaining why some falls result in fracture whereas others do not. The total number of clinical risk factors,69 precise fall-mechanisms and bone strength of the person who falls, will also influence whether the fall will result in hip fracture. Factors related to falling The British National Institute for Clinical Excellence (NICE) has identified up to 400 potential risk factors for falls among community-dwelling people.77 The most comprehensive method of classifying fall risk factors is proposed by Lord et al,78 who divided fall risk factors as follows: Sociodemographic, mobility and balance, sensory and neuromuscular, medical, drugrelated and environmental factors. Having a history of falls is one of the strongest predictors of falls and hip fracture among older among communitydwellers.69, 76 Factors indicative of impaired muscle function, such as impaired mobility, walking aid use and physical disabilities are also proven predictors for falls and subsequent fractures.69, 76, 79 The presence of diseases 16 affecting neuromuscular function, such as stroke or Parkinson’s disease are in turn risk factors for similar reasons.80 Other shared risk factors for falls and hip fracture among community-dwellers include: increased age, female sex, previous fractures, and balance impairment.81-83 76, 84 There is also evidence for the causal relationship between psychotropic drugs which act on the central nervous system such as benzodiazepines, anti-depressants and anti-convulsants, on falls and subsequent hip fracture.6, 85 Furthermore, if medications from multiple categories are used, the likelihood of falling increases.86 The relative effect of fall risk factors may differ in relation to place of residence, and factors which predispose to falls, as well as the impact of these factors, appears to differ between residential care and community settings.87 In many cases, risk factors can be thought to be inter-related in a fall situation, as is the case for example with psychotropic medications and their effect on postural control. Such medications have been shown to increase the risk for both falls and hip fracture, possibly due to their effects on the older person’s neuromuscular system, which in turn affects gait and postural reactions.85 Precise fall mechanisms, such as fall direction and velocity, are also known to influence the impact force transmitted to the greater trochanter and therefore the subsequent fracture risk. A sideways fall with direct hip impact, for example, is a very strong predictor of hip fracture, with risk increases between 6- and 20-fold being reported, dependent on health status.88-90 Similarly, whether or not a person manages to break the fall with an outstretched arm will also determine whether the impact energy from the fall can be dispersed.90, 91 Factors related to the bone or impact on the bone Bone mineral density (BMD) is a major determinant of hip fracture as for every standard deviation drop in BMD, there is a continuous (2.6-fold) increase in the risk for hip fracture.92, 93 The assessment of areal BMD by means of dual-energy X-ray absorptiometry (DXA) is the most widely used clinical measure of hip fracture risk.94 Hip fractures are, for this reason, often referred to in the literature as osteoporotic or fragility fractures, both terms which imply that an underlying weakened bone structure is influential in fracture causation. This nomenclature is somewhat controversial, as studies have shown that a minority of people with hip fracture have BMD values which can be defined as osteoporotic (BMD T score ≤ 2.5).15, 95, 96 BMD measurements, at the total hip, are reported to explain 28 % of the risk for hip fracture in women.97 When the cut-off point is raised in order to include those with less advanced bone density losses, (BMD T score ≤ 1.5) 51% of female hip fracture cases are reportedly explained.97 However, while those who have a low BMD may have a high risk for hip fracture, those who have 17 a high BMD do not necessarily have a low risk for hip fracture.98 The predictive ability of BMD measures also appears to weaken in line with increasing age,99 which is a factor of importance with late-onset fractures such as hip fracture. Another common classification of hip fractures in the literature is that of low-trauma fracture, a term which implies that the causative trauma was a force insufficient to break normal healthy bone, such as the force incurred from a fall from standing height.100 Neither can this classification be considered optimal, due to the fact that high trauma hip fractures have also been associated with reduced bone mineral density.101 The energy absorption of soft tissues surrounding the hip102 or the landing surface103, 104 are also factors which can affect the extent of impact to the hip during a sideways fall, which is weakest at this loading point.105 Environmental factors Environmental factors can contribute to hip fracture by influencing both fall risk, such as in tripping or slipping situations, or by affecting the force of the impact on the bone, which can vary in relation to the landing surface.103 The extent to which environmental hazards are reported to explain fall occurrence appears to vary in accordance with study design. Studies exploring older people’s perceptions, on the one hand, commonly describe environmental factors in fall causation,106, 107 whereas evidence from cohort studies appears inconsistent with regard to these associations.108-111 The transient nature of environmental factors and the mediating effects of functional capacity on exposure to, and the impact of, environmental factors on falls are both possible factors which may explain these disparities.108 With the exception of physical function, very little is reported about the presence of other health aspects, such as co-existing disease or medication status, among those who attribute their falls to environmental factors. Additionally, the traditional dichotomous division of fall risk factors as either extrinsic or intrinsic112114 fails to capture the role played by activity-related factors in the run-up to fall events. Evidence is also lacking concerning the extent to which place of residence, which in older ages is often a proxy for general health, or actual weather conditions influence patterns in hip fracture occurrence. 18 The importance of investigating different perspectives Fall-related hip fractures have multiple aetiologies and multi-faceted problems call for multiple means of inquiry. The ultimate aim of investigating the circumstances of fall-related hip fracture is to identify patterns which may then be susceptible to change. Similarly, estimating the impact of predisposing factors, such as disease, on hip fracture using epidemiological methods is a way to calculate risk, and in doing so identify individuals who are at high risk. Having identified those who are at-risk for falls or physical inactivity, the next step may often involve encouraging these individuals to change elements of their behavior, by for example becoming more physically active or by using a walking aid. It is however important to remember that the older person in question may not necessarily perceive themselves as being at-risk for negative health outcomes such as falls or fractures.115, 116 For these reasons, it is essential to build an understanding of how older people perceive the events which are being promoted or prevented. Findings from qualitative studies investigating physical activity and fall prevention beliefs give rise for concern in these areas. It has been reported, for example, that many physically inactive older people report ‘lack of interest’ or being ‘too old’ as barriers to exercise.117, 118 Additionally, it appears that even older people who have fallen previously tend to underestimate their risk of falling and disregard fall preventive advice as information not personally relevant to them.116, 119 Taking account of the preferences of the patient is an inherent element of patient-centered care, and may be especially important in the assessment and rehabilitation of older people who are faced with managing multiple health conditions.120 This thesis aims to take these factors into consideration and thereby help to fill gaps in the current literature. 19 Rationale for the current thesis In order to tailor exercise interventions which appeal to older people we need to understand the beliefs they hold in relation to physical activity. It is reported that many older people are unaware of the ways in which moderate intensity physical activity can benefit health117 or prevent falls.119, 121 Over half of those offered to join exercise programs refuse115 and it is reportedly the way in which exercise is promoted, which makes older people feel stigmatized and therefore disregard the information.122 Although there is extensive evidence for the risks posed by predisposing factors for hip fracture, risk estimates have often been calculated from study populations which have not accounted for the oldest old or the cognitively impaired. Similarly, there is a scarcity of studies which report the relative as well as the absolute risks posed by co-existing disease for hip fracture occurrence. Despite this fact, clinicians are advised to quote the absolute risks when informing patients about treatment options.123 To date, hip fracture research has focused on bone-related aspects whereas there is a scarcity of studies which investigate the where, when and how of the fall events which precede hip fractures in over 90 % of cases. It is therefore feasible to focus hip fracture prevention on the investigation of the circumstances surrounding the fall events which precede hip fracture. It is not known, for example, how factors such as health and residential status effect when hip fracture occurs and it is unclear whether season or actual weather conditions influence hip fracture patterns. Additionally, qualitative studies have highlighted the diversity in causal perceptions amongst older adults who fall and quantitative studies have described predisposing factors for hip fracture, but these factors have rarely been investigated in relation to each other. Placing fall descriptions within the health context of the person who falls may help identify important patterns, which can, in turn, inform preventive efforts. 20 Aims The overall aims of this thesis were: To explore the physical activity perceptions of older adults, to describe the characteristics of people with hip fracture as well as the situational circumstances of fall-related hip fracture events. The specific aims of Studies I−IV were: I. To explore and describe the physical activity perceptions held by older urban adults in Sweden and Ireland. II. To investigate whether the timing of hip fracture occurrence varied in relation to the subject’s health, residential status, or fall location and to examine the effect of season and weather on hip fracture incidence. III. To investigate the circumstances surrounding fall-related hip fracture and to describe fall circumstances in relation to participants’ health and functional characteristics. IV. To report the relative and absolute risks posed by a comprehensive range of disease categories for the occurrence of hip fracture, using a population-based cohort design. 21 Methods This work has an exploratory approach which incorporates research into two domains affecting the health and well-being of elderly people, namely hip fracture occurrence and physical activity beliefs. The aims of the thesis varied in nature from describing subjective experiences to calculating risks estimates which are applicable at a population level. In consideration that different aspects of reality lend themselves to different methods of inquiry,124 the adoption of different paradigmatic traditions, or ‘viewing positions’125 were required. A paradigm or methodology is the ‘basic set of beliefs,126 brought to the study by the researcher which influence the techniques used to answer the specific research questions. Whereas methodologies have been defined as ‘thinking tools’, methods on the other hand have been described as ‘doing tools’127 are not uniformly linked to paradigms. The various investigations incorporated in this thesis are underpinned by naturalistic, postpositivistic and pragmatic methodologies which in turn utilize qualitative, epidemiological, and mixed-method research designs. The chosen research methods of each design then translate each approach into practice. Naturalism using interview technique Study I was carried out in a naturalistic tradition as questions regarding people’s beliefs in relation to physical activity lend themselves to qualitative approaches. Qualitative inquiry is concerned with exploring the experiences and social contexts of participants.128 The ontological belief of such an approach is that multiple realities exist, and these realities are constructed through lived experiences.129 Face to face interviews were used as the method to access perceptions of PA in Study I and descriptions of fall circumstances in study III. In the naturalistic tradition, knowledge obtained through interviews is considered contextual and open to multiple interpretations.129 Resultantly, researcher objectivity is considered neither necessary nor advisable,130 as subjective experiences of the researcher are considered part of the process of interpretation. The analysis of interviews in studies I and III focused on the development of categories, derived inductively from the data.131 The choice of the qualitative method of analysis gives rise to findings which can vary on a continuum in relation to the degree of transformation of the 22 data. This may range from descriptive results at lower levels of abstraction to those which progress to more interpretative or explanatory domains.132 Postpositivism and epidemiology Studies II and IV had quantitative approaches which are rooted in postpositivistic traditions. Postpositivistic inquiry has a deterministic philosophy where causes or exposures determine the effects or outcomes and these relationships can be deductively ascertained by objective neutral processes.131 In order to answer the specific research questions in these studies, epidemiological methods were employed, which are concerned with measuring the occurrence of disease in relation to the population at risk. The axiological beliefs in this tradition are that bias, whether random or systematic in nature, is an error and attempts should therefore be made to control for these errors.133 The use of epidemiological methods and statistics in studies II and IV had the view of adopting an estimation approach, which involves reporting the magnitude and precision of an estimate, by means of the point estimate and confidence intervals respectively.133 This approach differs from that adopted when hypothesis testing is in focus. Mixed methods Study III was a mixed methods study, the guiding interpretative framework of which is pragmatism. The assumptions of pragmatism are that data collection and analysis is determined by nature of the study aims.134 Pragmatists believe that a richer analysis of the phenomenon can be achieved by tapping the relative strengths of positivistic and naturalistic methods.135 The integration of methodologies is therefore allowed, and pragmatism adheres, in general, to the belief that ‘a false dichotomy exists’136 between qualitative and quantitative data. The rationale for this approach in study III was that of complementarity, whereby participant descriptions of the fall which preceded hip fracture could provide insight into one aspect of fall circumstances and quantitative health data could elaborate and enhance these descriptions.134 23 Theoretical perspectives Investigation of multifactorial phenomena may be aided by the use of theories and schematic models which portray how multilevel factors interact to result in health outcomes among older people. Such models are based on the assumption that patterns of health and well-being are affected by a dynamic interplay between biological, behavioral and environmental factors.137 The use of more comprehensive conceptual models in relation to health outcomes also serves to highlight multiple points of possible intervention. The below described theories and associated models provided a theoretical basis from which hip fracture occurrence and PA perceptions were viewed throughout this work. International Classification of Function, Disability and Health The dynamic interaction of different domains on health is reflected in the World Health Organizations’ model: International Classification of Function, Disability and Health (ICF).138 The ICF is a framework for describing health and disability and commonly used in rehabilitative medicine. In the current work, the ICF was used as a general conceptual model concerning the areas PA perceptions and fall-related hip fracture. The ideology of the ICF model (Figure 1) is that biological, individual and social perspectives interact to result in health outcomes. A person’s activities and participation in everyday life can therefore be viewed from such biological, psychosocial and social perspectives.139 The fall risk triangle The fall risk triangle140 is an example of a health model which was designed specifically in relation to falls and can be used in the investigation of fall risk or planning of fall prevention at individual or group level (Figure 2). This schematic depiction of the interplay between multiple factors outlines three factor groups of pertinence in falls investigation; individual capacity, environmental and activity-related factors. In the current thesis, the model was used as a thought structure in the design of the interview protocol and questionnaire relating to fall events. Terminology in the fall risk triangle is in line with language used in the ICF framework whereby: ’Capacity’ incorporates aspects of physical, cognitive, medical status such as for example, balance, muscle strength or co-existing disease: ‘Activity’ incorporates the activity at the time of the fall and the way in which it was performed, and ‘Environment’ relates to physical and psychological aspects, such as flooring, lighting or stress. 24 Environment Activity Capacity Figure 2. The fall risk triangle Ethical considerations All participants in study I gave their informed consent to participate in the study. Ethical permission was not necessary according to the current legislation at that time. Ethical approval was sought and approved by the Regional Ethical Review board in Uppsala for studies II-IV. Qualitative interviews are by nature intrusive as they seek to access personal experiences and, in the case of hip fracture events, have the potential to evoke traumatic memories. A range of precautions were therefore undertaken to ensure the ease of all interviewed participants. Potential participants with hip fracture were firstly identified by a review of hospital records. The responsible nursing staff member for each patient was then consulted regarding the medical status of the individual in relation to the study inclusion criteria. When appropriate, patients were approached by the author, who explained the nature of the study and clarified her role as researcher and therefore not a member of the hospital staff responsible for the persons’ health care. People then received written information of the study and were given a day or two to review before being re-approached for confirmation of whether they wished to participate. Participants in study I chose a time and place for the interviews, and in study II and III, chose a suitable time and location within the hospital setting. All participants were reminded that interviews were voluntary and that withdrawal from the study would not be questioned. 25 Participants Participants included in this thesis came from two separate study populations, a population of healthy community-dwelling older adults in Dublin and Stockholm (Study I), and a population-based cohort of people admitted to Uppsala University hospital with hip fracture (Study II-IV). An additional study population was used in study IV and incorporated the source population of all people over the age of 50 years (born in 1960 or before), who were registered as residing in Uppsala County during the study period. An overview of study participants, design and methods is displayed in table 1. Study I Purposive sampling141 was the method used to ensure a diversity among 15 Swedish and 15 Irish participants with regard to physical activity level, age, gender and geographic location within the respective cities. Voluntary organizations such as active retirement groups and day centers for the elderly were used as a source for recruiting participants to the study. Active participants were recruited most often through active retirement groups, whereas contact with the lesser active participants tended to be made at day centers which served more as a social meeting point, and did not have physical activity on their agenda. The author, with the help of contact people from these organizations then selected potential participants according to the specific sampling criteria. To be eligible for participation in the study participants were required to be: 65 years and older, living in their own homes and functionally independent regarding activities of daily living. Cognitively impaired people were not eligible for inclusion, a factor which was judged by the contact person to whom participants were familiar. Study II and IV The study population in stuies II and IV consisted of people aged 50 years and older, who were consecutively recruited on the basis of admission to Uppsala University Hospital with radiographically confirmed hip fracture, during the 12-month period 8th June 2009–8th June 2010.Those with pathological fractures (n=5), periprosthetic fractures (n=12) and cases where surgical treatment did not occur at Uppsala University Hospital (n=15) were excluded, and four people died post-surgery, leaving 484 people with hip fracture in the study population. (This number then sank to 477 in study IV as a result of further cross-checking of hip fracture cases). As previously mentioned, the source population of all people who were residing in Uppsala county (n=117 494) in 2010, constituted an additional population in study IV. 26 Study III This study population consisted of 125 individuals who were consecutively included from people, aged 50 years and older, admitted to Uppsala University Hospital with radiographically confirmed hip fracture, during 10-months (September−June) of the above described 12-month period (Study II and IV). In addition, participants were required to fit the following criteria: at least partial memory of the fall, verbal ability to recount fall details and a Mini-mental state examination (MMSE) score ≥24 points. In consideration of the study aims, those with reduced consciousness, post-operative confusion or medical instability were not considered for participation. Figure 3. Flow chart over inclusion of participants in studies II−IV. Colour scheme: Study II (light/dark grey), Study III (yellow), Study IV (orange). 1Further case identification lead to the removal of 7 cases between study II and IV. 27 Population-based cohort 484 people with hip fracture, aged ≥ 50 years, admitted to an acute hospital setting, consecutively included during a one-year period. Medical journal access Interview administered questionnaires Face to face and telephone interviews Logistic regression Poisson regression Population proportional attributable risk (PPAR) Generalized linear models Purposive 30 community-dwelling adults aged ≥ 65 years, 15 living in Dublin, 15 living in Stockholm. In-depth interviews Thematic text analysis Sampling strategy Study population Data collection Data analysis 28 Population-based cohort study Qualitative Study II Design Study I Table 1. Overview of design and methods for study I − IV Data for disease diagnosis registered in the Swedish National Patient Register during the observation period 19642010 for study populations. Calculation of age and sex-stratified incidence proportions of hip fracture among exposed and unexposed to each disease category. Calculation of age- and sexstandardized risk ratios (RR’s) and risk differences (RD’s) by sex and 5-year age category. Individual interviews Interview-administered questionnaire. Hand dynamometer. Qualitative content analysis Descriptive statistics Analysis of variance for age Fisher’s exact test i) 477 people with hip fracture, aged ≥ 50 years, admitted to an acute hospital setting, consecutively included during a one-year period. ii) Source population of Uppsala County, aged ≥ 50 years during the study period. Population-based cohort and the underlying population in the region. Population-based cohort study Study IV 125 people with hip fracture, aged ≥ 50 years, MMSE score ≥24, who had memory and ability to verbally recount fall details. Purposive Concurrent triangulation mixed methods Study III Data collection and analysis Study I − A qualitative study Interviews were conducted mostly in participant homes or in the meeting place of their groups, in accordance with their individual wishes. The 15 interviews in Dublin were conducted in English and the 15 interviews in Stockholm were conducted in Swedish and lasted between 30 to 75 minutes. All interviews commenced with the same opening question – ‘What are your thoughts on physical activity?’ The participants were then encouraged to speak freely around specific themes relating to physical activity such as motivators and barriers, available information and if and how they believed physical activity to affect health. Transcribed interview material in this study was content-analyzed in accordance with the ‘thematic framework’ approach.142 Each transcript was firstly individually reviewed to identify the relevant topics that related to participants’ perceptions of physical activity, these were termed initial codes. These initial codes were then applied to all transcripts one at a time by a process of labeling. This analysis yielded preliminary interpretations of each participant’s perceptions, which were reanalyzed and checked against the transcribed material. Sub-themes and main themes were discussed and chosen by the authors so as to most accurately encompass and reflect participants’ perceptions during earlier stages of the analysis. Study II −A population-based cohort study Data collection occurred during hospital stay. Sociodemographic details and data concerning the fracture type, month of fracture, surgical reparation, comorbidity and medication were gathered from subjects’ medical records. Place of residence prior to the hip fracture was defined as either communitydwelling or residential care facility (RCF), whereby RCF was a collective term for serviced apartments and nursing homes for elderly people in greater need of medical care. Based on place of residence and fall location data, subjects were divided into three groups; community-dwellers who fractured outdoors, community-dwellers who fractured indoors, and those who lived in a RCF at the time of fracture. Diseases were defined according to International Classification of Diseases (ICD-10) categories as follows: Cerebrovascular diseases (I6), Ischaemic heart disease (I20-I25), Hypertension (I12 I15 I10.9 I11.9), Dementia (F00-F03, F05.1) and Depression (F32-F33, F20.4 F25.1 F31.3-F31.5 F41.2 F92.0). We further calculated the number of comorbidities using Charlson’s Comorbidity Index143 as a measure of overall comorbidity for each subject. Drugs were defined according to the Anatomical Therapeutic Chemical (ATC) classification system as: Benzodiazepines 29 (N05BA), Sedatives (N02A, N05B-C), Antidepressants (N06A, N06CA-B) and Diuretics (C03C). We defined Polypharmacy as the use of five or more medications and psychotropic drugs as the use of at least one drug from the categories N05A-C or N06A-C. The time of day at which the fracture occurred, estimated to the closest hour, was reported by 334 hip fracture cases (78 % of eligible cases) included in an interview sub-study. Verbal and written informed consent was obtained from these subjects or from their family members or care providers, in the case of cognitive impairment (MMSE score ≤ 24 points). All interviews were carried out within one week of the fracture event. When subjects could not personally relate the time of the fall event which lead to hip fracture, due to; reduced consciousness, language difficulties, delirium or cognitive impairment, this information was collected by telephone interview with a family member when the subject lived in the community (n= 75), or from a care nurse from the subject’s RCF (n=90). Weather data concerning daily snow depth (cm) for Uppsala County for the study period was obtained on request from the Swedish Meteorological and Hydrological Institute. Statistical analysis Logistic regression analysis was used to determine the age-adjusted odds ratios (OR) as a measure for the prevalence of comorbidities and medication usage between the three above-described subject groups. Logistic regression was also used to estimate the effect of psychotropic drugs or dementia (or both) on nighttime fractures and ORs were adjusted for the effects of age and Charlson comorbidity index. Poisson regressions were used to compare the fracture incidence rate between days where the ground was snow-covered and days of no snow cover and presented as incidence rate ratios (IRR). The population proportional attributable risk (PPAR) of hip fracture due to snow coverage for the entire sample/community-dwelling subjects was estimated using the formula: PPAR = Prevalenceexposure (RR-1)/ (1+ Prevalenceexposure (RR-1)), where Prevalence exposure is prevalence of the exposure (days with snow cover) and RR the risk ratio obtained through Poisson regression. Variation in time of fracture was modeled using generalized linear models with a Poisson distribution, by applying combinations of sine and cosine terms with varying wavelength up to the sixth harmonic or combinations of linear, quadratic and cube functions both for the 24-hour period and for the months of the year.144 The Bayesian Information Criteria (BIC) statistic was used as a criterion for model selection as it resolves the problem of overfitting, by introducing a penalty term for the number of parameters in the model. In situations where only one peak was observed in the model, we used a 30 directional statistical method to evaluate seasonal variation in fracture occurrence.145 These analyses were performed separately for the three groups based on living status and fall location, as well as for those with and without dementia/psychotropic drugs. Stata version 11.0 (Stata Corp., College Stn, TX, USA) was used for the statistical analyses. Study III − A concurrent triangulation mixed method study Collection of the qualitative and quantitative data With this design, qualitative and quantitative data were collected concurrently, analysed separately and then merged. Data concerning fall circumstances were collected during semi-structured interviews, performed by the same investigator, where participants were encouraged to speak freely while describing fall events. An interview guide was used and commenced with the open the question: ‘Can you describe for me what happened when you fell and broke your hip?’ These open-ended questions were then followed by probing questions which aimed to investigate the eventual influence of other specific environmental, individual or situational factors which may have played a role in the fall occurrence. Interviews varied in length from 8 to 25 minutes and were recorded and transcribed verbatim. Each interview was then followed by an interview-administered questionnaire which contained closed-answer questions concerning more specific details of the fall. This section of the questionnaire regarding fall details was developed with reference to previous hip fracture studies on fall mechanisms. 91, 146 Health data questions were incorporated into the interview-administered questionnaire and were developed with reference to previously existing health-related questionnaires tested in earlier epidemiological studies of fractures.9, 147, 148Questions related to prior performance of personal activities of daily living, mobility, balance, previous falls and PA participation were categorized as described in the section ‘Analysis of the health data’. Hand grip strength was tested, by the same investigator, using the ‘Baseline’ hydraulic hand dynamometer (Fabrication Enterprises Inc.) with the elbow flexed at 90 degrees and the participant sitting in a high-lying position. Three maximal grip contractions of the dominant hand were performed, the mean value of which was then calculated. Data concerning pre-fracture comorbidity and medication were retrieved from medical records. Comorbidity was categorized as 0, 1 and ≥ 2 chronic diseases, based on Charlson’s unweighted comorbidity score, calculated from ICD-10 diagnoses.143, 149 Fall-riskincreasing drugs (FRIDs) were categorized as: no FRIDs, psychotropic FRIDs (PsyFRID; ATC codes N02A, N03A, N04A-B, N05A-C, N06A), cardiovascular FRIDs (CvdFRID; ATC codes C01A, C01BA, C01D, C02, C03, C07-C09, G04CA) and concomitant use of PsyFRID and CvdFRID. 31 Analysis of the interviews Qualitative content analysis was the method used due to analyse the interview data where a step-by-step analysis is used to highlight patterns in the data according to phenomenon being studied.150, 151 Content analysis has been described by Krippendorff as “the use of replicable and valid method for making specific inferences from text,” 152 with the overall aim of making a condensed and broad description of a particular phenomenon. Our interview data also consisted of relatively short interviews which further motivated the choice of a method which allows data to be interpreted at various levels of abstraction. Based upon the study aims as well as the nature of the interview material, we chose to perform a manifest analysis, which aims to analyse content close to the text.153 We also chose to approach the text inductively thereby without preconceived categories.154 The analysis was performed as described by Graneheim and Lundman153 in the following stages: i) The interviews were read through several times to get a sense of the whole ii) Domains relating to areas of the text relating to causal factors for the falls were heighted iii) The meaning units of analysis were selected in direct relation to the study aims and involved sections of the text referring to perceived cause of the fall iv) Meaning units were then condensed and part of the texts were labeled to derive codes v) Codes of similar nature were then grouped to form sub-categories and then four main categories which can be identified as a thread throughout the codes.153 The analysis process involved a back and forth movement between the selected codes and the interviews as a whole. The first author was responsible for the primary coding of all interviews and the last author acted as a peer to the first author through a process of review and discussion of the credibility of categories in the data. Analysis of the health data Pre-fracture mobility was divided into the categories ‘Low’, ‘Moderate’ or ‘High’ which were defined as follows: Low (required a walking aid indoors); Moderate (required a walking aid outdoors only), High (no walking aid required). Participation in personal activities of daily living was categorized as independent /dependent in ≥ 1 activity of personal care, whereby for example participants who required assistance or supervision during bathing were considered dependent in one activity. With regards to PA participation, participants answering ‘hardly any exercise’ to the question ‘How much exercise did you engage in prior to the fracture?’ were classified as ‘Sedentary’, with other divisions including ‘Light exercise’ (Lighter exercise such as regular walks or gardening) or ‘Hard physical exercise’ (Do you engage in hard physical training or sport?). This question was based on a questionnaire, created in collaboration with the Swedish national institute of health which measures lifetime physical activity, 155 which has been used in previous cohort studies of older Swedish populations.9, 32 Absolute grip strength values 32 were compared with published normative grip strength data,156 and stratified according to age, gender and height. ‘Low’ grip strength incorporated values between the 10th and the 5th percentile of normative data and ‘Abnormally low’ grip strength as those under the 5th percentile of normative data. In this calculation, 7 individuals (1 man and 6 women) lacked height data so for these we used the median height in the sex-specific age group in our sample (women ≥ 65 years: 164 cm; men ≥ 65 years: 175 cm). This imputation was only used for the definition of low grip strength. The two forms of data were then merged in the results section using a joint display of the data 157 and then interpreted in the discussion section. To facilitate the integration of the data sets, qualitative fall circumstance categories were treated as categorical variables. We organized the data in relation to the fall circumstance categories, which were viewed in relation to variables of health and function using cross tabulation and descriptive statistics. This triangulation of data occurred during both the analysis and interpretation phases with the purpose of interrelating and identifying patterns in the data.158 Study IV − A population-based cohort study Outcome and exposures The outcome under investigation was admission for radiographically confirmed hip fracture (S72.0–S72.2 according to ICD-10) to Uppsala university hospital, during the 12-month period, 8th June 2009–8th June 2010. Each hip fracture case was individually validated by crosschecking medical records which ruled out misclassification of events. Pathological (n=5), periprosthetic (n=12), and those fractures where surgical treatment did not occur at Uppsala University Hospital (n=15) were excluded. Exposures in the current study included all diseases registered under the primary and secondary diagnoses in the Swedish National Patient Register (NRP). The NPR contains data concerning disease diagnosis registered during hospital admissions from 1964 and onwards. Diagnoses in broad categories according to the ICD-10 were as follows; infectious and parasitic diseases (A00-B99); malignant tumors (C00-C97); non-malignant tumors (D00D48); diseases of blood/blood forming organs (D50-D89); endocrine, nutritional and metabolic disorders (E00-E90); mental and behavioural disorders (F00-F99); diseases of the; nervous system (G00-G99); eye (H00-H59); ear (H60-H95); circulatory system (I00-I99); respiratory system (J00-J99); digestive system (K00-K93); skin and subcutaneous tissue (L00-L99); musculoskeletal system (M00-M99); genitourinary system (N00-N99); previous injury (S00-T98) and fracture injury (S 12/22/32/42/52/62/72/82), which was 33 a subcategory of injury. Data were extracted such that each diagnosis could be counted once per individual during the observation period 1964-2010. For subjects with two hip fractures during the observation period (n= 7), the date of the second fracture was used as the index date, to prevent diseases being diagnosed more than once for the same individual. Data extraction was performed in a similar way for both the hip fracture cohort and for the source population. Statistical analysis Using the 477 subjects with hip fracture during a one-year period as outcome, the underlying population was included as a reference population in a cohort analysis. Age was divided into 5-year strata and subjects aged 85 years and above were included in an additional category. The prevalence of each disease category was calculated for both the hip fracture cohort and source population. Age and sex-stratified incidence proportions of hip fracture were calculated among those exposed and unexposed to each disease category. These calculations were then used to estimate the age- and sexstandardized risk ratios (RR’s) and risk differences (RD’s) of hip fracture in the total population, and separately by sex, and age category (50-69; 70-79; 80+) using total number of exposed as internal weights. Statistical analyses were performed using Stata 11.0 (Stata Corp., College Station, TX, USA). 34 Findings Perceptions of physical activity (Study I) Participants were divided into three groups in relation to how physically active they described themselves as being: Active (involved in planned PA of non-functional nature, at least three times a week), moderately active (engaged in some weekly PA which was often unscheduled and functional in nature) and inactive (not engaged in PA of any kind outside of the home). Three central themes were identified in relation to participants’ perceptions of PA. These were: PA as self-perception, PA as interaction and PA as health promotion. Active participants often claimed not to need external motivators to be physically active as, for them, PA was a lifestyle and an inherent part of selfidentity. The uptake of PA in retirement years was also discussed by others as a means of establishing a new self-identity. For these ‘late starters’, becoming physically active was perceived as a means by which they could invest in themselves, as well as a way to adapt to major lifestyle changes which occurred in connection with retirement or children leaving the home. However, not all participants adapted to retirement in the same way, as inactive men from both cultures appeared to identify strongly with former occupations and spoke of how the consequences of their previous working life had negatively affected their health and thus explained their current levels of inactivity. All participants acknowledged the physical or mental health benefits associated with PA, but variations existed in the extent to which people believed that PA could affect their personal health. More active participants had in general, strong beliefs about the physical and psychological health benefits of PA. They viewed symptoms from chronic conditions as a motivating factor to remain physically active and had altered the nature of their PA to adapt to declines in capacity in the attempt to remain active. Lesser active people, on the other hand, spoke of how symptoms such as joint pain or breathlessness reduced or prevented them from being physically active. Some of these lesser active participants expressed feeling unsure and uninformed regarding whether physical exertion was advised or not in the presence of symptoms of chronic disease. Some moderately active participants expressed skepticism 35 towards the targeting of certain forms of exercise towards older people. They believed gym-based exercise to be nonsensical and placed a higher value on exercise that was of functional nature. Cultural differences were most evident in terms of the interactive role played by PA. Social interaction with peers was stressed to a greater extent among Irish participants whereas outdoor pursuits in natural environments featured more strongly as a recurrent motivator to PA among Swedish participants, of all PA levels. Physically inactive participants (from both cultures) often expressed a desire for greater social contact, but felt confined to their home environment by physical ailments. When and where hip fractures occur (Study II) A total of 484 hip fracture cases were included in this study. Of the total cases, 338 (69.8 %) were female. The mean age at fracture was 82.7 years in women (range, 51–104) and 79.5 years in men (range, 52–98). In total, 375 (80.3 %) fractures occurred indoors and of those which occurred in residential care, all except for six cases, occurred indoors. Gender was not seen to be associated with either type of hip fracture or fall location. We divided subjects into three groups, based on place of residence and fall location at the time of fracture; community-dwellers who fractured outdoors (n=62, 18 %), community-dwellers who fractured indoors (n= 234, 50 %), and subjects who fractured in a residential care facility (n= 151, 32 %). In relation to health status, residential care dwellers had a higher prevalence of disease and medication use than those living in community settings. When controlled for the effects of higher age, subjects living in RCFs were more likely to suffer from dementia (OR: 11.34; 95% CI; 6.26−20.57) and to be receiving a greater number of fall-risk-increasing medications (OR: 3.35; 95% CI; 2.07–5.45), than community-dwellers who fractured indoors. In relation to the time at which fracture occurred, community-dwelling subjects showed a peak in fracture occurrence between 12 and 2 o’clock in the afternoon, regardless of whether the fall occurred indoors or outdoors. Those living in RCFs, on the other hand, showed no daytime peak in the fitted frequency model and demonstrated a greater tendency to fracture during evening and night-time hours (Figure 4) than community-dwellers who fractured indoors (OR: 1.37, 95% CI; .86−2.10). 36 a b 20 c 15 15 10 10 5 5 0 Frequency Frequency 20 0 0 3 6 9 12 15 Time of day 18 21 24 0 3 6 9 12 15 Time of day 18 21 24 0 3 6 9 12 15 Time of day 18 21 24 Figure 4. Time of fracture according to residence and fall location. The fitted functions were: Panel a (Community-dwellers who fractured indoors, n=166); Panel b (Community-dwellers who fractured outdoors, n=62) and panel c (Residential care facility dwellers, n=104). In consideration of the high prevalence of dementia and use of psychotropic drugs among RCF dwellers, we examined the association between nighttime fractures and psychotropic drugs. Subjects using psychotropic drugs, were also twice as likely to fracture during nighttime hours (Adj. OR: 2.20, 95% CI; 1.12–4.30) compared to those not receiving these medications (Figure 5). Even when subjects with a dementia diagnosis were excluded from the analysis (n= 75), the associations between psychotropic drugs and nighttime fracture remained (OR: 2.91, 95% CI; 1.40–6.0). This trend was also observed, albeit with a weaker association among community-dwellings subjects (Adj. OR: 2.31, 95% CI; .98–5.42). a b 25 c 20 20 15 15 10 10 5 5 0 Frequency Frequency 25 0 0 3 6 9 12 15 Time of day 18 21 24 0 3 6 9 12 15 Time of day 18 21 24 0 3 6 9 12 15 Time of day 18 21 24 Figure 5. Time of fracture according to psychotropic drug use. The fitted functions were: Panel a (subjects without dementia not using psychotropic drugs, n=136); Panel b (Subjects using psychotropic drugs, n=181) and Panel c (Subjects using psychotropic drugs without a dementia diagnosis, n=123). 37 Although we did not observe a seasonal pattern in hip fracture variation, based on month, we reported that on days of snow cover, communitydwellers, as a whole, had a heightened hip fracture incidence rate (IRR: 1.32, 95% CI; 1.05–65). This increased risk was seen both indoors and outdoors. The population proportional attributable risk of hip fractures that occurred due to snow was calculated as 5.5% for the entire subject group and 9% for community-dwelling subjects. Circumstances surrounding fall-related hip fractures (Study III) Fall circumstances Participant descriptions of fall circumstances are presented in relation to two main domains, Activity at the time of the fall (Table 2 and 3) and Nature of the fall (Table 4 and Figure 6). Table 2. Activity at the time of the fall of the 125 interviewed participants Activity, n (%) Indoor 83 (66.4) Outdoor 42 (33.6) Positional change Sit-to-walk1 Stand-to-sit2 23 (27.7) 20 (24.1) 3 (3.6) 1 (2.3) 1 (2.3) 0 24 (19.2) 21 (16.8) 3 (2.4) Standing Standing still Standing & bending/reaching 13 (15.6) 6 (7.2) 7 (8.4) 3 (7.1) 0 3 (7.1) 16 (12.8) 6 (4.8) 10 (8.0) Walking Forward walking Turning while walking 41 (49.4) 29 (34.9) 12 (14.5) 31 (73.8) 29 (69.1) 2 (4.8) 71 (56.8) 57 (45.6) 14 (11.2) 7 (8.4) 3 (3.6) 4 (4.8) 7 (16.6) 1 (2.3) 6 (14.3) 14(11.2) 4 (3.2) 10 (8.0) Balance challenging Walking on stairs ‘Hazardous’3 1 Total n=125 Falls during the initial 3 meters of walking having risen from a sitting/lying position, 2Falls while transferring from a standing to a sitting/lying position, 3Falls during complex activities, such as from: bikes, stools, ladders or while running or sledging. 38 Nature of the fall Indoor falls (n=83, 66%): Environmental (n= 32, 26%); falls described as primarily precipitated by environmental conditions or objects, which included the subcategories: trips over; mats (n=5), thresholds (n= 3), and; household objects (n=10), slips on wet surfaces (n=8) and inadequate footwear, (n=7). Physiological (n=35, 28%); falls described as precipitated by physiological factors, with no environmental component and included the subcategories: self-induced disequilibrium − a symptomless loss of balance during normal body movements (n=14); falls proceeded by physical symptoms such as dizziness or faintness (n=13) and reduced function or pain of the lower limb (n=6). Activity-related indoor (n=8, 6%); falls precipitated by the complex nature of the activities engaged in, such as standing on one leg or on kitchen stools. Outdoor falls (n=42, 34%): Trips & slips, snow (n=20, 16%); slipping and tripping on snowy or icy surfaces. Trips & slips, no snow (n=12, 10%); slipping and tripping on snow and icefree surfaces; Activity-related outdoor (n=8, 6%); falls during high speeds (n=2) or from ladders (n=3) or bikes (n=3). The category Unknown (n=10, 8%) incorporated falls of unexplained nature occurring both indoors (8/10) and outdoors (2/10). Patterns between fall circumstances and health characteristics Those who fell during positional change had the poorest functional status Participants whose fall-related fracture occurred during positional change appeared to be those with greatest functional limitations. Support for this pattern in the quantitative data was primarily seen in relation to mobility and limitations in personal activities of daily living (Table 3). Additionally, this group more frequently rated themselves as being sedentary, having ‘bad’ balance and having fallen previously. In terms disease prevalence and medication use, patterns were somewhat divergent as differences were not significant from other groups. Nevertheless, support for this pattern was present in the interview data as the majority of those who fell during positional change frequently described physiological symptoms of chronic conditions or poor health as precipitants for their falls. 39 Table 3. Activity at the time of the fall in relation to health characteristics of 125 interviewed participants. Health characteristic Age, years (mean (SD)) Gender (female) Mobility1 Low2 Moderate3 High4 P-ADL Participation1 Dependent in ≥ 1 activity Balance1 Self-rated balance (bad), (n=114) Fear of falling (yes), (n=122) Previous falls ≥1 fall previous year, (n=122) Physical activity1 (n=118) Sedentary Nr. of chronic diseases 5 0 1 ≥2 Fall-risk-increasing drugs (FRIDs) 0 FRIDS Cardiovascular (Cvd) FRIDs Psychotropic (Psy) FRIDs Concomitant Cvd & Psy FRIDs Positional change n=24 81.4 (10.3) 16 (66.7) Standing Walking n=16 78.7 (9.3) 11 (68.7) n=71 79.4 (8.7) 52 (73.3) Balancechallenging n=14 76.7 (9.1) 10 (71.4) 15 (62.5) 6 (25.0) 3 (12.5) 3 (18.8) 7 (43.7) 6 (37.5) 17 (23.9) 16 (22.5) 38 (53.5) 1 (7.4) 2 (14.3) 11 (78.6) 7 (29.2) 0 6 (8.5) 0 16 (69.6) 9 (56.3) 28 (45.2) 6 (46.2) 10 (41.7) 5 (31.3) 19 (27.9) 2 (14.3) 14 (58.3) 6 (40.0) 33 (47.8) 7 (50.0) 9 (37.5) 1 (6.3) 11 (15.5) 1 (7.4) 12 (50.0) 7 (29.2) 5 (20.8) 10 (62.5) 5 (31.3) 1 (6.3) 35 (49.3) 26 (36.6) 10 (14.1) 10 (71.4) 3 (21.4) 1 (7.1) 4 (16.7) 10 (41.7) 4 (25.0) 4 (25.0) 13 (18.3) 29 (40.8) 3 (21.4) 5 (35.7) 2 (8.3) 8 (33.3) 3 (18.8) 5 (31.3) 6 (8.4) 23 (32.4) 2 (14.3) 4 (28.6) Units are expressed as number (percentage) unless otherwise stated. 1 Reported performance of pre-fracture status. 2Required a walking aid indoors. 3Required a walking aid outdoors. 4No walking aid required. 5According to Charlson’s comorbidity index. 40 Those who described environmental fall factors (indoors) had moderate physical function but high levels of comorbidity and medication use The two main described categories of the nature of indoor falls included physiological and environmental and are illustrated in Figure 6. When compared to those describing falls of physiological nature, those describing environmentally precipitated falls reported higher levels of physical function Although not statistically different, these participants less frequently reported their balance as ‘bad’ (p=0.075) and appeared to have higher levels of mobility (p=0.461) and functional independence (p=0.477). Support for stronger beliefs in functional performance was frequently found in interview data where these participants commonly described collisions with furniture or tripping on cords, whilst cleaning indoors, or slipping on surfaces although often moving unhindered in the home. Despite a higher level of selfreported physical function, the group environmental had nonetheless a similar prevalence of comorbidity, fall-risk-increasing drug prescription and previous falls as those perceiving falls as caused by physiological factors (Figure 6). Mobility (low) P-ADL* (dependent in ≥1) Self-rated balance (bad) Previous falls ( ≥ 1 previous year) Physical activity (sedentary) Grip strength (abnorm low) Chronic diseases ( ≥2) Fall-risk-increasing drugs ( ≥2) 0 5 10 Physiological, n=35 15 20 25 Environmental, n=32 Figure 6. Nature of the fall in relation to health characteristics for the two main indoor categories, Physiological and Environmental. * P-ADL (Personal activities of daily living) 41 Those who fell outdoors in snow-free environments appeared to have poorer health than other outdoor groups Those fracturing outdoors in snow-free environments (Trips & slips, no snow) appeared to have lower levels of mobility, a higher incidence of previous falls and showed tendencies towards higher prevalence of disease and use of fall-risk-increasing drugs, compared with other outdoor groups (Table 4). These participants described their falls in terms of environmental factors, often involving unsuitable footwear and/or trips on uneven paths, tree roots or curbsides which were either misjudged or not visible. Not surprisingly, those falling during more balance challenging activities and those describing activity-related falls reported the least functional limitations and appeared to be those most physically active, both indoors and outdoors. 42 Table 4. Nature of the fall (outdoors) in relation to health characteristics of interviewed participants (n= 401). Health characteristic Age group 50−80 years 80+ years Gender (female) Mobility2 Low 3 Moderate4 High5 P-ADL participation2 Dependent in ≥1activity Balance2 Self-rated balance (bad), (n=114) Fear of falling (yes), (n=122) Previous falls ≥1 fall previous year, (n=122) Physical activity2 (n=118) Sedentary Number of chronic diseases 6 0 1 ≥2 Fall-risk-increasing drugs (FRIDs) 0 FRIDs CvdFRIDs PsyFRIDs Concomitant Cvd & PsyFRIDs Trip/slips (snow) n=20 Trip/slips (no snow) n=12 Activityrelated n=8 16 (80.0) 4 (20.0) 14 (70.0) 6 (50.0) 6 (50.0) 7 (58.3) 5 (62.5) 3 (37.5) 5 (62.5) 0 5 (25.0) 15 (75.0) 3 (25.0) 1 (8.3) 8 (66.7) 0 2 (25.0) 6 (75.0) 2 (15.4) 0 0 7 (36.8) 4 (20.0) 6 (54.5) 3 (30.0) 5 (62.5) 0 6 (30.0) 7 (58.3) 2 (25.0) 1 (5.3) 1 (8.3) 0 15 (75.0) 5 (25.0) 0 6 (50.0) 5 (41.7) 1 (8.3) 3 (37.5) 4 (50.0) 1 (12.5) 8 (40.0) 7 (35.0) 0 5 (25.0) 3 (25.0) 3 (25.0) 1 (8.3) 5 (41.7) 1 (12.5) 6 (75.0) 1 (12.5) 0 Units are expressed as number (percentage) unless otherwise stated. 1 Two outdoor falls categorised as ‘Unknown’ are not illustrated in the table. 2 Reported performance of pre-fracture status.3 Required a walking aid indoors. 4Required a walking aid outdoors only. 5No walking aid required. 6 According to Charlson’s unweighted comorbidity index. 43 The impact of disease on hip fracture risk (Study IV) The prevalence of disease categories in the hip fracture and in the underlying population is shown in table 5. All disease categories were associated with an increased risk of hip fracture and highest risk ratios (RR) were observed for Circulatory system diseases (RR: 6.28, 95% CI; 5.06–7.79) and Previous injuries (RR: 5.82, 95% CI; 4.80–7.05) which had a 6-fold increase in hip fracture risk and the subgroup Previous fractures (RR: 5.12, 95% CI; 4.18– 6.28) (Figure 7). An approximately 5-fold increase in risk was seen for disorders of the digestive and genitourinary systems. In absolute terms, conditions with highest impact for hip fracture occurrence per 1000 individual included; previous fracture (41/1000); previous injury (27/1000) mental disorders (26/1000), non-malignant tumors (25/1000); genitourinary and eye disorders (both 24/1000), and digestive system diseases (23/1000). No major differences were observed across disease exposures between men and women and combined estimates are therefore presented in the illustrations. In relation to age, risk ratios for all disease categories were highest among the youngest age groups (50−69 years) and decreased in line with increasing age (Figure 8). Despite larger observed risk ratios for disease exposures among younger individuals, the impact of hip fracture in absolute terms was lower among this group as demonstrated by lower risk differences for all diseases in the younger age groups. 44 A00-B99 C00-C97 D00-D48 D50-D89 E00-E90 F00-F99 G00-G99 H00-H59 H60-H95 I00-I99 J00-J99 K00-K93 L00-L99 M00-M99 N00-N99 S00-T98 S12-82a ICD-10 code S12/22/32/42/52/62/72/82- fracture of the neck/thorax/lumbar spine/pelvis/upper limb/lower limb/foot a Infectious and parasitic diseases Malignant tumors Non-malignant tumors Diseases of blood/blood forming organs Endocrine, nutritional and metabolic disorders Mental and behavioural disorders Nervous system diseases Diseases of the eye Diseases of the ear Circulatory system diseases Diseases of the respiratory system Diseases of the digestive system Diseases of the skin and subcutaneous tissue Diseases of the musculoskeletal system /connective tissue Diseases of the genitourinary system Injury Fracture injurya Disease category 45 Prevalence, n (%) Hip fracture Source popula(n=477) tion (n=117 494) 106 (22.2) 4588 (3.8) 76 (15.9) 3867 (3.2) 79 (16.6) 2398 (2.0) 89 (18.7) 2996 (2.5) 162 (34.0) 7245 (6.0) 135 (28.3) 4013 (3.3) 101 (21.2) 3812 (3.2) 64 (13.4) 1850 (1.5) 33 (6.9) 1019 (0.8) 319 (66.9) 12709 (10.5) 140 (29.4) 5471 (4.5) 210 (44.0) 7270 (6.0) 32 (6.7) 1858 (1.6) 187 (39.2) 7316 (6.1) 224 (47.0) 7355 (6.1) 246 (51.6) 7559 (6.2) 145 (30.4) 2853 (2.3) Table 5. Prevalence of disease categories in the both the hip fracture cohort and the source population in Uppsala county. 1 2 3 4 5 6 Risk ratio (95% CI) 7 8 0 10 20 30 40 Risk difference per 1000 (95% CI) 50 46 Figure 7. Age- and sex-standardized risk ratios (RR’s) and risk differences (RD’s) of all disease categories (per 1000) and 95% confidence intervals (CI) of all disease categories for hip fracture. Fracture injury Injury Diseases of the genitourinary system Diseases of the musculoskeletal system/connective tissue Diseases of the skin and subcutaneous tissue Diseases of the digestive system Diseases of the respiratory system Circulatory system diseases Diseases of the ear Diseases of the eye Nervous system diseases Mental and behavioural disorders Endocrine, nutritional and metabolic disorders Diseases of blood/blood forming organs Non−malignant tumors Malignant tumors Infectious and parasitic diseases 1 10 20 30 40 50 60 Risk ratio (95% CI) Age 50−69 1 5 10 15 20 1 5 10 15 20 Risk ratio (95% CI) Risk ratio (95% CI) Age 70−79 Age 80+ 47 Figure 8. Age- and sex-standardized risk ratios (RR’s) (per 1000) and 95% confidence intervals (CI) of all disease categories for hip fracture according to the age categories 50−69 years, 70−79 years and 80+ years. Fracture injury Injury Diseases of the genitourinary system Diseases of the musculoskeletal system/connective tissue Diseases of the skin and subcutaneous tissue Diseases of the digestive system Diseases of the respiratory system Circulatory system diseases Diseases of the ear Diseases of the eye Nervous system diseases Mental and behavioural disorders Endocrine, nutritional and metabolic disorders Diseases of blood/blood forming organs Non−malignant tumors Malignant tumors Infectious and parasitic diseases Methodological Considerations Mixed methods This thesis, as a whole, was inspired by a mixed methods approach, which represents a primary methodological strength. The main rationale of using mixed methods is to provide a comprehensive description of the events under investigation. In the current thesis the choice of this approach was further motivated by several other important factors. Health research should support clinical decision making, and effective clinical reasoning relies on drawing from different types of knowledge.131 Additionally, health care aims to adopt a holistic and patient-centered perspective, one aspect of which involves increase patient involvement.120 Fall-related hip fracture, like many health conditions, often involves a complex interaction of multiple contributing factors, which in turn requires a multifaceted approach of investigation.134 No single research method can explain all the facts, which motivates the choice of mixed method approaches, whereby additional insight about the research problem can be gained from integrating knowledge from different sources.159 Use of the Swedish National Patient Register as a source of data, concerning disease incidence, had the advantage of capturing health data from the entire source population. Individual identification of all hip fracture cases in our cohort also ruled out case misclassification which can occur in register-based data, when people with a history of hip fracture are falsely reregistered as new cases upon re-admission to hospital within the same year as the fracture occurred.160 Furthermore, sampling methods in cohort studies which require active participation, will on the other hand, create a bias towards the younger, healthier members of a population who are physically or cognitively capable of visiting a research center. An additional advantage of register-based data was the ability to calculate the absolute risk of disease categories on hip fracture, which better indicates the public health impact of disease, and is the estimate which should, if possible, be communicated to patients.123 A further motivation for using mixed methods is that the strength of one method can compensate for the weaknesses of another.134 The use of interviews, on the one hand, allowed us to access information concerning fall circumstances of those who were cognitively capable of recounting fall details. The use of epidemiological methods, on the other hand, enabled the inclusion of those not capable of recounting fall details due to physical and mental disease, a group commonly excluded in hip fracture research.67, 161 Relying on self-reports of fall circumstances The majority of fall-related hip fractures occur unobserved in community settings,162 which means that accounts from those who experience falls are often the only source of this information. Nevertheless, the collection of self48 reported data regarding fall circumstances is often criticised as being potentially inaccurate, unobjective and therefore undependable as a source of information. These criticisms are grounded on proposals that older people are either not capable, due to cognitive impairment or poor recall, or not willing, due to feelings of vulnerability, to recount details of their falls. Cognitive impairment was a factor of relevance in our sample, as post-operative delirium is known to be highly prevalent among hospitalized older people who have recently undergone reparative hip fracture surgery.163 This factor was taken into consideration during the inclusion process to study III and less than one third of all patients with hip fracture were capable of verbally recounting details of the fall. Participants who were interviewed in study III can therefore be thought to represent those, primarily community-dwelling, people with hip fracture who have the lowest burden of physical and mental illness. The sparse representation of residential care dwellers among those interviewed (n=5) therefore reflects the high prevalence of both postoperatively acute and chronic forms of cognitive impairment in this group. This, in turn, means that we are unable to generalize findings from study III to those living in institutional care. For those who were capable of describing the fall, it should be further considered that falls can be emotive subjects which participants may have wished to distance themselves from.122 Criticism of older people’s fall descriptions as unobjective is complicated however for several reasons. Even if fall descriptions cannot always be considered as exact accounts of the chain of events, they still highlight important subjective beliefs concerning the nature of the fall event, which may both provide clues as to intervention strategies and may also influence how the individual reasons in relation to future fall prevention. Additionally, even in studies where falls are captured on video and are thereby objectively accounted for,164, 165 analysis of these fall scenarios is still reliant on the subjective opinions of one or several, albeit qualified, individuals. Trustworthiness Study I and part of study III involved the collection and analysis of interview data and when ensuring scientific rigor in the treatment of text data, the use of different procedures are required.157 In order to achieve credibility, specific measures were taken during the various stages of data collection and analysis. During the interviews, a process of dialogical validation was used, whereby participants’ responses to certain questions were checked by posing follow-up questions such as ‘Have I understood correctly when I say that you answered that’.166 Detailed descriptions were also provided regarding steps in the process of both thematic analysis (study I) and qualitative content analysis (Study III). Thematic analysis and qualitative content analysis are largely similar processes, with minor differences existing mostly in relation to the terminology used to describe the analytic process. Citations from 49 the interviews, in both studies, have been used in the articles to allow the reader to make links between interview data and analytic interpretations. This process also aids the reader when deciding whether findings can be transferred to other settings.129 Furthermore, in order to ensure validity of the analytic process different forms of triangulation were used. Researcher triangulation was achieved during the analysis of interview transcripts during continual peer debriefing sessions between the authors (B. Leavy and AC Åberg).The co-author (AC Åberg) analyzed selected interviews and compared and checked the interpretation of the first author. During these sessions, sub-themes and main themes were reviewed and discrepancies discussed, so that categories would most accurately encompass and reflect the interview data. The quality and richness of text data determines the extent to which theoretical abstraction is possible during the analysis process.153 Qualitative interviews are concerned with accessing the experiences and meanings people attribute to dimensions of their lives.130, 131 In study I, this involved exploring general beliefs regarding PA, whereas in study III, the aim of the interviews was to explore fall events which occurred unexpectedly and instantaneously. This probably explains why many of the interviews in study III were short in length, a factor, in combination with the large number (125) of individual interviews, shaped our choice of analysis method. We considered a manifest analysis, with a lower level of theoretical abstraction, to be the most suitable method to enable an authentic representation of the participant descriptions and to avoid misdirected interpretations of the interview transcripts. Data triangulation of fall circumstances (study III) was also possible by comparing interview data with answers from closed-ended questions in questionnaire as well as medical journal data from the time of hospital admission. This allowed for a cross-checking and comparing of data from different sources which related to the same fall situation in study III.141 The use of single item questions to capture health data In light of the multifactorial nature of hip fracture occurrence, we aimed in studies II-III to investigate fall circumstances while also provide a comprehensive picture of the participants’ health and functional status. In order therefore to cover both these aspects, within a time period considered realistic for hospitalized older people who have recently undergone surgery, we chose to base our data concerning functional performance, primarily on the use of single item questions. Examples of such questions relate to physical activity and fear of falling. The disadvantage of this approach is that such questions lack sensitivity, and in the case of the dichotomous question − ‘Before your hip fracture, were you afraid of falling? (yes/no), a ‘yes’ answer does not differentiate between excessive or rational fear, or whether 50 fear related to the risks, or the consequences of falling.167 For this purpose, the use of standardized assessment tools such as, for example, the Falls Efficacy Scale−International,168 or the Physical Activity Scale for the Elderly,169 could have captured aspects and gradients of diversity which are not reflected in our data pertaining to these domains. On the other hand, the choice to use single item questions items is simple, less time consuming and may reflect clinic situations where the comprehensive health backgrounds of older people are often charted in a single visit to a doctor of physiotherapist. Separating the effects of disease and medication Confounding by indication is a factor deserving attention when interpreting the observed associations between psychotropic medications and hip fracture in study II. Psychotropic drugs are prescribed as symptom management for underlying conditions such as dementia, depression, anxiety, pain or sleeping problems, all conditions which also predispose to falls and hip fracture.76, 170 Stratification according to dementia status in study II allowed us to separate the effects of fall-risk-increasing drugs from dementia on night-time falls, but the confounding effects of other health conditions cannot be ruled out. It is possible that among community-dwelling older people in our sample, depression and sleep disorders were more prevalent underlying factors for psychotropic drug prescription and could therefore have confounded the associations observed between time of the fall and medication use.171, 172 Similarly, we were unable to separate the effects of underlying disease from the effects of medications routinely prescribed to treat those diseases in Study IV. It is possible therefore that the observed absolute risk posed by mental disorders for hip fracture (26/1000) is somewhat explained by fallrisk-increasing medications used to treat these conditions. 51 General Discussion Summary of results Being physically active in natural, outdoor environments was a recurrent motivator to physical activity among Swedish participants, regardless of their described level of activity. The importance of exercise which was functional in nature was also stressed by moderately and lesser active community-dwelling participants. Additionally, the uptake of physical activity in retirement years was described as a means of establishing a new self-identify. In relation to hip fracture patterns, the time of day at which hip fracture occurred differed in accordance with place of residence and these 24-hour patterns may be further explained by medication use. Both residential and community-dwelling people without a dementia diagnosis, who were receiving psychotropic medication were at increased risk for hip fracture during night-time hours. Community-dwelling people with the poorest health characteristics tended to fall and fracture during changes of position indoors, and often described individual health factors as precipitators for their fall. The majority of people, nonetheless, describe their fall in terms of environmental factors, despite having relatively high levels of co-existing disease and medication use. All disease categories were seen to be positively associated with hip fracture, with cardiovascular disease and previous injury having the highest relative and absolute risks for fracture. The role played by outdoor environments for physical activity beliefs and hip fracture patterns Being outdoors as a motivating factor for physical activity The importance of nature and the outdoors as a motivating factor for physical activity was apparent when subjective perceptions were explored (Study I). The attraction of outdoor environments was prominent when culturespecific motivating factors for physical activity were considered, whereby outdoor activities were valued highly especially among Swedish participants, of all activity levels. Physical activity in natural surroundings was described as therapeutic as it provided the opportunity for thought and reflection and was described as something which positively affected well-being. The added 52 benefits of being active in natural or ’green’ environments, as opposed to indoor environments, are reported to primarily involve positive effects on mental health, such as decreased tension and depression, among adults in general173, 174 and to some extent among older adults.175 It also appears that older people who engage in outdoor activity achieve higher levels of PA, compared to those who do not.176 The skepticism towards gym-based training, which we observed among certain participants in study I, serves furthermore to highlight the pivotal role played by outdoor walks among certain older people, concerning whether they will achieve the minimum levels of PA which are required for the prevention of disease. Outdoor environments and the risk for hip fracture The physical activity and health patterns of people with hip fracture were somewhat reflected in the location and circumstances surrounding the fall events which led to fracture. When falls were viewed solely in relation to location (indoor/outdoor) in study II, those who fractured outdoors had the lowest levels of comorbid disease and medication use. A further breakdown of the circumstances surrounding falls in study III however, revealed the more heterogeneous health status of those who fractured outdoors. Compared to those who fractured indoors, outdoor groups generally reported higher levels of physical activity and had a lower prevalence of disease. However, we also identified a somewhat frailer group, in terms of disease burden and medication status, who also reported a higher prevalence of previous falls compared to other outdoor groups. These individuals also fell while walking under less challenging, snow-free conditions. This group of less vigorous adults, who still maintain physical activity outdoors has not previously received attention in the falls literature, possibly due to several reasons. Outdoor falls among older adults have been the focus of fewer investigations than indoor falls.177 Those studies which have focused specifically on outdoor falls, have often comprised of healthy, relatively younger community-dwelling populations who may not necessarily represent those at risk for hip fracture.177, 178 Other studies which dichotomized falls according to indoor/outdoor fall location have demonstrated separate risk factor profiles for indoor and outdoor falls,179-181 but have not considered sub-group variations among those who fall outdoors. In the literature, indoor falls are associated with indicators of poorer health and function whereas outdoor falls have been viewed as markers of a ‘vigorous’, physically active elderly group.181, 182 Additionally, indoors falls are reported to predict future limitations in physical activity, whereas outdoor falls have not.63, 179 In consideration of the threat of functional limitation following hip fracture63 and that the group of outdoor fallers, described above, also appears to share risk factors which predispose them to future mobility limitation,179 we propose that such individuals require focused efforts during hip fracture rehabilitation. Longer 53 follow-up physiotherapy efforts in the community, that aim to ensure a return to physical activity in outdoor environments, is one such example. Developing walking programs for fall prevention Walking was the most commonly described form of physical activity among the study population, many of whom stressed the importance that physical activity be functional in nature (Study I). At a national level, walking is reported as the most frequently engaged in physical activity among Swedish older adults.39 Similarly, walking is recommended as a way for older people to meet recommended guidelines of 150 minutes of activity per week.5 It is not surprising therefore that walking is the activity most frequently engaged in at the time of hip fracture. While outdoor walks provide emotional as well as physical benefits, they also place greater demands on the coordination of sensory and motoric functions as well as the need to divide attention. For older people with functional limitations, this increased demand on attentional resources can result in poorer gait performance, thus resulting in falls.183 The extent to which walking is effective in falls prevention however is contested, as there is weak evidence from intervention trials for the effect of walking-based programs alone on balance gain among community-dwelling older people.184 The effect of walking on balance appears to be mediated by levels of frailty and fall risk, as positive effects are reported for those at lowrisk for falls,185 whereas negative effects, such as increased fall incidence, is reported among those with a high risk for falls who enrolled in walking intervention programs.186, 187 Additionally, due to the biomechanical properties of the femur, it appears that the weight-bearing effects conferred by walking are insufficient in stimulating bone formation in the area of the femoral cortex which is the weakest in a sideways fall situation.188 A recent metaanalysis reports some positive effects of walking on BMD at the hip, but notes that due to inconsistent results from randomized trials, walking as a single intervention may not be clinically effective at reducing fractures.189 Similarly, best practice guidelines for physical activity programs aimed at preventing falls state that the walking component: may not be crucial, should not be prioritized over balance training and, at worst, can increase fall rates among high-risk individuals.190 So, although walking is the preferred physical activity among older people, there is weak evidence for its preventive effect on falls and subsequent fractures, the threat of which increase with age. The development of walking programs which sufficiently challenge components of balance, as well as muscle strength and weight-bearing, could help bridge the gap between walking as the most preferred exercise form among older people and that which has efficacy enough to be included in fall pre54 vention programs. Group walking programs may be an effective way of motivating lesser active older people who are often socially restricted due to mobility limitations. For less functionally impaired older people, more challenging walking programs in different outdoor terrains and inclines may be more suitable. The gradual introduction of a mixed-impact weight-bearing component, which has been reported to have positive effects on bone density levels at the femoral neck,191 may also have added benefits. Such initiatives could serve to challenge balance, and possibly bone strength, while at the same time be perceived as relevant and meaningful forms of physical activity by older people. Patterns in fall circumstances reflect the health and functional status of people with hip fracture Night-time falls and the use of psychotropic medication Examination of the times at which fall-related hip fractures occurred, within different locational settings, revealed interesting findings in relation to participants’ health characteristics (Study II). We observed a higher night-time incidence of hip fracture in residential care settings, a pattern which we observed to be further associated with psychotropic medication use. This is a previously unreported observation concerning hip fracture patterns. There is support in the literature for the high levels of psychotropic drug prescription in residential care settings.192-194 This medication use is often concomitant with a high prevalence of dementia, itself the most frequent reason for admission to a care homes.195 Whereas others have linked night-time falls and cognitive impairment in psychogeriatric settings,196 we found night-time hip fractures to be associated with psychotropic medications, regardless of the presence of dementia or residential setting. This finding has implications for fracture prevention as medication prescription is a risk factor which may, in many cases, be modified. Randomized trials have demonstrated reductions in falls and improvements in muscle strength and balance when psychotropic drugs are withdrawn.197, 198 Sedatives and anti-depressants were among the most commonly prescribed fall-risk-increasing drugs at the time of hospital admission among community-dwellers in our sample (study II). Antidepressants have been shown to negatively affect sleep patterns 199 and gait patterns,200 as well as cause postural hypotension among older people, which are three possible explanations for the associations between night-time hip fracture and psychotropic drugs which we observed. Confounding by indication is also a factor for consideration in this instance, as depression, even when untreated by medication, is also associated with sleep disturbances,172 as well as delayed reactions, both of which are known risk factors for falls. These factors may in turn explain why depressive symptoms are repeatedly 55 associated with falls in older people, regardless of residential care setting.201 Our findings reinforce those of others, which highlight the importance of carefully weighing the potential advantages of prescribing fall-riskincreasing drugs to older people, against the known disadvantages in terms of increased risk for falls and hip fracture. This is especially relevant when prescribing psychotropic drugs for older people with existing fall risk factors such as: previous falls, mobility limitation and cognitive impairment. Activity at the time of the fall and health status When participants’ health was viewed in relation to the activity they were engaged in at the time of the fall, those who fell during positional change had the poorest overall health status. Studies which have measured gait parameters in older people during the sit-to-walk phase have demonstrated that those with fear of falling tend to adopt a wider step width.202 This postural adjustment along with a slower walking speed then results in a higher sideways velocity.202 Additionally, a fall in a sideways direction strongly predicts the risk of hip fracture occurrence.89 Our findings, in the light of others, highlight possible ways to tailor fall prevention for older people with functional limitations. Focusing, for example, on the assessment and training of muscle strength and power during specific locomotor tasks, such as sit-towalk, during physiotherapy sessions could be an effective starting point. These findings also highlight the importance that physiotherapists and general practitioners ask simple questions regarding eventual symptoms which precede falls, which may provide clues as to causation. Whereas dizziness was frequently described among those who fractured indoors, an equal amount of participants described a symptomless loss of balance and in ten cases, precipitating factors for the fall remained unknown to those who had fractured. Asymptomatic orthostatic hypotension is reported to account for a larger proportion of cases of postural hypotension than the symptomatic form203 and could partly explain falls which are described as involving a symptomless loss of balance or for unknown reasons. There is also evidence that such symptoms can be caused by the effects of drugs with antihypertensive effects,204 which explains why certain cardiovascular medications are categorized as fall-risk-increasing drugs. Even if orthostatic hypotension may be difficult to assess or treat in the clinical setting,205 simple physical or behavioral compensatory strategies can be taught to patients who present with orthostatic intolerance and a history of falls.206 Possible techniques which oppose orthostasis and can be taught as a part of tailored fall prevention include: muscle tensing of the lower larger muscle groups, heel raises, bending forward, leg crossing and breathing exercises.206 56 People in the current study who described their fall in relation to their movements or activities (activity-related) at the time of the fall had, in general, fewer functional limitations. This group also tended to explain falls in terms of their own behavior, as activity-related indoor falls often involved descriptions of hurrying or undertaking more complicated tasks within the home. It is worth noting that this group consisted entirely of female participants, so the influence of gender cannot be ruled out, as men and women have been reported to reason differently in relation to falling and fall risk.116 Explorative studies report, for example, that older women are more likely to blame themselves when describing fall events.207 Alternatively, these fall descriptions may also reflect the time spent engaging in specific tasks as women generally spend more time performing household duties even in higher ages.208 Greater fall rates have been previously reported amongst women while performing household activities,209 a factor which is often explained by the increased exposure time at risk in indoor environments. Furthermore, psychological traits such as risk-taking behavior have been identified as an independent risk factor for falls, but may be difficult to assess from self-reported behavior.210 Such factors have not been in focus in the current thesis but are relevant topics for future investigation. Falls in outdoor environments and good perceived health Falling outdoors is associated with good health and higher physical activity levels, both in our investigations (study II and III) and in those of others.179, 180 It may also be the case that less ‘robust’ older people who fall and fracture outdoors have good perceived health and thereby attribute falls to environmental factors. This is a possible interpretation as to why those who fell outdoors in less challenging (snow-free) circumstances explained their falls in terms of the environment and not in terms of their health or function, despite their relatively high levels of disease, medications and previous falls (Study III). Not only has physical activity in natural environments been associated with emotional benefits such as reduced tension and depression,173 but also been connected with better perceived health in general populations.174 In turn, older people with good perceived health have also been reported as being less likely to perceive themselves as being at risk of falls.116 People who were capable of engaging in outdoor walks may therefore have better perceived health, regardless of having a similar burden of co-existing disease and drug prescription as those whose fall occurred indoors. Gender-related differences It is also possible that gender differences in fall perceptions partly explained why those who fractured outdoors under less challenging circumstances did 57 not describe their fall in terms of health factors. This group comprised of a relatively larger proportion of male participants than other groups, and others have shown that older men are more likely to have a low perceived risk of falling116 and are less likely to ‘self-blame’ in relation to falls than women are.207 The greater propensity for men to fall outdoors, which we observed, has also been previously reported.209 This pattern may, like that of female participants falling during household activities, be explained by increased exposure time in outdoor environments. Gender differences have however not been in focus in the current thesis, and subgroup analysis of fall patterns in relation to health status, may have been hindered by the small total number of men in study III. Nonetheless, we did observe (Study IV) that in relation to the disease categories with the highest impact on hip fracture risk (Previous injury and Circulatory disease), the relative risks were somewhat higher for male subjects whereas the absolute risks were higher in both cases among females. The impact of co-existing disease on hip fracture and physical activity beliefs Disease as a predisposing risk factor for hip fracture Investigation of fall descriptions (study III) highlighted ways in which symptoms of disease acted as precipitating factors in the acute onset of fall-related hip fractures. By comparing the disease history of the entire hip fracture cohort to that of the source population in the region (study IV), we could calculate estimates for co-existing disease as predisposing risk factors for hip fracture. We observed positive associations between all disease categories and hip fracture, a pattern supported by those who have investigated a smaller range of disease exposures.211, 212 In the literature, a greater focus has been placed on calculating the relative risks posed by medication use for falls and hip fractures, than that posed by underlying disease.6, 85, 86 This is despite evidence that co-existing disease may pose a stronger population proportionate attributable risk for falls, than medications do.213 Additionally, inclusion of disease-related data has been shown to substantially improve the predictive ability of tools developed to identify individuals at high risk for hip fracture.214 Nonetheless, it should also be considered that the impact of coexisting disease may be reported indirectly when the exposures under investigation include functional impairments, such as gait or balance, and the risk they pose for falls.83, 215 Nevertheless, it is recognized that adults with a high prevalence of co-existing disease, more frequently experience falls and hip fracture216, 217 and that an overlap exists between comorbid disease, disability and frailty in older populations.218 It is however, uncommon that the separate effects of underlying disease categories for hip fracture are reported, and this 58 knowledge should serve as a first step in identifying those who are at high risk and deserve further assessment. Previous injury as a risk factor for hip fracture After cardiovascular disease, previous injury and the sub-group previous fracture were the disease categories with the highest relative and absolute risks for hip fracture. Having a history of fracture is a well-established predictive factor for future hip fracture,219, 220 and is a factor which may signal reduced bone mineral density as well as an increased tendency to fall. The category ‘Previous injury’ can be considered to largely incorporate prior falls which required hospitalization, as the vast majority of injury-related hospitalizations among older people are fall-related.221, 222 That previous falls are predictive of future falls is well supported in the literature,76, 106 as well as implemented in clinical practice where questions relating to falls are incorporated in several formal fall risk assessment tools, such as the Downton fall risk index,223 STRATIFY,224 and the Fall Risk Assessment Tool.225 Additionally, the Swedish Association of Local Authorities recommends that, upon admission to hospital environments, all people aged 65 years and older be asked whether they have fallen in the previous year, as one of the two screening questions which initially assess fall risk.226 Perceptions of the relevance of previous falls Whether older adults share the understandings of researchers and clinicians regarding the relevance of previous falls however is unclear. Approximately half of all participants who were interviewed (study III) had experienced a fall in the year previous to fracture. Nevertheless, a relatively smaller proportion of participants answered ‘yes’ to having had a fear of falling prior to the fracture and this was especially the case among certain groups who fractured outdoors. It has been previously reported that community-dwelling older people perceive their risk of falls as low, in spite of having a history of falls.116, 227 Although we do not have data concerning the extent to which previous falls were followed up in our cohort, others report that older adults tend not to seek medical help for falls as 75−80 % of non-injurious falls go unreported to health professionals.228-230 When help-seeking behaviors among older people have been explored, it appears that previous falls are often rationalized as not being a legitimate enough reason to warrant seeking medical help.231 This is in accordance with evidence that older people are less likely to seek medical help or engage in preventive behavior, when they perceive that symptoms are a normal part of aging.232 As we observed (study III), even when falls have severe consequences, in the majority of cases, people do not perceive these events as being related to personal health factors, which may further explain why seeking medical advice for non59 injurious falls is considered unnecessary. Another reason for concern is that not communicating about falls also coincides with activity curtailment233 and older people commonly describe avoiding future falls by ‘taking care’, a precaution which often involves a level of physical activity avoidance.121, 234, 235 Our findings, in light of those of others, highlight the importance that health professionals working with older people include questions pertaining to prior falls when recording health details, as this information may not be considered relevant on the part of the older person. Charting previous falls and including fall incidence in the follow-up assessments of the effects of rehabilitation or medical treatments should be given priority among physiotherapists, nurses and doctors working in primary and community care settings. Such approaches will also help to convey the perspective that falls in later life are often symptomatic of underlying conditions which can be treated, and not an inevitable consequence of ageing, which is a commonly held belief among those who fall119, 235, 236 and was often expressed by participants during the interviews in study III. The impact of cardiovascular and other diseases We observed a relative risk of 6.28 (95% CI; 5.06−7.79) for circulatory diseases, which incorporates the combined effects of hypertensive, heart and cerebrovascular diseases. Others have reported relative risk increases for stroke ranging between 2− and 12−fold, and for heart failure ranging between 1.3−3 Risk ratio/Hazard ratio increases, dependent on age, sex and time since disease onset.147, 237, 238 212 Functional impairments caused by stroke increase the propensity for falls,76 and for fall impact on the paretic side where the bone density of the femoral neck is often compromised due to reduced weight-bearing capacity on the hemiplegic leg.239 Cardiovascular conditions such as congestive heart failure or ischaemic heart disease, which are captured well by register data, were prevalent in approximately one fifth of subjects (according to medical journal data) in Study II. The circulatory effects of heart disease, or of the drugs which treat disease, on cerebral perfusion include for example postural hypotension, which was reported most frequently by those who fractured indoors in study III. Other diseases categories such as genitourinary and digestive conditions were highly prevalent among people with hip fracture and posed a similar absolute risk (approximately 22/1000 individuals) as that posed by cardiovascular diseases, the most prevalent of all diseases. Urinary incontinence is a previously established risk factor for falls,76 but often overlooked when predisposing factors for hip fracture are investigated.69, 79, 212, 240 Previous investigations of the impact of urinary incontinence on fractures, have focused on community-dwelling women and report lower risk estimates in relation to ‘non spine’ fractures (Relative hazard:1.34, 95% CI;1.06−1.69)241 60 and hip fractures (RR: 2.42, 95% CI;1.23−4.74). The higher risk estimates reported in our study may reflect our strict population-based design which served to highlight conditions more prevalent among older, frailer individuals in residential care facilities, a group often excluded in epidemiological studies which base risk calculations on community-dwelling samples. Urinary incontinence is defined as a geriatric syndrome which, like falling, is a condition involving the accumulation of impairments in multiple bodily systems.242 For these reasons falls and urinary incontinence share many predisposing risk factors such as impairments in mobility, cognition and medication use and depression.2, 242 Perceptions of disease in relation to falls and physical activity In terms of participant perceptions (Study III), the presence of disease was, in the majority of cases, not discussed in relation to the occurrence of the fall which led to hip fracture. Interestingly, that medication could have potentially contributed to the fall event was not either discussed by any participant during the interviews. It was only in cases of indoor falls when obvious symptoms such as dizziness or joint pain were experienced that co-existing conditions, such as low blood pressure and arthritis, were discussed as precipitating factors. We also observed that those who had no explanation for their fall had a high prevalence of underlying disease, as well as previous falls. It has been reported that the transient nature of certain cardiovascular conditions, such as postural hypotension and syncope, can be methodologically difficult to capture in study trials.78, 243 Our findings lend further support to these observations, as those who described unexplained falls, often in relation to positional change, did not discuss underlying health conditions as a potential factors in fall causation. This in turn may explain conflicting evidence concerning the impact of postural hypotension or drop attacks on falls and hip fracture. Encouraging older adults, especially those with repeated and unexplained falls, to reflect upon potential contributory factors for the fall may itself be a first step in fall prevention. Although rarely investigated, one study has shown that beliefs about fall causation are predictive of functional outcome following the fracture.244 Another study, which performed follow-up interviews to detect how fall beliefs changed over time, report that older people who reflected and aimed to understand why they fell also developed strategies to prevent falls, whereas those who neither reflected or understood why their fall occurred went on to restrict their activities over time.235 In relation to physical activity perceptions, there was diversity in the way in which symptoms of disease were perceived as barriers or motivating factors. Whereas disease symptoms such as joint pain or breathlessness caused the cessation or reduction of certain physical activities among lesser active par61 ticipants; active people, on the other hand, interpreted similar symptoms as a motivating factor to stay physically active. As is the case with help-seeking behaviours previously described, beliefs about aging have also been shown to be associated with physical activity engagement in later life. 245 Interestingly, participants in study I, regardless of physical activity level, did not discuss the avoidance of disease as a motivating factor to being physically active. This finding is once again a reminder that those promoting exercise in community-settings will be better served to focus on positive messages which appeal to older people’s desire to remain independent as opposed to those focusing on risk avoidance. 62 Main Conclusions Exposure to natural, outdoor environments is an important motivating factor for physical activity among Swedish older adults, regardless of their described level of activity. Physical activity which is functional in nature is that perceived as most meaningful to certain groups of older people and the uptake of physical activity in retirement years can provide a way of establishing a new self-identity and adapt to lifestyle changes (Study I). People living in residential care facilities are more likely to fracture their hip during night-time hours, unlike community-dwelling people, who experience a daytime peak in hip fracture occurrence. Hip fractures which occur during night-time hours also appear to be associated with the use of psychotropic medications, regardless of the presence of dementia, or whether the person lives in residential care or in the community. Community-dwelling people with poor health and function may be at greatest risk for hip fractures during positional changes indoors, such as sit-to-walk, and these people also perceive their fracture to be precipitated by poor health. The majority of people with hip fracture however, describe their falls as being precipitated by environmental factors, despite having relatively high levels of co-existing disease and medication use (Study II and III). All disease categories appear to be positively associated with hip fracture, and cardiovascular disease and previous injury pose the highest relative and absolute risks for fracture. Less prevalent conditions however, such as genitourinary, digestive and eye diseases pose absolute risks of a similar impact to that posed by cardiovascular diseases (Study IV). The use of a mixed method approach to investigate the multiple perspectives of hip fracture occurrence can reveal possible contributory factors for these events and ways in which they may interact. Such knowledge may in turn inform the tailoring of fall prevention strategies for those at risk. 63 Clinical implications Our findings emphasise that knowledge of the context of hip fracture events is essential to understanding the problem. In the clinical setting, posing question concerning the fall circumstances of older people need not be time or resource consuming, and may reveal patterns relating to health and function which contributed to the fall. Similarly, taking people’s fall perceptions into consideration may also be a key factor in understanding how best to help them avoid the next potential fall event. That those who live in residential care are more prone to hip fracture during night-time hours, may signal important structural aspects, such as staffing levels and medication routines, which are akin to modification in these settings. Medication prescription is a modifiable risk factor for falls and psychotropic drug withdrawal has been shown to be effective in falls prevention.197 This strategy may also be especially important among frail nursing home dwellers, where evidence is still lacking for the effectiveness of exercise interventions in falls reduction.246 Additionally, higher staffing levels and education appear to be related to lower levels of psychotropic drug prescription in nursing homes,247 which is a factor of relevance in the context of findings in this thesis. Communitydwelling people with functional limitations also require information regarding the increased fall risks of drugs such as anti-depressants or sedatives, so that they can make informed decisions regarding their potential benefits and dangers. People hospitalized with hip fracture, and older people who fall recurrently are a heterogonous group, who therefore require different caring pathways in terms of rehabilitation and fall prevention. Among community-dwelling people with frail health, the most effective starting point for rehabilitation or fall prevention may involve focused strength and balance training, aimed at specifically improving capacity during positional transfers. A recent study which incorporated graded tasks such as sit-to-stand reports improvements in strength and balance among older subjects.248 High-intensity functional exercises have also been reported to prevent functional decline in residential care settings among people with dementia.249 64 Most importantly, health care professionals working with older adults need to systematically document fall histories and apply follow-up strategies concerning falls, as this information may not be considered relevant by the person who has experienced them. Although not investigated in direct relation to each other in the current thesis, findings regarding physical activity perceptions in study I are applicable to the promotion of exercise as a means of fall risk reduction. Communitydwellers in study I did not articulate that it was avoidance of disease or negative health outcomes, that motivated them to stay physically active. These tendencies are also reflected in reports that messages concerning fall avoidance are less attractive to those ‘at-risk’, than messages which appeal to the desire to remain strong and independent.122 ‘Staying healthy and strong’ is a message which should be kept in mind by health professionals who work with older people. Similarly, older people who are clinically defined as being ‘at-risk’ may neither perceive themselves as being at risk, nor consider risk avoidance to have precedence over maintaining functional independence. Indeed exposing themselves to risk may be an important component of good perceived health. Exercise promotion amongst those at the transitional pre-retirement stage, focusing on functional exercise and challenging walking programs in outdoor environments, are strategies which could appeal to older populations. If walking programs can be developed that sufficiently challenge balance, strength and endurance, this could bridge the gap between walking as the exercise of choice among older people and that which has adequate efficacy for inclusion in fall prevention programs. The way in which older adults perceive PA or falls may differ from the opinions of professionals working in PA promotion or falls prevention. For the older person, physical health benefits may, for example, not be the primary motivator to staying physically active and maintaining independence may be more important than evaluating the risk for falls in certain situations. Nevertheless, a merging of knowledge from both perspectives should benefit the development of effective strategies to promote physical activity and prevent fall-related hip fracture, which also appeal to older adults. 65 Acknowledgments First of all, I wish to thank all the participants for generously sharing their experiences, whose personal accounts were the foundation for a large part of this work. Writing this thesis was not something I did alone but depended on the guidance, support and encouragement of certain people, who I wish to thank: Liisa Byberg, my main supervisor – not just for your insightful guidance on all things epidemiological and statistical but for the way in which you gave it. Thank you for always having the time and for always making me feel that I hadn’t asked the question before (when I knew of course that I had). I am grateful for getting the chance to learn from someone who is so truly competent and kind. Karl Michaëlsson, my co-supervisor – whose enthusiasm to look beyond the boney aspects of hip fracture was inspiring and contagious. I appreciate that you trusted in my capability from the start and enabled me to work on this full-time. I am mostly thankful however, for your generous praise and words of encouragement over the years, which have often made all the difference. Anna Cristina Åberg, my co-supervisor – for introducing me to the world of research and for your valuable skills in the qualitative field, astute attention to detail, yet sound advice that I should always try to lift the story. Thank you for the time and effort you have dedicated to helping me find my research feet, especially in the early stages of this journey. Håkan Melhus, my co-supervisor – for your interested, engaged and sharp feedback on all the manuscripts. Thanks to Carina Fredriksson for helping out with the data collection when the work-load (and my body weight) got heavy. I’d also like to thank the doctoral students, some of whom have made it to the other side, who I’ve gotten to know along the way. Ylva Cedervall, thanks for all the support about all things research and non-research related. You have been a great help! Thanks as well to Erika Ax, Erika Olsson and Johanna Törmä for all the laughs during the time and feedback on the final manuscript. To Mwenya, Markus and Lisa for taking the time to read the draft, and give valuable feedback when I needed it. To Alison, Louise and Yvonne who are always till closest ones, despite 1000 miles. To all my friends in Stockholm for good times and practical help with the puzzle. 66 Agneta Wetterberg and all the paramedical staff at Brommageriatriken- an inspiring (and fun) workplace, where I was let loose into the Swedish healthcare system (despite some rusty language skills in the early days). Special thanks to Åke Seiger and everyone at the FoUU department at Stockholms Sjukhem, for providing me with a safe haven, to work during the final half year, when working at the kitchen table no longer was an option. Tack till C, som har blivit en god vän och en riktig träningsinspiration till alla över (och under) 80 år. To my parents, for being wise enough to smile and nod to the many decisions I’ve made, but most of all for instilling the belief that it can be done. To Aisling for breaking the academic glass ceiling, before any of the rest of us had a notion what it involved, to Aaron for the images over the years, and Cillian, who will be getting this book as a Christmas present. Thanks to Linus Gårdfeldt for excellent (and painstaking) transcription work – not everyone gets a poet for that part. To Niklas, the best choice I’ve made. For all the love and encouragement throughout and for being the only one I know with an instantly wise answer to questions ranging from − ‘What’s the point of it all?’ to ‘What’s the correct use of the semi-colon?’ Thank you for always being on my side. Lastly, to those of whom I am most proud, my daughters Elsa and Moira, my favourite people and the ones who make it all worthwhile. 67 References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] 68 Lamb SE, Jorstad-Stein EC, Hauer K, Becker C. Development of a common outcome data set for fall injury prevention trials: the Prevention of Falls Network Europe consensus. J Am Geriatr Soc. 2005;53: 1618-1622. Inouye SK, Studenski S, Tinetti ME, Kuchel GA. Geriatric syndromes: clinical, research, and policy implications of a core geriatric concept. J Am Geriatr Soc. 2007;55: 780-791. World Health Organisation. Global Recommendations on Physical Actvitiy for Health. Switzerland: 2010. Hallal PC, Andersen LB, Bull FC, et al. Global physical activity levels: surveillance progress, pitfalls, and prospects. Lancet. 2012;380: 247-257. Nelson ME, Rejeski WJ, Blair SN, et al. Physical activity and public health in older adults: recommendation from the American College of Sports Medicine and the American Heart Association. Medicine & Science in Sports & Exercise 2007;39: 1435-1445. Ensrud KE, Blackwell TL, Mangione CM, et al. Central nervous system-active medications and risk for falls in older women. J Am Geriatr Soc. 2002;50: 1629-1637. Chou CH, Hwang CL, Wu YT. Effect of exercise on physical function, daily living activities, and quality of life in the frail older adults: a meta-analysis. Arch Phys Med Rehabil. 2012;93: 237-244. Lynch NA, Metter EJ, Lindle RS, et al. Muscle quality. I. Ageassociated differences between arm and leg muscle groups. Journal of applied physiology. 1999;86: 188-194. Michaelsson K, Olofsson H, Jensevik K, et al. Leisure physical activity and the risk of fracture in men. PLoS Med. 2007;4: e199. Moayyeri A. The association between physical activity and osteoporotic fractures: a review of the evidence and implications for future research. Annals of epidemiology. 2008;18: 827-835. Cauley JA, Harrison SL, Cawthon PM, et al. Objective measures of physical activity, fractures and falls: the osteoporotic fractures in men study. J Am Geriatr Soc. 2013;61: 1080-1088. Magaziner J, Fredman L, Hawkes W, et al. Changes in functional status attributable to hip fracture: a comparison of hip fracture patients to community-dwelling aged. Am J Epidemiol. 2003;157: 1023-1031. Tinetti ME, Williams CS. Falls, injuries due to falls, and the risk of admission to a nursing home. N Engl J Med. 1997;337: 1279-1284. [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] Michaelsson K, Nordstrom P, Nordstrom A, et al. Impact of hip fracture on mortality: a cohort study in hip fracture discordant identical twins. J Bone Miner Res. 2014;29: 424-431. Jarvinen TL, Sievanen H, Khan KM, Heinonen A, Kannus P. Shifting the focus in fracture prevention from osteoporosis to falls. BMJ. 2008;336: 124-126. Sambrook P, Cooper C. Osteoporosis. Lancet. 2006;367: 2010-2018. Lyles KW, Colon-Emeric CS, Magaziner JS, et al. Zoledronic Acid in Reducing Clinical Fracture and Mortality after Hip Fracture. N Engl J Med. 2007;357: nihpa40967. Siris ES, Chen YT, Abbott TA, et al. Bone mineral density thresholds for pharmacological intervention to prevent fractures. Archives of internal medicine. 2004;164: 1108-1112. Wells GA, Cranney A, Peterson J, et al. Alendronate for the primary and secondary prevention of osteoporotic fractures in postmenopausal women. Cochrane Database Syst Rev. 2008: CD001155. Wells G, Cranney A, Peterson J, et al. Risedronate for the primary and secondary prevention of osteoporotic fractures in postmenopausal women. Cochrane Database Syst Rev. 2008: CD004523. McClung MR, Boonen S, Torring O, et al. Effect of denosumab treatment on the risk of fractures in subgroups of women with postmenopausal osteoporosis. J Bone Miner Res. 2012;27: 211-218. Sherrington C, Whitney JC, Lord SR, Herbert RD, Cumming RG, Close JC. Effective exercise for the prevention of falls: a systematic review and meta-analysis. J Am Geriatr Soc. 2008;56: 2234-2243. Gillespie LD, Robertson MC, Gillespie WJ, et al. Interventions for preventing falls in older people living in the community. Cochrane Database Syst Rev. 2009: CD007146. El-Khoury F, Cassou B, Charles MA, Dargent-Molina P. The effect of fall prevention exercise programmes on fall induced injuries in community dwelling older adults: systematic review and metaanalysis of randomised controlled trials. BMJ. 2013;347: f6234. Surgeon General Atlanta. Physical activity and health. Volume 2009, 1996. Pedersen BK, Saltin B. Evidence for prescribing exercise as therapy in chronic disease. Scandinavian Journal of Medicine & Science in Sports 2006;16 Suppl 1: 3-63. Thompson PD, Buchner D, Pina IL, et al. Exercise and physical activity in the prevention and treatment of atherosclerotic cardiovascular disease: a statement from the Council on Clinical Cardiology (Subcommittee on Exercise, Rehabilitation, and Prevention) and the Council on Nutrition, Physical Activity, and Metabolism (Subcommittee on Physical Activity). Circulation. 2003;107: 3109-3116. 69 [28] [29] [30] [31] [32] [33] [34] [35] [36] [37] [38] [39] [40] [41] 70 Diabetes Prevention Program Research G, Knowler WC, Fowler SE, et al. 10-year follow-up of diabetes incidence and weight loss in the Diabetes Prevention Program Outcomes Study. Lancet. 2009;374: 1677-1686. Strawbridge WJ, Deleger S, Roberts RE, Kaplan GA. Physical activity reduces the risk of subsequent depression for older adults. Am J Epidemiol. 2002;156: 328-334. Schnohr P, Lange P, Scharling H, Jensen JS. Long-term physical activity in leisure time and mortality from coronary heart disease, stroke, respiratory diseases, and cancer. The Copenhagen City Heart Study. European journal of cardiovascular prevention and rehabilitation : official journal of the European Society of Cardiology, Working Groups on Epidemiology & Prevention and Cardiac Rehabilitation and Exercise Physiology. 2006;13: 173-179. van der Bij AK, Laurant MG, Wensing M. Effectiveness of physical activity interventions for older adults: a review. American journal of preventive medicine. 2002;22: 120-133. Byberg L, Melhus H, Gedeborg R, et al. Total mortality after changes in leisure time physical activity in 50 year old men: 35 year follow-up of population based cohort. BMJ. 2009;338: b688. Lord SR, Ward JA. Age-associated differences in sensori-motor function and balance in community dwelling women. Age Ageing. 1994;23: 452-460. Fries JF. Measuring and monitoring success in compressing morbidity. Ann Intern Med. 2003;139: 455-459. Liu CJ, Latham NK. Progressive resistance strength training for improving physical function in older adults. Cochrane Database Syst Rev. 2009: CD002759. Arnold P, Bautmans I. The influence of strength training on muscle activation in elderly persons: a systematic review and meta-analysis. Experimental gerontology. 2014;58: 58-68. Hamilton CJ, Swan VJ, Jamal SA. The effects of exercise and physical activity participation on bone mass and geometry in postmenopausal women: a systematic review of pQCT studies. Osteoporos Int. 2010;21: 11-23. World Health Organisation. Global Health Risks: Mortality and Burden of Disease Attributable to Selected Major Risks. Geneva: 2009. The Swedish National Institute of Public Health. Living habits, report 2010 (Levnadsvanor, lägesrapport 2010). Statens folkhälsoinstitut: 2010. The Swedish National institute of Public Health. Living habits, report 2011 (Levnadsvanor, lägesrapport 2011). Östersund: Statens Folkhälsoinstitut, 2012. Harris TJ, Owen CG, Victor CR, Adams R, Cook DG. What factors are associated with physical activity in older people, assessed objectively by accelerometry? Br J Sports Med. 2009;43: 442-450. [42] [43] [44] [45] [46] [47] [48] [49] [50] [51] [52] [53] [54] [55] Arnardottir NY, Koster A, Van Domelen DR, et al. Objective measurements of daily physical activity patterns and sedentary behaviour in older adults: Age, Gene/Environment SusceptibilityReykjavik Study. Age Ageing. 2013;42: 222-229. Lim K, Taylor L. Factors associated with physical activity among older people--a population-based study. Prev Med. 2005;40: 33-40. Lees FD, Clark PG, Nigg CR, Newman P. Barriers to exercise behavior among older adults: a focus-group study. Journal of Aging and Physical Activity. 2005;13: 23-33. Van Holle V, Deforche B, Van Cauwenberg J, et al. Relationship between the physical environment and different domains of physical activity in European adults: a systematic review. BMC public health. 2012;12: 807. Horne M, Skelton DA, Speed S, Todd C. Perceived barriers to initiating and maintaining physical activity among South Asian and White British adults in their 60s living in the United Kingdom: a qualitative study. Ethnicity & health. 2013. Belza B, Walwick J, Shiu-Thornton S, Schwartz S, Taylor M, LoGerfo J. Older adult perspectives on physical activity and exercise: voices from multiple cultures. Preventing Chronic Disease. 2004;1: A09. Schmidt JA, Gruman C, King MB, Wolfson LI. Attrition in an exercise intervention: a comparison of early and later dropouts. J Am Geriatr Soc. 2000;48: 952-960. Nyman SR, Victor CR. Older people's participation in and engagement with falls prevention interventions in community settings: an augment to the Cochrane systematic review. Age Ageing. 2012;41: 16-23. Johnell O, Kanis JA, Jonsson B, Oden A, Johansson H, De Laet C. The burden of hospitalised fractures in Sweden. Osteoporos Int. 2005;16: 222-228. Johnell O, Kanis JA. An estimate of the worldwide prevalence and disability associated with osteoporotic fractures. Osteoporos Int. 2006;17: 1726-1733. Kanis JA, Oden A, McCloskey EV, et al. A systematic review of hip fracture incidence and probability of fracture worldwide. Osteoporos Int. 2012;23: 2239-2256. Kanis JA, Johnell O, Oden A, Jonsson B, Dawson A, Dere W. Risk of hip fracture derived from relative risks: an analysis applied to the population of Sweden. Osteoporos Int. 2000;11: 120-127. Kanis JA, Johnell O, De Laet C, Jonsson B, Oden A, Ogelsby AK. International variations in hip fracture probabilities: implications for risk assessment. J Bone Miner Res. 2002;17: 1237-1244. Roy DK, Pye SR, Lunt M, et al. Falls explain between-center differences in the incidence of limb fracture across Europe. Bone. 2002;31: 712-717. 71 [56] [57] [58] [59] [60] [61] [62] [63] [64] [65] [66] [67] [68] 72 Melhus H, Snellman G, Gedeborg R, et al. Plasma 25hydroxyvitamin D levels and fracture risk in a community-based cohort of elderly men in Sweden. The Journal of clinical endocrinology and metabolism. 2010;95: 2637-2645. Eklund F, Nordstrom A, Neovius M, Svensson O, Nordstrom P. Variation in fracture rates by country may not be explained by differences in bone mass. Calcif Tissue Int. 2009;85: 10-16. Norton R, Campbell AJ, Reid IR, et al. Residential status and risk of hip fracture. Age Ageing. 1999;28: 135-139. Roche JJ, Wenn RT, Sahota O, Moran CG. Effect of comorbidities and postoperative complications on mortality after hip fracture in elderly people: prospective observational cohort study. BMJ. 2005;331: 1374. Johnell O, Kanis JA. An estimate of the worldwide prevalence, mortality and disability associated with hip fracture. Osteoporos Int. 2004;15: 897-902. Leibson CL, Tosteson AN, Gabriel SE, Ransom JE, Melton LJ. Mortality, disability, and nursing home use for persons with and without hip fracture: a population-based study. J Am Geriatr Soc. 2002;50: 1644-1650. Abrahamsen B, van Staa T, Ariely R, Olson M, Cooper C. Excess mortality following hip fracture: a systematic epidemiological review. Osteoporos Int. 2009;20: 1633-1650. Salpakoski A, Tormakangas T, Edgren J, et al. Walking recovery after a hip fracture: a prospective follow-up study among community-dwelling over 60-year old men and women. BioMed research international. 2014;2014: 289549. Holt G, Smith R, Duncan K, Hutchison JD, Gregori A. Gender differences in epidemiology and outcome after hip fracture: evidence from the Scottish Hip Fracture Audit. The Journal of bone and joint surgery British volume. 2008;90: 480-483. Borgstrom F, Zethraeus N, Johnell O, et al. Costs and quality of life associated with osteoporosis-related fractures in Sweden. Osteoporos Int. 2006;17: 637-650. Visschedijk J, van Balen R, Hertogh C, Achterberg W. Fear of falling in patients with hip fractures: prevalence and related psychological factors. Journal of the American Medical Directors Association. 2013;14: 218-220. Visschedijk J, Achterberg W, Van Balen R, Hertogh C. Fear of falling after hip fracture: a systematic review of measurement instruments, prevalence, interventions, and related factors. J Am Geriatr Soc. 2010;58: 1739-1748. Givens JL, Sanft TB, Marcantonio ER. Functional recovery after hip fracture: the combined effects of depressive symptoms, cognitive impairment, and delirium. J Am Geriatr Soc. 2008;56: 1075-1079. [69] [70] [71] [72] [73] [74] [75] [76] [77] [78] [79] [80] [81] [82] [83] Cummings SR, Nevitt MC, Browner WS, et al. Risk factors for hip fracture in white women. Study of Osteoporotic Fractures Research Group. N Engl J Med. 1995;332: 767-773. Mussolino ME, Looker AC, Madans JH, Langlois JA, Orwoll ES. Risk factors for hip fracture in white men: the NHANES I Epidemiologic Follow-up Study. J Bone Miner Res. 1998;13: 918924. Chen JS, Simpson JM, March LM, et al. Risk factors for fracture following a fall among older people in residential care facilities in Australia. J Am Geriatr Soc. 2008;56: 2020-2026. Cauley JA, Lui LY, Genant HK, et al. Risk factors for severity and type of the hip fracture. J Bone Miner Res. 2009;24: 943-955. Grisso JA, Kelsey JL, Strom BL, et al. Risk factors for falls as a cause of hip fracture in women. The Northeast Hip Fracture Study Group. N Engl J Med. 1991;324: 1326-1331. Kauppi M, Stenholm S, Impivaara O, Maki J, Heliovaara M, Jula A. Fall-related risk factors and heel quantitative ultrasound in the assessment of hip fracture risk: a 10-year follow-up of a nationally representative adult population sample. Osteoporos Int. 2014;25: 1685-1695. Masud T, Morris RO. Epidemiology of falls. Age Ageing. 2001;30 Suppl 4: 3-7. Deandrea S, Lucenteforte E, Bravi F, Foschi R, La Vecchia C, Negri E. Risk factors for falls in community-dwelling older people: a systematic review and meta-analysis. Epidemiology. 2010;21: 658668. NICE. Clinical Practice Guideline for the Assessment and Prevention of Falls in Older People. London: Royal College of Nursing, 2004. Lord S, Sherrington C, Menz HB, Close J. Falls in Older People. Cambridge, United Kingdom. : Cambridge University Press, 2007. Albrand G, Munoz F, Sornay-Rendu E, DuBoeuf F, Delmas PD. Independent predictors of all osteoporosis-related fractures in healthy postmenopausal women: the OFELY study. Bone. 2003;32: 78-85. Stolze H, Klebe S, Zechlin C, Baecker C, Friege L, Deuschl G. Falls in frequent neurological diseases--prevalence, risk factors and aetiology. Journal of neurology. 2004;251: 79-84. Kanis JA, Johnell O, De Laet C, et al. A meta-analysis of previous fracture and subsequent fracture risk. Bone. 2004;35: 375-382. Wagner H, Melhus H, Gedeborg R, Pedersen NL, Michaelsson K. Simply ask them about their balance--future fracture risk in a nationwide cohort study of twins. Am J Epidemiol. 2009;169: 143149. Muir SW, Berg K, Chesworth B, Klar N, Speechley M. Quantifying the magnitude of risk for balance impairment on falls in community- 73 [84] [85] [86] [87] [88] [89] [90] [91] [92] [93] [94] [95] [96] [97] [98] 74 dwelling older adults: a systematic review and meta-analysis. Journal of clinical epidemiology. 2010;63: 389-406. Dargent-Molina P, Favier F, Grandjean H, et al. Fall-related factors and risk of hip fracture: the EPIDOS prospective study. Lancet. 1996;348: 145-149. Hartikainen S, Lonnroos E, Louhivuori K. Medication as a risk factor for falls: critical systematic review. J Gerontol A Biol Sci Med Sci. 2007;62: 1172-1181. Kragh A, Elmstahl S, Atroshi I. Older adults' medication use 6 months before and after hip fracture: a population-based cohort study. J Am Geriatr Soc. 2011;59: 863-868. Chen JS, Sambrook PN, Simpson JM, et al. Risk factors for hip fracture among institutionalised older people. Age Ageing. 2009;38: 429-434. Greenspan SL, Myers ER, Kiel DP, Parker RA, Hayes WC, Resnick NM. Fall direction, bone mineral density, and function: risk factors for hip fracture in frail nursing home elderly. Am J Med. 1998;104: 539-545. Kannus P, Leiponen P, Parkkari J, Palvanen M, Jarvinen M. A sideways fall and hip fracture. Bone. 2006;39: 383-384. Wei TS, Hu CH, Wang SH, Hwang KL. Fall characteristics, functional mobility and bone mineral density as risk factors of hip fracture in the community-dwelling ambulatory elderly. Osteoporos Int. 2001;12: 1050-1055. Parkkari J, Kannus P, Palvanen M, et al. Majority of hip fractures occur as a result of a fall and impact on the greater trochanter of the femur: a prospective controlled hip fracture study with 206 consecutive patients. Calcif Tissue Int. 1999;65: 183-187. Marshall D, Johnell O, Wedel H. Meta-analysis of how well measures of bone mineral density predict occurrence of osteoporotic fractures. BMJ. 1996;312: 1254-1259. Kanis JA. Diagnosis of osteoporosis and assessment of fracture risk. Lancet. 2002;359: 1929-1936. Cummings SR, Bates D, Black DM. Clinical use of bone densitometry: scientific review. JAMA. 2002;288: 1889-1897. Stone KL, Seeley DG, Lui LY, et al. BMD at multiple sites and risk of fracture of multiple types: long-term results from the Study of Osteoporotic Fractures. J Bone Miner Res. 2003;18: 1947-1954. Wainwright SA, Marshall LM, Ensrud KE, et al. Hip fracture in women without osteoporosis. The Journal of clinical endocrinology and metabolism. 2005;90: 2787-2793. Stone KL, Seeley DG, Lui LY, et al. BMD at multiple sites and risk of fracture of multiple types: long-term results from the Study of Osteoporotic Fractures. J Bone Miner Res. 2003;18: 1947-1954. Roberts BJ, Thrall E, Muller JA, Bouxsein ML. Comparison of hip fracture risk prediction by femoral aBMD to experimentally measured factor of risk. Bone. 2010;46: 742-746. [99] [100] [101] [102] [103] [104] [105] [106] [107] [108] [109] [110] [111] [112] Johnell O, Kanis JA, Oden A, et al. Predictive value of BMD for hip and other fractures. J Bone Miner Res. 2005;20: 1185-1194. Melton LJ, 3rd, Thamer M, Ray NF, et al. Fractures attributable to osteoporosis: report from the National Osteoporosis Foundation. J Bone Miner Res. 1997;12: 16-23. Mackey DC, Lui LY, Cawthon PM, et al. High-trauma fractures and low bone mineral density in older women and men. JAMA. 2007;298: 2381-2388. Bouxsein ML, Szulc P, Munoz F, Thrall E, Sornay-Rendu E, Delmas PD. Contribution of trochanteric soft tissues to fall force estimates, the factor of risk, and prediction of hip fracture risk. J Bone Miner Res. 2007;22: 825-831. Laing AC, Robinovitch SN. Low stiffness floors can attenuate fallrelated femoral impact forces by up to 50% without substantially impairing balance in older women. Accident; analysis and prevention. 2009;41: 642-650. Simpson AH, Lamb S, Roberts PJ, Gardner TN, Evans JG. Does the type of flooring affect the risk of hip fracture? Age Ageing. 2004;33: 242-246. Keyak JH, Skinner HB, Fleming JA. Effect of force direction on femoral fracture load for two types of loading conditions. J Orthop Res. 2001;19: 539-544. Berg WP, Alessio HM, Mills EM, Tong C. Circumstances and consequences of falls in independent community-dwelling older adults. Age Ageing. 1997;26: 261-268. Zecevic AA, Salmoni AW, Speechley M, Vandervoort AA. Defining a fall and reasons for falling: comparisons among the views of seniors, health care providers, and the research literature. Gerontologist. 2006;46: 367-376. Lord SR, Menz HB, Sherrington C. Home environment risk factors for falls in older people and the efficacy of home modifications. Age Ageing. 2006;35 Suppl 2: ii55-ii59. Letts L, Moreland J, Richardson J, et al. The physical environment as a fall risk factor in older adults: Systematic review and metaanalysis of cross-sectional and cohort studies. Australian occupational therapy journal. 2010;57: 51-64. Gill TM, Williams CS, Tinetti ME. Environmental hazards and the risk of nonsyncopal falls in the homes of community-living older persons. Medical care. 2000;38: 1174-1183. Allander E, Gullberg B, Johnell O, Kanis JA, Ranstam J, Elffors L. Circumstances around the fall in a multinational hip fracture risk study: a diverse pattern for prevention. MEDOS Study Group). Mediterranean Osteoporosis Study. Accident; analysis and prevention. 1998;30: 607-616. Norton R, Campbell AJ, Lee-Joe T, Robinson E, Butler M. Circumstances of falls resulting in hip fractures among older people. J Am Geriatr Soc. 1997;45: 1108-1112. 75 [113] [114] [115] [116] [117] [118] [119] [120] [121] [122] [123] [124] [125] [126] 76 Formiga F, Lopez-Soto A, Duaso E, et al. Characteristics of falls producing hip fractures in nonagenarians. J Nutr Health Aging. 2008;12: 664-667. Bueno-Cavanillas A, Padilla-Ruiz F, Jimenez-Moleon JJ, PeinadoAlonso CA, Galvez-Vargas R. Risk factors in falls among the elderly according to extrinsic and intrinsic precipitating causes. European journal of epidemiology. 2000;16: 849-859. Yardley L, Donovan-Hall M, Francis K, Todd C. Attitudes and beliefs that predict older people's intention to undertake strength and balance training. J Gerontol B Psychol Sci Soc Sci. 2007;62: P119125. Hughes K, van Beurden E, Eakin EG, et al. Older persons' perception of risk of falling: implications for fall-prevention campaigns. Am J Public Health. 2008;98: 351-357. Zunft HJ, Friebe D, Seppelt B, et al. Perceived benefits and barriers to physical activity in a nationally representative sample in the European Union. Public Health Nutrition. 1999;2: 153-160. Gellert P, Witham MD, Crombie IK, Donnan PT, McMurdo ME, Sniehotta FF. The role of perceived barriers and objectively measured physical activity in adults aged 65-100. Age Ageing. 2015. Yardley L, Donovan-Hall M, Francis K, Todd C. Older people's views of advice about falls prevention: a qualitative study. Health Educ Res. 2006;21: 508-517. American Geriatrics Society Expert Panel on the Care of Older Adults with M. Patient-centered care for older adults with multiple chronic conditions: a stepwise approach from the American Geriatrics Society: American Geriatrics Society Expert Panel on the Care of Older Adults with Multimorbidity. J Am Geriatr Soc. 2012;60: 1957-1968. Simpson JM, Darwin C, Marsh N. What are older people prepared to do to avoid falling? A qualitative study in London. British journal of community nursing. 2003;8: 152, 154-159. Yardley L, Bishop FL, Beyer N, et al. Older people's views of fallsprevention interventions in six European countries. Gerontologist. 2006;46: 650-660. Apter AJ, Paasche-Orlow MK, Remillard JT, et al. Numeracy and communication with patients: they are counting on us. J Gen Intern Med. 2008;23: 2117-2124. Wolfer J. Aspects of “reality” and ways of knowing in nursing: In search of an integrating paradigm. Journal of Nursing Scholarship. 1993;25: 141-146. Sandelowski M. Combining qualitative and quantitative sampling, data collection, and analysis techniques in mixed-method studies. Research in nursing & health. 2000;23: 246-255. Guba E, Lincoln Y. Competing paradigms in qualitative research. Handbook of qualitatvie research. California: Sage, 1994, pp. 105117. [127] [128] [129] [130] [131] [132] [133] [134] [135] [136] [137] [138] [139] [140] [141] [142] [143] [144] [145] [146] Giddings LS, Grant BM. A Trojan horse for positivism?: a critique of mixed methods research. ANS Advances in nursing science. 2007;30: 52-60. Strauss A, Corbin J. Basics of qualitative research. California, 1990. Creswell JW. Qualitative inquiry and Research Design: Choosing among five approaches. 3rd ed. London: Sage Publications, 2013. Hewitt J. Ethical components of researcher researched relationships in qualitative interviewing. Qualitative health research. 2007;17: 1149-1159. Fossey E, Harvey C, McDermott F, Davidson L. Understanding and evaluating qualitative research. The Australian and New Zealand journal of psychiatry. 2002;36: 717-732. Sandelowski M, Barroso J. Classifying the findings in qualitative studies. Qualitative health research. 2003;13: 905-923. Rothman KJ. Epidemiology: An introduction 2nd Edition. New York: Oxfor University Press, 2012. Andrew S, Halcomb EJ. Mixed Methods for Nursing and the Health Sciences. Oxford, U.K.: Wiley-Blackwell, 2009. Rossman G, Wilsson B. Numbers and Words: Combining Quantitative and Qualitative Methods in a Single Large-Scale Evaluation Study Evaluation Review. 1985;9: 627-643. Denzin NK. The Research Act: A Theoretical Introduction to Sociological Methods. New York: Aldine, 1970. Satariano WA. Epidemiology of aging: An ecological approach. Massachusetts: Jones and Bartlett, 2006. World Health Organisation International classification of functioning, disability and health. Geneva: WHO, 2001. Pless M, Granlund M. Handbok i att använda ICF och ICF-CY. Lund: Studentlitteratur AB, 2011. Lundin-Olsson L, Jensen J. Fall och fallskador i: Systematiskt arbete för äldres säkerhet. Om fall, trafikolyckor och bränder. Stockholm och Karlstad: Räddningsverket, 2007. Polit D, Beck C. Nursing Research, Principles and Methods (7th edition): Lippincott Williams & Wilkins, 2004. Ritchie J, Lewis J. Qualitative Research Practice. A guide for social science students and researchers. London: Sage publications, 2003. Quan H, Sundararajan V, Halfon P, et al. Coding algorithms for defining comorbidities in ICD-9-CM and ICD-10 administrative data. Medical care. 2005;43: 1130-1139. Cox NJ. Speaking Stata: In praise of trigonometric predictors. The stata journal. 2006;6: 561-579. Rothman K. Episheet- Spreadsheets for the analysis of Epidemiologic Data. Rothman, K, 2008. Greenspan SL, Myers ER, Maitland LA, Resnick NM, Hayes WC. Fall severity and bone mineral density as risk factors for hip fracture in ambulatory elderly. JAMA. 1994;271: 128-133. 77 [147] [148] [149] [150] [151] [152] [153] [154] [155] [156] [157] [158] [159] [160] [161] [162] [163] 78 Sennerby U, Melhus H, Gedeborg R, et al. Cardiovascular diseases and risk of hip fracture. JAMA. 2009;302: 1666-1673. Michaelsson K, Melhus H, Ferm H, Ahlbom A, Pedersen NL. Genetic liability to fractures in the elderly. Archives of internal medicine. 2005;165: 1825-1830. Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. Journal of chronic diseases. 1987;40: 373-383. Hsieh HF, Shannon SE. Three approaches to qualitative content analysis. Qualitative health research. 2005;15: 1277-1288. Morgan DL. Qualitative content analysis: a guide to paths not taken. Qualitative health research. 1993;3: 112-121. Krippendorff K. Content analysis: An introduction to its methodology. Beverly Hills, CA: Sage, 1980. Graneheim UH, Lundman B. Qualitative content analysis in nursing research: concepts, procedures and measures to achieve trustworthiness. Nurse education today. 2004;24: 105-112. Kondracki NL, Wellman NS, Amundson DR. Content analysis: review of methods and their applications in nutrition education. Journal of nutrition education and behavior. 2002;34: 224-230. Saltin B, Grimby G. Physiological analysis of middle-aged and old former athletes. Comparison with still active athletes of the same ages. Circulation. 1968;38: 1104-1115. Spruit MA, Sillen MJ, Groenen MT, Wouters EF, Franssen FM. New normative values for handgrip strength: results from the UK Biobank. Journal of the American Medical Directors Association. 2013;14: 775 e775-711. Creswell JW. Research design: Qualitative, Quantitative and Mixed Method approaches. 4th edition ed. London.: Sage, 2014. Tashakkori A, Teddlie C. Mixed methodology: Combining qualitative and quantitative approaches. Thousand Oaks, CA: Sage, 1998. Creswell JW, Plano Clark VL. Designing and conducting mixed methods research. Thousand Oaks, CA.: Sage, 2007. Gedeborg R, Engquist H, Berglund L, Michaelsson K. Identification of incident injuries in hospital discharge registers. Epidemiology. 2008;19: 860-867. Hebert-Davies J, Laflamme GY, Rouleau D, Health, investigators F. Bias towards dementia: are hip fracture trials excluding too many patients? A systematic review. Injury. 2012;43: 1978-1984. Ranhoff AH, Holvik K, Martinsen MI, Domaas K, Solheim LF. Older hip fracture patients: three groups with different needs. BMC geriatrics. 2010;10: 65. Dasgupta M, Dumbrell AC. Preoperative risk assessment for delirium after noncardiac surgery: a systematic review. J Am Geriatr Soc. 2006;54: 1578-1589. [164] [165] [166] [167] [168] [169] [170] [171] [172] [173] [174] [175] [176] [177] Robinovitch SN, Feldman F, Yang Y, et al. Video capture of the circumstances of falls in elderly people residing in long-term care: an observational study. Lancet. 2013;381: 47-54. Vlaeyen E, Deschodt M, Debard G, et al. Fall incidents unraveled: a series of 26 video-based real-life fall events in three frail older persons. BMC geriatrics. 2013;13: 103. Kvale S, Brinkman S. Interviews: Learning the craft of qualitative interviewing. 2nd ed. Los Angelas: Sage publications, 2009. Yardley L, Smith H. A prospective study of the relationship between feared consequences of falling and avoidance of activity in community-living older people. Gerontologist. 2002;42: 17-23. Yardley L, Beyer N, Hauer K, Kempen G, Piot-Ziegler C, Todd C. Development and initial validation of the Falls Efficacy ScaleInternational (FES-I). Age Ageing. 2005;34: 614-619. Washburn RA, Smith KW, Jette AM, Janney CA. The Physical Activity Scale for the Elderly (PASE): development and evaluation. Journal of clinical epidemiology. 1993;46: 153-162. Sato Y, Kanoko T, Satoh K, Iwamoto J. Risk factors for hip fracture among elderly patients with Alzheimer's disease. Journal of the neurological sciences. 2004;223: 107-112. Whooley MA, Kip KE, Cauley JA, Ensrud KE, Nevitt MC, Browner WS. Depression, falls, and risk of fracture in older women. Study of Osteoporotic Fractures Research Group. Archives of internal medicine. 1999;159: 484-490. Paudel ML, Taylor BC, Diem SJ, et al. Association between depressive symptoms and sleep disturbances in community-dwelling older men. J Am Geriatr Soc. 2008;56: 1228-1235. Thompson Coon J, Boddy K, Stein K, Whear R, Barton J, Depledge MH. Does participating in physical activity in outdoor natural environments have a greater effect on physical and mental wellbeing than physical activity indoors? A systematic review. Environmental science & technology. 2011;45: 1761-1772. Pasanen TP, Tyrvainen L, Korpela KM. The relationship between perceived health and physical activity indoors, outdoors in built environments, and outdoors in nature. Applied psychology Health and well-being. 2014;6: 324-346. Kerr J, Marshall S, Godbole S, et al. The relationship between outdoor activity and health in older adults using GPS. International journal of environmental research and public health. 2012;9: 46154625. Kerr J, Sallis JF, Saelens BE, et al. Outdoor physical activity and self rated health in older adults living in two regions of the U.S. The international journal of behavioral nutrition and physical activity. 2012;9: 89. Nyman SR, Ballinger C, Phillips JE, Newton R. Characteristics of outdoor falls among older people: a qualitative study. BMC geriatrics. 2013;13: 125. 79 [178] [179] [180] [181] [182] [183] [184] [185] [186] [187] [188] [189] [190] [191] 80 Li W, Keegan TH, Sternfeld B, Sidney S, Quesenberry CP, Jr., Kelsey JL. Outdoor falls among middle-aged and older adults: a neglected public health problem. Am J Public Health. 2006;96: 1192-1200. Manty M, Heinonen A, Viljanen A, et al. Outdoor and indoor falls as predictors of mobility limitation in older women. Age Ageing. 2009;38: 757-761. Bergland A, Jarnlo GB, Laake K. Predictors of falls in the elderly by location. Aging clinical and experimental research. 2003;15: 43-50. Kelsey JL, Berry SD, Procter-Gray E, et al. Indoor and outdoor falls in older adults are different: the maintenance of balance, independent living, intellect, and Zest in the Elderly of Boston Study. J Am Geriatr Soc. 2010;58: 2135-2141. Bath PA, Morgan K. Differential risk factor profiles for indoor and outdoor falls in older people living at home in Nottingham, UK. European journal of epidemiology. 1999;15: 65-73. Beauchet O, Annweiler C, Allali G, Berrut G, Herrmann FR, Dubost V. Recurrent falls and dual task-related decrease in walking speed: is there a relationship? J Am Geriatr Soc. 2008;56: 1265-1269. Howe TE, Rochester L, Neil F, Skelton DA, Ballinger C. Exercise for improving balance in older people. Cochrane Database Syst Rev. 2011: CD004963. Okubo Y, Osuka Y, Jung S, et al. Walking can be more effective than balance training in fall prevention among community-dwelling older adults. Geriatrics & gerontology international. 2015. Okubo Y, Seino S, Yabushita N, et al. Longitudinal association between habitual walking and fall occurrences among communitydwelling older adults: analyzing the different risks of falling. Archives of gerontology and geriatrics. 2015;60: 45-51. Faber MJ, Bosscher RJ, Chin APMJ, van Wieringen PC. Effects of exercise programs on falls and mobility in frail and pre-frail older adults: A multicenter randomized controlled trial. Arch Phys Med Rehabil. 2006;87: 885-896. Turner CH. Bone strength: current concepts. Annals of the New York Academy of Sciences. 2006;1068: 429-446. Martyn-St James M, Carroll S. Meta-analysis of walking for preservation of bone mineral density in postmenopausal women. Bone. 2008;43: 521-531. Sherrington C, Tiedemann A, Fairhall N, Close JC, Lord SR. Exercise to prevent falls in older adults: an updated meta-analysis and best practice recommendations. New South Wales public health bulletin. 2011;22: 78-83. Martyn-St James M, Carroll S. A meta-analysis of impact exercise on postmenopausal bone loss: the case for mixed loading exercise programmes. Br J Sports Med. 2009;43: 898-908. [192] [193] [194] [195] [196] [197] [198] [199] [200] [201] [202] [203] [204] Johnell K, Fastbom J. Comparison of prescription drug use between community-dwelling and institutionalized elderly in Sweden. Drugs Aging. 2012;29: 751-758. Formiga F, Lopez-Soto A, Duaso E, et al. Differences in the characteristics of elderly patients suffering from hip fracture due to falls according to place of residence. Journal of the American Medical Directors Association. 2007;8: 533-537. Bennett A, Gnjidic D, Gillett M, et al. Prevalence and impact of fallrisk-increasing drugs, polypharmacy, and drug-drug interactions in robust versus frail hospitalised falls patients: a prospective cohort study. Drugs Aging. 2014;31: 225-232. Nordberg G, Wimo A, Jonsson L, et al. Time use and costs of institutionalised elderly persons with or without dementia: results from the Nordanstig cohort in the Kungsholmen Project--a population based study in Sweden. International journal of geriatric psychiatry. 2007;22: 639-648. Eriksson S, Strandberg S, Gustafson Y, Lundin-Olsson L. Circumstances surrounding falls in patients with dementia in a psychogeriatric ward. Archives of gerontology and geriatrics. 2009;49: 80-87. van der Velde N, Stricker BH, Pols HA, van der Cammen TJ. Risk of falls after withdrawal of fall-risk-increasing drugs: a prospective cohort study. British journal of clinical pharmacology. 2007;63: 232-237. Nurminen J, Puustinen J, Lahteenmaki R, et al. Handgrip strength and balance in older adults following withdrawal from long-term use of temazepam, zopiclone or zolpidem as hypnotics. BMC geriatrics. 2014;14: 121. Ensrud KE, Blackwell TL, Ancoli-Israel S, et al. Use of selective serotonin reuptake inhibitors and sleep disturbances in communitydwelling older women. J Am Geriatr Soc. 2006;54: 1508-1515. Draganich LF, Zacny J, Klafta J, Karrison T. The effects of antidepressants on obstructed and unobstructed gait in healthy elderly people. J Gerontol A Biol Sci Med Sci. 2001;56: M36-41. Kvelde T, McVeigh C, Toson B, et al. Depressive symptomatology as a risk factor for falls in older people: systematic review and metaanalysis. J Am Geriatr Soc. 2013;61: 694-706. Aberg AC, Frykberg GE, Halvorsen K. Medio-lateral stability of sitto-walk performance in older individuals with and without fear of falling. Gait & posture. 2010;31: 438-443. Benvenuto LJ, Krakoff LR. Morbidity and mortality of orthostatic hypotension: implications for management of cardiovascular disease. American journal of hypertension. 2011;24: 135-144. Poon IO, Braun U. High prevalence of orthostatic hypotension and its correlation with potentially causative medications among elderly veterans. Journal of clinical pharmacy and therapeutics. 2005;30: 173-178. 81 [205] [206] [207] [208] [209] [210] [211] [212] [213] [214] [215] [216] [217] [218] [219] 82 Feldstein C, Weder AB. Orthostatic hypotension: a common, serious and underrecognized problem in hospitalized patients. Journal of the American Society of Hypertension : JASH. 2012;6: 27-39. Wieling W, van Dijk N, Thijs RD, de Lange FJ, Krediet CT, Halliwill JR. Physical countermeasures to increase orthostatic tolerance. Journal of internal medicine. 2015;277: 69-82. Horton K. Gender and the risk of falling: a sociological approach. J Adv Nurs. 2007;57: 69-76. Belanger M, Townsend N, Foster C. Age-related differences in physical activity profiles of English adults. Prev Med. 2011;52: 247249. Duckham RL, Procter-Gray E, Hannan MT, Leveille SG, Lipsitz LA, Li W. Sex differences in circumstances and consequences of outdoor and indoor falls in older adults in the MOBILIZE Boston cohort study. BMC geriatrics. 2013;13: 133. Butler AA, Lord SR, Taylor JL, Fitzpatrick RC. Ability Versus Hazard: Risk-Taking and Falls in Older People. J Gerontol A Biol Sci Med Sci. 2014. Jorgensen TS, Hansen AH, Sahlberg M, et al. Falls and comorbidity: the pathway to fractures. Scandinavian journal of public health. 2014;42: 287-294. Reyes C, Estrada P, Nogues X, et al. The impact of common comorbidities (as measured using the Charlson index) on hip fracture risk in elderly men: a population-based cohort study. Osteoporos Int. 2014;25: 1751-1758. Lawlor DA, Patel R, Ebrahim S. Association between falls in elderly women and chronic diseases and drug use: cross sectional study. BMJ. 2003;327: 712-717. Byberg L, Gedeborg R, Cars T, et al. Prediction of fracture risk in men: a cohort study. J Bone Miner Res. 2012;27: 797-807. Quach L, Galica AM, Jones RN, et al. The nonlinear relationship between gait speed and falls: the Maintenance of Balance, Independent Living, Intellect, and Zest in the Elderly of Boston Study. J Am Geriatr Soc. 2011;59: 1069-1073. Lee PG, Cigolle C, Blaum C. The co-occurrence of chronic diseases and geriatric syndromes: the health and retirement study. J Am Geriatr Soc. 2009;57: 511-516. Marengoni A, Rizzuto D, Wang HX, Winblad B, Fratiglioni L. Patterns of chronic multimorbidity in the elderly population. J Am Geriatr Soc. 2009;57: 225-230. Fried LP, Tangen CM, Walston J, et al. Frailty in older adults: evidence for a phenotype. J Gerontol A Biol Sci Med Sci. 2001;56: M146-156. Lewis CE, Ewing SK, Taylor BC, et al. Predictors of non-spine fracture in elderly men: the MrOS study. J Bone Miner Res. 2007;22: 211-219. [220] [221] [222] [223] [224] [225] [226] [227] [228] [229] [230] [231] [232] Drake MT, Murad MH, Mauck KF, et al. Clinical review. Risk factors for low bone mass-related fractures in men: a systematic review and meta-analysis. The Journal of clinical endocrinology and metabolism. 2012;97: 1861-1870. Saveman BI, Bjornstig U. Unintentional injuries among older adults in northern Sweden--a one-year population-based study. Scandinavian journal of caring sciences. 2011;25: 185-193. safety Kcfs. Injuries among older people in Sweden. Karlstad: The Swedish civil contingencies agency 2009. Rosendahl E, Lundin-Olsson L, Kallin K, Jensen J, Gustafson Y, Nyberg L. Prediction of falls among older people in residential care facilities by the Downton index. Aging clinical and experimental research. 2003;15: 142-147. Oliver D, Britton M, Seed P, Martin FC, Hopper AH. Development and evaluation of evidence based risk assessment tool (STRATIFY) to predict which elderly inpatients will fall: case-control and cohort studies. BMJ. 1997;315: 1049-1053. Nandy S, Parsons S, Cryer C, et al. Development and preliminary examination of the predictive validity of the Falls Risk Assessment Tool (FRAT) for use in primary care. J Public Health (Oxf). 2004;26: 138-143. Falls and Fall Injuries - measures for prevention (National satsning för ökad patientsäkerhet). Stockholm: The Swedish Associations of Local Authorities, 2010. Gill T, Taylor AW, Pengelly A. A population-based survey of factors relating to the prevalence of falls in older people. Gerontology. 2005;51: 340-345. Stel VS, Smit JH, Pluijm SM, Lips P. Consequences of falling in older men and women and risk factors for health service use and functional decline. Age Ageing. 2004;33: 58-65. Roe B, Howell F, Riniotis K, Beech R, Crome P, Ong BN. Older people and falls: health status, quality of life, lifestyle, care networks, prevention and views on service use following a recent fall. Journal of clinical nursing. 2009;18: 2261-2272. Todd C, Skelton DA. What are the main risk factors for falls among older people and what are the most effective interventions to prevent these falls? Copenhagen: WHO Regional Office for Europe (Health Evidence Network report), 2004. Dollard J, Braunack-Mayer A, Horton K, Vanlint S. Why older women do or do not seek help from the GP after a fall: a qualitative study. Family practice. 2014;31: 222-228. Sarkisian CA, Hays RD, Mangione CM. Do older adults expect to age successfully? The association between expectations regarding aging and beliefs regarding healthcare seeking among older adults. J Am Geriatr Soc. 2002;50: 1837-1843. 83 [233] [234] [235] [236] [237] [238] [239] [240] [241] [242] [243] [244] [245] [246] [247] 84 Howland J, Lachman ME, Peterson EW, Cote J, Kasten L, Jette A. Covariates of fear of falling and associated activity curtailment. Gerontologist. 1998;38: 549-555. Lee F, Mackenzie L, James C. Perceptions of older people living in the community about their fear of falling. Disability and rehabilitation. 2008;30: 1803-1811. Roe B, Howell F, Riniotis K, Beech R, Crome P, Ong BN. Older people's experience of falls: understanding, interpretation and autonomy. J Adv Nurs. 2008;63: 586-596. Horne M, Speed S, Skelton D, Todd C. What do communitydwelling Caucasian and South Asian 60-70 year olds think about exercise for fall prevention? Age Ageing. 2009;38: 68-73. Kanis J, Oden A, Johnell O. Acute and long-term increase in fracture risk after hospitalization for stroke. Stroke; a journal of cerebral circulation. 2001;32: 702-706. Pouwels S, Lalmohamed A, Leufkens B, et al. Risk of hip/femur fracture after stroke: a population-based case-control study. Stroke; a journal of cerebral circulation. 2009;40: 3281-3285. Carda S, Cisari C, Invernizzi M, Bevilacqua M. Osteoporosis after stroke: a review of the causes and potential treatments. Cerebrovascular diseases. 2009;28: 191-200. Dargent-Molina P, Douchin MN, Cormier C, Meunier PJ, Breart G, Group ES. Use of clinical risk factors in elderly women with low bone mineral density to identify women at higher risk of hip fracture: The EPIDOS prospective study. Osteoporos Int. 2002;13: 593-599. Brown JS, Vittinghoff E, Wyman JF, et al. Urinary incontinence: does it increase risk for falls and fractures? Study of Osteoporotic Fractures Research Group. J Am Geriatr Soc. 2000;48: 721-725. Tinetti ME, Inouye SK, Gill TM, Doucette JT. Shared risk factors for falls, incontinence, and functional dependence. Unifying the approach to geriatric syndromes. JAMA. 1995;273: 1348-1353. Kenny RA, Bhangu J, King-Kallimanis BL. Epidemiology of syncope/collapse in younger and older Western patient populations. Progress in cardiovascular diseases. 2013;55: 357-363. Borkan JM, Quirk M, Sullivan M. Finding meaning after the fall: injury narratives from elderly hip fracture patients. Social science & medicine. 1991;33: 947-957. Sarkisian CA, Prohaska TR, Wong MD, Hirsch S, Mangione CM. The relationship between expectations for aging and physical activity among older adults. J Gen Intern Med. 2005;20: 911-915. Cameron ID, Gillespie LD, Robertson MC, et al. Interventions for preventing falls in older people in care facilities and hospitals. Cochrane Database Syst Rev. 2012;12: CD005465. Svarstad BL, Mount JK. Chronic benzodiazepine use in nursing homes: effects of federal guidelines, resident mix, and nurse staffing. J Am Geriatr Soc. 2001;49: 1673-1678. [248] [249] Seo HS, Lee JH, Park YH. Effects of a Task-specific Exercise Program on Balance, Mobility, and Muscle Strength in the Elderly. Journal of physical therapy science. 2014;26: 1693-1695. Littbrand H, Lundin-Olsson L, Gustafson Y, Rosendahl E. The effect of a high-intensity functional exercise program on activities of daily living: a randomized controlled trial in residential care facilities. J Am Geriatr Soc. 2009;57: 1741-1749. 85 Acta Universitatis Upsaliensis Digital Comprehensive Summaries of Uppsala Dissertations from the Faculty of Medicine 1082 Editor: The Dean of the Faculty of Medicine A doctoral dissertation from the Faculty of Medicine, Uppsala University, is usually a summary of a number of papers. A few copies of the complete dissertation are kept at major Swedish research libraries, while the summary alone is distributed internationally through the series Digital Comprehensive Summaries of Uppsala Dissertations from the Faculty of Medicine. (Prior to January, 2005, the series was published under the title “Comprehensive Summaries of Uppsala Dissertations from the Faculty of Medicine”.) Distribution: publications.uu.se urn:nbn:se:uu:diva-247286 ACTA UNIVERSITATIS UPSALIENSIS UPPSALA 2015