Risk factors for falls among older adults in India: A systematic review and meta‐analysis

Abstract Background and Aim Falls are common among older adults in India. Several primary studies on its risk factors have been conducted in India. However, no systematic review has been conducted on this topic. Thus, the objective of this systematic review was to synthesize the existing evidence on the risk factors for falls among older adults in India. Methods JBI and Preferred Reporting Items for Systematic Reviews and Meta‐Analyse guidelines were followed, and two independent reviewers were involved in the process. This review included observational studies conducted among older adults (aged ≥ 60 years) residing in India, reporting any risk factor for falls as exposure and unintentional fall as the outcome. MEDLINE, EMBASE, PsycInfo, CINAHL, and ProQuest Dissertations and Theses were searched until September 24, 2020. Where possible, data were synthesized using random‐effects meta‐analysis. Results The literature search yielded 3445 records. Twenty‐two studies met the inclusion criteria of this systematic review, and 19 studies were included in the meta‐analysis. Out of the 22 included studies in the systematic review, 12 (out of 18) cross‐sectional studies, two case–control studies, and two cohort studies met more than 70% criteria in the respective Joanna Briggs Institute (JBI) checklists. Risk factors for falls among older adults in India included sociodemographic factors, environmental factors, lifestyle factors, physical and/or mental health conditions, and medical interventions. Conclusions This systematic review and meta‐analysis provided a holistic picture of the problem in India by considering a range of risk factors such as sociodemographic, environmental, lifestyle, physical and/or mental health conditions and medical intervention. These findings could be used to develop falls prevention interventions for older adults in India. Systematic Review and Meta‐Analysis Registration The systematic review and meta‐analysis protocol was registered with PROSPERO (registration number‐CRD42020204818).


| INTRODUCTION
Falls are events that lead to a person coming to rest inadvertently at a lower level. 1 Falls commonly occur in adults aged 60 years or more. 1,2 India is the second most populated country, and the number of older adults is estimated to be 137 million in 2021. 3 The number of falls among older adults is increasing with the transition in demographics over time. 4,5 The pooled prevalence of falls among older adults in India is estimated to be 31% (95% confidence interval [CI]: 23%-39%). 6 Falls can have a negative long-term impact on the physical and psychological health and socioeconomic condition of the individual. [7][8][9][10][11][12][13][14][15][16][17][18] Impact on health includes morbidity and even mortality in severe cases. [7][8][9][10][11][12][13][14][15][16] Physical health consequences include injuries and fractures and reduced activities of daily living. 2,7,8,11 In India, every year, nearly 1.5-2 million older people suffer injuries due to falls, and 1 million succumb to death due to falls. 16 Psychological health consequences include depression, anxiety, the fear of falling, and the lack of self-confidence. 9,[11][12][13]17 Social consequences include the lack of social interaction leading to isolation. 9 Economic consequences include increased health and social care costs. 18 All these can take a toll on the overall quality of life. 9,11 Disability-adjusted life years (DALYs) lost due to falls are also high. 15 Several primary studies have been conducted in India to determine the risk factors for falls among older adults. 5,[19][20][21][22][23][24][25] However, no systematic review has been conducted on this topic.
Thus, the objective of this systematic review was to synthesize the existing evidence on the risk factors for falls among older adults in India. The intention was to provide a holistic picture of the problem in India by considering a range of risk factors such as sociodemographic, environmental, lifestyle, physical and/or mental health conditions, and medical intervention. These findings could be used to develop falls prevention interventions for older adults in India.

| METHODS
The systematic review process adhered to the Joanna Briggs Institute (JBI) systematic reviews of etiology and risk guidelines 26

and Preferred
Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA). 27 The review protocol was registered with PROSPERO (registration number: CRD42020204818). Two reviewers were involved in the process and independently screened the titles and abstracts and full texts of studies, assessed the methodological quality of studies, and extracted data from the studies (I. B. and B. A.). Any disagreements that arose between them were resolved through discussion. If consensus was not reached, a third reviewer was involved (K. C.).

| Population
The systematic review included studies conducted among older adults (aged ≥ 60 years) residing in India. A study was also eligible if the mean age of the participants was ≥60 years. Furthermore, if the study findings were stratified by age, required data were extracted from the relevant age group, that is, adults aged ≥60 years. If it was not possible to extract these findings, the study was excluded.
Studies conducted in any setting, such as community, residential care, primary care, secondary care, and tertiary care, were eligible.

| Exposure
Studies reporting any risk factors for falls as exposure were included.

| Outcome
Studies reporting unintentional falls as outcomes were included (i.e., the actual occurrence of falls and not the risk or fear of falls).
Studies reporting falls due to accidents or intentional actions like self-harm or domestic violence were excluded.

| Study design
Observational studies (cohort, case-control, and cross-sectional studies) were included. Dissertations and Theses. An initial limited search was carried out on MEDLINE and EMBASE databases using the keywords: "risk factors," "falls," and "India." The titles and abstracts of the studies were screened for keywords, and the index terms used to describe the article were also identified. The search results were inspected to ensure that relevant articles were identified. Based on this, the search strategy for each database was developed in consultation with a senior research librarian and are detailed in the Supporting Information File: Appendix 1. All the databases were searched on September 24, 2020.

| DATABASES AND SEARCH STRATEGY
No date or language restrictions were applied. The reference list of all the identified reviews and studies selected for inclusion in the systematic review were screened for additional studies.

| STUDY SELECTION
Retrieved studies were collated and uploaded onto EndNote X9 (Clarivate Analytics), a reference management software. 28 After the removal of duplicate studies, the titles, and abstracts of the remaining studies were screened for eligibility using the inclusion criteria.
Studies identified as potentially eligible or those without an abstract had their full texts retrieved. Full texts of the studies were assessed for eligibility. Those that did not meet the inclusion criteria were excluded, and the reasons for exclusion are reported in the Supporting Information File: Appendix 2.

| METHODOLOGICAL QUALITY ASSESSMENT
The included studies were critically assessed using the JBI checklists for observational studies. 26,27,29 As recommended by JBI, a cut-off score was not used to include/exclude studies. Hence, all studies irrespective of their methodological quality were included.

| DATA EXTRACTION
Data were extracted from the included studies using a predeveloped and pretested data extraction, and we used Microsoft Word for this purpose. The following information was extracted: author and year of the study, name of the Indian state, study design, study period, study setting (e.g., community, primary care, secondary care, tertiary care), sample size, population characteristics (mean age [in years], number of females), risk factors explored, the definition of falls and assessment of falls (e.g., self-reported/using medical notes or reports). Where possible, odds ratios (ORs) were extracted along with 95% CIs. Adjusted ORs were preferred over crude ORs. If only raw data were presented, ORs and 95% CIs were calculated. In case of missing or insufficient data in the paper, the corresponding author was emailed twice and requested to share the same.

| METHODOLOGICAL QUALITY OF INCLUDED STUDIES
The total critical appraisal scores for each study are presented in Table 1.
Tables 2-4 report the detailed critical appraisal of the included studies.
Two cohort studies attained more than 70% JBI criteria, that is, answered affirmatively to at least eight questions on the BISWAS ET AL. | 3 of 16 checklist. 42,43 The two groups for comparison were similar in characteristics and recruited from the same population in both the studies. 42,43 Measurement of exposures was done in a valid and reliable way and clearly described in both the studies. 42,43 Both the studies identified confounding factors and used multiple logistic regression analysis to deal with confounding. 42,43 The patients were free of the outcome (i.e., no falls) before inclusion in the studies and used standard definitions of falls. 42,43 The follow-up time was at least 1 year which was sufficient to assess falls. 42,43 In one study, there was no information on the follow-up of patients, and the strategies to address incomplete follow-up were also not described. 43 Appropriate statistical analysis was used as both the studies utilized regression analysis. 42,43 Both the case-control studies attained more than 70% JBI criteria, that is, answered affirmatively to at least seven questions on the checklist. 5,33 Cases and controls were not matched appropriately in one study. 33 For each of the studies, the same criteria were used for the identification of cases and controls. 5,33 It was unclear if the validity of exposure measurement was done in a standard, valid and reliable way. 5,33 However, measurement of exposure was done using the same method for cases and controls. 5,33 Both the studies identified confounders and used multivariable logistic regression analysis to deal with the potential confounding variables. Standard definitions of falls were used to assess falls in a standard, valid and reliable way for both cases and controls. 5,33 The exposure period of interest was at least 6 months in both the studies, which was enough to assess falls. Appropriate statistical analyses were used as multivariable regression analyses were conducted in both the studies. 5,33 Twelve out of 18 cross-sectional studies included in the systematic review attained more than 70% JBI criteria, that is, answered affirmatively to at least six questions on the checklist. 19,24,25,31,32,[34][35][36][37][39][40][41] All the studies reported inclusion criteria except one 22 and study settings and patients except one. 38 The PRISMA flow diagram of the identification, screening, and eligibility of the included articles.      measurement of exposure was unclear in three studies 20,21,39 and was not described in three studies. 22,24,40 All the studies defined falls succinctly except three. 21,34,40 Five studies did not identify the confounders and strategies to deal with the same. 20,22,23,35,38 However, studies that mentioned confounders reported age and sex as the most common confounders. In the four studies with insufficient statistical analyses, multivariable logistic regression could have been conducted. 21 Total % of "yes" to each critical appraisal question 100 (2) 100 (2) 100 (2) 100 (0) 100 (2) 100 (2) 100 (2) 100 (2) 50(1) 50 (1) 100 (2) Abbreviations: N, no; U, unclear; Y, yes. 1. Were the two groups similar and recruited from the same population?
2. Were the exposures measured similarly to assign people to both exposed and unexposed groups?
3. Was the exposure measured in a valid and reliable way? 4. Were confounding factors identified? 5. Were strategies to deal with confounding factors stated? 6. Were the groups/participants free of the outcome at the start of the study (or at the moment of exposure)? 7. Were the outcomes measured in a valid and reliable way?
8. Was the follow-up time reported and sufficient to be long enough for outcomes to occur? 9. Was follow-up complete, and if not, were the reasons to loss to follow-up described and explored? 10. Were strategies to address incomplete follow-up utilized? 11. Was appropriate statistical analysis used?

T A B L E 3 Critical appraisal results of case-control studies
Total % of "yes" to critical appraisal questions
2. Were cases and controls matched appropriately?
3. Were the same criteria used for the identification of cases and controls?
4. Was exposure measured in a standard, valid, and reliable way? 5. Was exposure measured in the same way for cases and controls? 6. Were confounding factors identified? 7. Were strategies to deal with confounding factors stated? 8. Were outcomes assessed in a standard, valid and reliable way for cases and controls? 9. Was the exposure period of interest long enough to be meaningful? 10   Globally, age is a well-known risk factor for falls. [45][46][47][48] In this review, age was found to be a significant risk factor in the metaanalysis conducted for cross-sectional studies, however, not for cohort studies. In terms of the hierarchy of study designs, cohort studies are considered better than cross-sectional studies. However, in this case, there were only two cohort studies, and the statistical heterogeneity was high (84%). On the other hand, there were nine cross-sectional studies, and the statistical heterogeneity F I G U R E 5 Summary forest plot of the association between physical and/or mental health conditions and falls.
F I G U R E 6 Summary forest plot of the association between medical interventions and falls.
was 43%. It should also be noted that we included only those studies that focused on older adults, and the age range was already narrow. In this review, physical activity was found to be a risk factor. Intuitively, one would expect the opposite, and this issue requires further investigation. The possible reason could be not following the recommended physical activity guidelines, quantity or quality wise. 84 In total, three studies could not be included in the metaanalysis. In two studies, it was not possible to estimate the ORs due to insufficient raw data, however, other relevant information was extracted. 20,21 Another study mentioned unique risk factors which were not reported in any other study. 22  and falls among older adults, 85 and research shows that South Asians are more likely to have diabetes. 86 Hence, the findings could be used by a range of stakeholders (including policymakers) in the South Asian region to develop falls prevention targeted interventions, depending on the exact risk factor. If there is more than one risk factor, a multifactorial intervention is recommended to prevent falls. 87,88 It should be noted that the "one-size-fits-all" concept should not be applied, and "need-sensitive" interventions are required. One such example could be yoga-based interventions. 89

| CONCLUSION
This systematic review and meta-analysis reported a wide range of risk factors for falls among older adults in India such as sociodemographic, environmental, lifestyle, physical and/or mental health condition, and medical intervention. These findings could be used to develop fall prevention interventions for older adults in India.