Alcohol and Select Medications as Fall Risk Factors in Community Dwelling Older Adults in Canada

Abstract

Abstract

Introduction: Falls are the leading cause of accidental injury in community dwelling older adults, often resulting in emergency room visits, hospitalization and early admission to long-term care. Studies have identified many risk factors for falls including increasing age, diagnosis of a chronic disease, poor vision, fear of falling, hazards in the home, alcohol and prescription medications. While studies have examined many risk factors associated with falls, there is limited information on whether prescription medications and alcohol are risk factors in community dwelling older adults. Purposes: The primary thesis objectives were to: i) conduct a literature review to synthesize the literature on whether alcohol and psychotropic, anti-depressant and anti-hypertensive medications are risk factors for falls in community dwelling older adults; ii) examine what health and social factors are associated with high-risk alcohol use in community-dwelling older adults in Canada,; and iii) examine whether alcohol and select medications are predictive of falls in community dwelling older adults. Methods: The first aspect of this thesis was to conduct a literature review on select medications (psychotropics, benzodiazepines or anti-psychotics, anti-depressants, and anti-hypertensives) and alcohol, both singly and in combination, on fall risk in community dwelling older adults using the following search terms: drug or medication, aged or elderly or older adult or senior, accidental falls or falls or falling, and alcohol or alcohol drinking. These search terms were entered into four databases (PubMed, EMBASE, CINAHL and SCOPUS). A total of 1,146 articles were retrieved and screened for inclusion. Studies were included if 1) was a primary study; 2) included community dwelling persons aged 60 years and older; 3) included alcohol use as an independent variable; 4) included medications of interest; 5) falls were the primary outcome variable; and 6) published in English. We excluded studies if they were: 1) review articles, conference proceedings, books, editorial, case studies or commentary; 2) if articles relied on qualitative data (interviews, focus groups); 3) if they were not in English; and 4) included institutionalized persons (e.g. living in LTC or hospital). Chapters 3 and 4 used data from the Canadian Injury Prevention Survey. The survey was distributed online to local, provincial and national organizations across Canada that cater to older adults (2016-2017) and collected information on demographics, perception of physical and mental health, fall history, alcohol use, use of psychotropic, anti-depressant and anti-hypertensive medication use, smoking status, diagnosis of a chronic disease, physical activity, usual sleep quantity, likelihood of daytime sleepiness, and executive duties. While data was collected on community dwelling individuals 45 years and older, only data of those 65 years and older was examined in the present thesis (n=2,281). Chapter 3 examined alcohol use in community dwelling older adults (n=2,279). Participants reported the number of drinks they consumed per week and the number of days per week they consumed alcohol. The purpose of this chapter was to: 1) examine alcohol use in community dwelling older adults in Canada using current alcohol consumption guidelines; 2) develop and test new alcohol consumption guidelines to determine if they better differentiate drinking habits on health outcomes; and 3) determine risk factors of high-risk drinking. Low risk drinking was defined as 1-2 drinks per week for both males and females. Moderate drinking was defined as 3-9 drinks per week for males and 3-6 drinks for females. High risk drinking limits were defined as 10 drinks per week for males and 7 drinks per week for females. Using logistic regression, we examined demographic and health factors to predict low and high-risk drinking. Multinomial regression was used to examine predictors of low, moderate and high-risk drinking. Chapter 4 examined the use of alcohol and select medications as risk factors for falls, both as independent predictors of falls, and together with alcohol (n=2,281). Participants reported prescription medication, alcohol use on a weekly basis and fall history. Medications examined included psychotropic (anxiolytics, anti-psychotics, hypnotics/sedatives), anti-depressants (selective serotonin reuptake inhibitors (SSRI), serotonin antagonist reuptake inhibitors (SARI), serotonin-norepinephrine reuptake inhibitors (SNRI), tricyclic anti-depressants, or other) and cardiovascular agents (diuretics, beta-blockers, angiotensin converting enzyme inhibitors, angiotensin II receptor antagonist, calcium channel blockers) on fall risk. Correlations were performed to determine health factors correlated with falls. Logistic regression determined medications predictive of falls. Results: The literature review (Chapter 2) found 29 observational studies that met the inclusion criteria after screening. Sample sizes ranged from 307 to 321,422. Twenty-six studies examined the effects of benzodiazepines, hypnotics/sedatives or anti-psychotic use on falls, and twenty found benzodiazepines, hypnotics/sedatives or anti-psychotics were predictive of falls. Thirteen studies examined anti-depressant use and falls; 10 studies found that anti-depressants are predictive of falls. Twelve studies examined anti-hypertensives as a risk factor for falls and four found anti-hypertensives were predictive of falls. Fifteen studies examined alcohol use and five found alcohol was a predictor of falls. Findings from Chapter 3 show that 70% of participants reported drinking at least once per week. Using the current Canadian alcohol consumption guidelines, 6.2% of older males and 10.1% of older females were classified as high-risk drinkers. Moderate drinking was reported by 32.8% of males and 23.3% of females. When comparing current alcohol consumption guidelines to the new additional category (moderate drinking), there were no significant difference on health outcomes. Findings from Chapter 4 show that 33.4% of older adults had a fall in the previous two years. In fallers, 5.3% reported psychotropic use, 10.4% reported anti-depressant use, and 51.6% reported anti-hypertensive use. Anti-hypertensive medication and alcohol use was not associated with falls. As a group, psychotropic drug use was predictive of falls after controlling for potential confounding factors (p<0.05). Anti-depressant sub-classes SSRIs, SARIs and SNRIs were predictive of falls on their own and when controlling for sex (p<0.05). Conclusions The literature review found benzodiazepines, hypnotics/sedatives, anti-psychotic and anti-depressants are predictive of falls in community dwelling older adults. Anti-hypertensive medication and alcohol use were not predictive of falls. Compared to females, males were more likely to be moderate drinkers, however, overall adding a moderate drinking category did not help differentiate health and social factors associated with alcohol consumption. Psychotropic drug use overall, as well as individual anti-depressant drug classes (i.e. SSRI, SNRI and SARI) were all predictive of falls after controlling for sex. Anti-hypertensives and alcohol were not predictors of falls in community dwelling older adults based on the survey data. Findings of the literature review and survey data confirm psychotropic and anti-depressant use increase fall risk in community dwelling older adults. Clinicians should exercise caution when prescribing psychotropic and anti-depressant medications to older adults.