Abstract
Background.
Despite wide-spread use of antihypertensives in older adults, the literature is unclear about their association with incident recurrent falls over time.
Methods.
Health, Aging and Body Composition study participants (n = 2,948) who were well functioning at baseline (1997) were followed to Year 7 (2004). The main outcome was recurrent falls (≥2) in the ensuing 12 months. Antihypertensive use was examined as: (a) any versus none, (b) long- versus short-term (≥2 vs <2 years), and by (c) summated standardized daily dose (SDD; 1 = maximum recommended daily dose for one antihypertensive), and (d) subclass.
Results.
Controlling for potential demographic, health status/behavior and access to care confounders, we found no increase in risk of recurrent falls in antihypertensive users compared to nonusers (adjusted odds ratio [AOR] = 1.13; 95% CI = 0.88–1.46), or those taking higher SDDs or for longer durations. Only those using a loop diuretic were found to have a modest increased risk of recurrent falls (AOR = 1.50; 95% CI = 1.11–2.03).
Conclusions.
Antihypertensive use overall was not statistically significantly associated with recurrent falls after adjusting for important confounders. Loop diuretic use may be associated with recurrent falls and needs further study.
Key Words: Falls, Medication, Epidemiology, Drug related
Hypertension (especially systolic) is the most common chronic condition in older adults (1). If un- or undertreated, hypertension can increase risk of stroke, heart failure, and premature death (1). Numerous randomized controlled trials demonstrate the effectiveness of thiazide diuretics, central alpha blockers, renin-angiotensin-aldosterone system inhibitors, and calcium channel blockers in reducing blood pressure in adults well into their eighties (1). Even though these drugs are generally well tolerated, as few as 50% of elders with hypertension use antihypertensives (often more than one), and fewer than 50% of these have adequate blood pressure control (2,3).
Suboptimal blood pressure control may derive, in part, from concerns about adverse drug effects, including falls. Indeed, 18%–40% of community-dwelling elderly fall yearly (4–6), of whom nearly 50% have recurrent falls (6,7). Recurrent falls (as opposed to single falls) may be more clinically important as they may increase the risk of physician visits, functional status decline, nursing home admission, and death (6). To date only three studies have looked at the risk of antihypertensives and recurrent falls in community dwelling elders, and none of them demonstrated an increased risk after controlling for other important factors (8–10). None of these studies examined individual antihypertensive classes, dosage, or duration.
Thus, the objective of the current longitudinal study was to assess the association of overall and specific classes of antihypertensive use, dose, and duration with recurrent falls in community-dwelling, initially well-functioning elders.
Methods
Study Design, Data Source, and Sample
We used data from the Health, Aging, and Body Composition (Health ABC) study, a population-based, prospective, longitudinal observational study of community-dwelling older adults (11). This study was approved by the University of California at San Francisco (UCSF), University of Pittsburgh, and University of Tennessee Memphis Institutional Review Boards, and informed consent was obtained from each participant prior to data collection. The baseline sample (1997/1998) included 3,075 black and white men and women aged 70–79 years who reported no difficulty walking ¼ mile, or climbing 10 steps and lived in specified zip codes surrounding Pittsburgh, Pennsylvania and Memphis, Tennessee (11). Thus, the sample for the current analysis included 2,948 well functioning older adults at Year 1 with complete medication use and fall data the following year followed through Year 6 for antihypertensive medication exposure and Year 7 for recurrent falls.
Data Collection and Management
Participants were seen annually during a clinic or home visit and detailed physiologic (including blood pressure) and self-report questionnaire measurements (including demographics, health behavior/status (including medications), and access to health care factors were collected (11). Detailed medication data were collected in clinic or at home about products taken in the previous month using a state of the art “brown bag” review method (12,13). A similar data collection approach was used for telephone interviews if participants were unable to be seen in person. Studies have shown that medication use information collected by either “brown bag” or telephone methods are highly accurate and concordant with information about dispensed prescription drugs in claims data (14,15). For all medications, the interviewer recorded the name, strength, dosage form, and the number of dosage forms the respondents said they had used the previous day, week, or month. The medication data collected for the Health ABC Study were edited and coded using the Iowa Drug Information System (IDIS) Drug Vocabulary and Thesaurus (16). IDIS is a hierarchical coding system with eight character unique codes for specific drug ingredients, chemical and therapeutic categories. The therapeutic category code allows drugs to be assigned to 1 of 20 major therapeutic classes and 200 subclasses based on an expanded version of the American Hospital Formulary Services format (17).
Teleform was used to create scannable forms for direct data entry. Missing and questionable values were highlighted by the software for visual review and online editing. Additional range checks and data cleaning were conducted at the UCSF Coordinating Center. De-identified SAS permanent data files were created for analysis.
Primary Outcome
The number of falls in which the participant landed on the floor or ground in the previous 12 months was assessed at the year following medication use data collection (eg, Year 1 antihypertensive use and previous year falls reported at Year 2) for each wave (five waves). The primary outcome was recurrent falls (≥2) in the ensuing 12 months following report of medication use. This method of fall recall (in the previous 12 months) has been shown to be highly specific (91%–95%) in comparison with that reported using more frequent assessments (18).
Primary Independent Variable
Medications used to treat hypertension were grouped into therapeutic subclasses: (a) beta blockers; (b) alpha blockers (peripheral and central); (c) loop diuretics; (d) thiazide diuretics; (e) potassium-sparing diuretics; (f) calcium channel blockers; (g) angiotensin converting enzyme inhibitors; and (h) angiotensin receptor blockers (19). Too few participants were taking vasodilators so this class was not assessed. Any use was defined as the use of any medication from the eight subclasses. To evaluate the possibility of a dose–response relationship, the daily dose was calculated for current users for each individual antihypertensive medication by multiplying the number of dosage forms taken the previous day by the strength of the medication reported at the interview. The daily dose was then converted to a standardized daily dose (SDD) by dividing it by the maximum effective dose per day as noted in a standard reference (19). Thus, a person taking 1.0 standardized antihypertensive drug unit would have taken the maximum recommended effective daily dose for one agent. The standardized dose was summated for all antihypertensives, regardless of class, taken daily. Finally, to examine the impact of duration, long-term use was operationally defined as ≥2 years and short-term use as less than 2 years (19).
Control Variables
To address potential confounding, we controlled for a number of demographic, health status/behavior, and access to care factors (6,7). Demographic factors included age, sex, race, site, education (less than a high school education, high school graduate, and postsecondary education), and marital status (never married, married, and previously married).
Health behaviors included smoking status and alcohol use. Health status factors included pulmonary disease, arthritis, urinary problems, cerebrovascular disease, bodily pain in the previous month (13,19), vision (excellent/good sight, fair sight, and poor to completely blind), and body mass index (underweight/normal, <24.9; overweight, 25.0–29.9; and obese, ≥30). Self-rated global health was dichotomized as excellent/very good/good versus fair/poor. In addition, time-varying covariates were created for depressive symptoms (Short Center for Epidemiologic Studies-Depression Scale >10), cognitive impairment (modified mini-mental state test [3MS] <80) and the total number of prescription medications (excluding dugs that increase falls, and antihypertensive drugs) per participant (7,20,21). Time varying exposure to any drug that increases the risk of falls (non-anticholinergic central nervous system medications, including benzodiazepines, antidepressants, antipsychotics, and opioids) was also assessed (7). Access to care factors included dichotomous variables for hospitalization in previous 12 months, private physician, prescription insurance, and flu shot in previous 12 months (3).
Self-reported hypertension was subdivided into controlled (<140/90 mmHg) and uncontrolled (≥140/90 mmHg) on a time-varying basis. Antihypertensives can be used for a number of comorbid conditions (1,22). Therefore, to mitigate potential confounding by indication, we included self-reported peripheral artery disease, benign prostatic hyperplasia symptoms, coronary heart disease, congestive heart failure, and diabetes.
Statistical Analysis
We used appropriate descriptive statistics for summarization and generalized estimating equations for eliciting the main findings (23–25). First, we assessed the unadjusted association between antihypertensive use and recurrent falls over time. Second, a priori covariates that may affect recurrent falls were included: site, heart failure, benign prostatic hypertrophy, cognitive impairment, depressive symptoms, self-reported hypertension (controlled/uncontrolled), and other drugs that increase the risk of falls. Each antihypertensive subclass was run as a separate model and controlled for antihypertensive subclass use other than the subclass being evaluated, with no use being the reference group. Finally, additional covariates were selected using a forward stepwise selection approach applied separately for each of three domains of covariates (demographic, health status/behavior, and access to health care). Specifically, stepwise detected covariates and those deemed important a priori were included in the final model. We also conducted sensitivity analyses by restricting the sample to only those with hypertension (whether controlled or uncontrolled) at baseline, and stratifying the analysis by any falls history and median gait speed at baseline. All analyses were conducted using SAS software (version 9.3; SAS Institute, Cary, NC).
Results
At baseline these well functioning participants had a mean (standard deviation) age of 73.6 (2.9) years, 51.6% were female, and 40.8% were black (Table 1). In addition, slightly more than half had self-reported hypertension, and nearly half of those with hypertension were uncontrolled.
Table 1.
Characteristics of the Sample at Baseline (n = 2,948)
| Variables | Antihypertensive User (n = 1,677), Mean ± SD or N, % | Not an Antihypertensive User (n = 1,271), Mean ± SD or N, % |
|---|---|---|
| Demographics | ||
| Female gender | 888 (53.0) | 634 (50.0) |
| Black race | 766 (45.7) | 437 (34.4) |
| Site (Pittsburgh) | 829 (49.4) | 637 (50.1) |
| Age | 73.6 (2.9) | 73.6 (2.8) |
| Education | ||
| Postsecondary | 694 (41.4) | 566 (44.5) |
| High school graduate | 557 (33.2) | 397 (31.2) |
| <High school | 423 (25.2) | 304 (24.0) |
| Married | 857 (51.1) | 674 (53.0) |
| Health behaviors | ||
| Current smoker | 153 (9.1) | 149 (11.7) |
| Alcohol use (≥1 drink per week) | 441 (26.3) | 406 (31.9) |
| Health status | ||
| Pulmonary disease | 183 (10.9) | 115 (9.0) |
| Arthritis | 1,006 (60.0) | 644 (50.7) |
| Urinary problems | 305 (18.2) | 190 (14.9) |
| Cerebrovascular disease | 173 (10.3) | 58 (4.6) |
| Bodily pain (any in past 30 d) | 1,152 (68.7) | 790 (62.2) |
| Vision problems | ||
| Excellent/good sight | 1,314 (78.4) | 1,032 (81.2) |
| Fair sight | 314 (18.7) | 205 (16.1) |
| Poor to completely blind | 47 (2.8) | 34 (2.7) |
| Body mass index | ||
| Underweight/normal (<24.9) | 458 (27.3) | 489 (38.5) |
| Overweight (25.0–29.9) | 711 (42.4) | 546 (43.0) |
| Obese (30+) | 508 (30.3) | 236 (18.6) |
| Excellent/very good/good self-rated health | 1,328 (79.2) | 1,151 (90.6) |
| Depressive symptoms (Short CES-D ≥ 10)* | 108 (6.4) | 60 (4.7) |
| Cognitive impairment (3MS < 80)* | 162 (9.7) | 122 (9.6) |
| Drugs that increase risk of falls* (benzodiazepines antidepressants, antipsychotics, opioids) | 246 (14.7) | 121 (9.5) |
| No. prescription medications*,† | 2.1 (2.1) | 1.3 (1.8) |
| Conditions where antihypertensives are indicated | ||
| Self-reported hypertension* | ||
| Controlled < 140/90 mmHg | 695 (41.4) | 80 (6.3) |
| Uncontrolled ≥140/90 mmHg | 625 (37.3) | 98 (7.7) |
| Peripheral arterial disease | 112 (6.7) | 37 (2.9) |
| Benign prostatic hyperplasia | 413 (24.6) | 279 (22.0) |
| Coronary heart disease | 491 (29.3) | 140 (11.0) |
| Congestive heart failure | 77 (4.6) | 8 (0.6) |
| Diabetes | 322 (19.2) | 118 (9.3) |
| Access to care | ||
| Hospitalization in previous 12 mo | 307 (18.3) | 128 (10.1) |
| Private physician | 1,352 (80.6) | 967 (76.1) |
| Prescription insurance | 1,109 (66.1) | 752 (59.2) |
| Flu shot in previous 12 mo | 1,225 (73.0) | 814 (64.0) |
Notes: 3MS = modified mini-mental status exam; CES-D = Center for Epidemiologic Studies Depression Scale; CNS = central nervous system; SD = standard deviation.
*Time varying.
†Excluding drugs that increase falls and antihypertensives.
Table 2 shows the prevalence of antihypertensive use over time. At baseline, over half reported the use of one or more antihypertensives, with calcium channel blockers being the most common class used (23%). At baseline, hydrochlorothiazide was the most common thiazide diuretic reported (94%). Also at baseline, 13.2% took a potassium sparing diuretic. The majority (93%) of potassium sparing diuretic use was with amiloride and triamterene, which are both epithelial sodium channel inhibitors (ENaC). ENaC use was almost always in combination with a thiazide diuretic (97%).
Table 2.
Antihypertensive Use Over Time*
| Antihypertensive Medication Use | Year 1 (n = 2,948), N (%) | Year 2 (n = 2,811), N (%) | Year 3 (n = 2,679), N (%) | Year 5 (n = 2,489), N (%) | Year 6 (n = 2,388), N (%) |
|---|---|---|---|---|---|
| Any use | 1,677 (56.9) | 1,636 (58.2) | 1,622 (60.5) | 1,676 (67.3) | 1,673 (70.1) |
| Long term (≥2 y) for any use | 1,143/1,677 (68.2) | 1,084/1,636 (66.3) | 1,073 (66.2) | 1,153/1,676 (68.8) | 1,059/1,673 (63.3) |
| SDD ≥ 2 | 283 (9.6) | 289 (10.3) | 253 (9.4) | 300 (12.4) | 305 (12.8) |
| SDD = 1–2 | 546 (18.5) | 555 (19.7) | 584 (21.8) | 668 (26.8) | 662 (27.7) |
| SDD < 1 | 820 (28.7) | 774 (28.1) | 770 (29.3) | 703 (28.4) | 695 (29.6) |
| Specific class use | — | — | — | — | — |
| Beta blockers | 395 (13.4) | 409 (14.5) | 434 (16.2) | 520 (20.9) | 572 (24.0) |
| Alpha blockers | 364 (12.4) | 317 (11.3) | 284 (10.6) | 264 (10.6) | 238 (10.0) |
| Loop diuretics | 178 (6.0) | 195 (6.9) | 207 (7.7) | 263 (10.6) | 265 (11.1) |
| Thiazide diuretics | 582 (19.7) | 593 (21.1) | 557 (20.8) | 599 (24.1) | 579 (24.3) |
| Calcium channel blockers | 678 (23.0) | 662 (23.6) | 642 (24.0) | 631 (25.4) | 611 (25.6) |
| ACE inhibitors | 443 (15.0) | 458 (16.3) | 479 (17.9) | 586 (23.5) | 606 (25.4) |
| ARB | 69 (2.3) | 95 (3.4) | 142 (5.3) | 205 (8.2) | 249 (10.4) |
| Vasodilators | 35 (1.2) | 20 (0.7) | 18 (0.7) | 9 (0.4) | 10 (0.4) |
Notes: ACE = angiotensin converting enzyme; ARB = angiotensin II receptor blocker; SDD = standardized daily dose of antihypertensives.
*No medication data were collected at Year 4 in the Health ABC Study.
At baseline, the overwhelming majority reported taking an antihypertensive for more than 2 years and in doses ≤2 SDDs. By Year 6, 70% took one or more antihypertensives, with calcium channel blockers persisting as the most common class used (25.6%).
At baseline, 21.2% of participants reported any fall in the previous year. At year 2, 8% of participants reported having recurrent falls in the previous year. This rate remained somewhat stable over the next four waves (7.5%–10.4%; Table 3).
Table 3.
Prevalence of Recurrent and Single Falls Over Time
| Variables | Year 2 (n = 2,948) n, % | Year 3 (n = 2,811) n, % | Year 4 (n = 2,679) n, % | Year 6 (n = 2,489) n, % | Year 7 (n = 2,388) n, % |
|---|---|---|---|---|---|
| 2+ falls (recurrent) | 240 (8.1) | 210 (7.5) | 229 (8.6) | 259 (10.4) | 206 (8.6) |
| 1 fall | 456 (15.5) | 391 (13.9) | 430 (16.1) | 396 (15.9) | 369 (15.5) |
Table 4 shows the results from the multivariable model, controlling for demographic, health behavior/status, and access to care factors. While any antihypertensive use was found to increase the risk of recurrent falls in bivariate anlyses (crude odds ratio [OR] 1.89, 95% confidence interval [CI] 1.61–2.21), after controlling for a number of variables, we found that the adjusted odds ratio was reduced and no long showed a statistically significant increase in risk of recurrent falls (adjusted OR 1.13, 95% CI 0.88–1.46). Covariates with AOR > 1.0 and p < .05 were use of non-antihypertensive drugs that may increase risk of falls, arthritis, pulmonary disease, poor vision to completely blind, alcohol use, cerebrovascular disease, and diabetes. Similar findings were seen in those taking higher SDDs or for longer durations.
Table 4.
Association Between Antihypertensive Use and Recurrent Falls Controlling for Covariates
| Antihypertensive Medication Use* | Unadjusted OR (95% CI) | Adj. OR (95% CI)† |
|---|---|---|
| Any use | 1.89 (1.61–2.21) | 1.13 (0.88–1.46) |
| Long duration (≥2 y) | 1.89 (1.60–2.24) | 1.15 (0.88–1.49) |
| Short duration | 1.89 (1.57–2.28) | 1.11 (0.83–1.49) |
| SDD ≥2 | 2.12 (1.69–2.69) | 1.21 (0.82–1.78) |
| SDD 1–2 | 1.93 (1.61–2.31) | 1.05 (0.78–1.42) |
| SDD <1 | 1.79 (1.48–2.16) | 1.16 (0.88–1.52) |
| Specific class use‡ | ||
| Beta blockers | 1.22 (1.01–1.46) | 1.06 (0.82–1.36) |
| Alpha blockers | 1.29 (1.03–1.63) | 1.25 (0.94–1.67) |
| Loop diuretics | 1.60 (1.27–2.01) | 1.50 (1.11–2.03) |
| Thiazide diuretics | 0.97 (0.82–1.15) | 0.87 (0.68–1.10) |
| Calcium channel blockers | 1.23 (1.04–1.45) | 1.06 (0.84–1.33) |
| ACE inhibitors | 1.22 (1.01–1.47) | 1.10 (0.85–1.42) |
| ARBs | 1.35 (1.03–1.76) | 1.17 (0.80–1.71) |
Notes: ACE = angiotensin converting enzyme; adj = adjusted; ARB = angiotensin II receptor blocker; CI = confidence interval; OR = odds ratio.
*Reference group is no use.
†Controlling for variables forced into the model (ie, site, heart failure, BPH, cognitive impairment, depressive symptoms, self-reported hypertension [controlled/uncontrolled], drugs that increase risk of falls) and those from forward selection procedures (ie, education, age, marital status, alcohol use, cerebrovascular disease, diabetes, pulmonary disease, arthritis, urinary problems, vision problems, total number of prescription medications, syncope).
‡Each antihypertensive subclass was run as a separate model and controlled for antihypertensive subclass use other than the subclass being evaluated.
Of the eight subclasses, only loop diuretics were found associated with recurrent falls (adjusted OR 1.50, 95% CI 1.11–2.03). Post hoc analyses revealed no significant dose-response relationship among loop diuretic users (SDD ≥ 1, adjusted OR 1.53, 95% CI 0.88–2.65; SDD < 1, adjusted OR 1.53, 95% CI 1.08–2.15). However, those taking a loop diuretic for ≥2 years had a significant increase in risk, but not for a shorter period of time (adjusted OR 1.64, 95% CI 1.10–2.45 and adjusted OR 1.38, 95% CI 0.94–2.03, respectively). Post hoc analyses were also conducted in which three variables were created to capture the use of thiazide and potassium diuretics alone or in combination. They revealed that thiazide diuretic use only (adjusted OR 1.05), potassium-sparing diuretic use only (adjusted OR 0.92), or both diuretic types (adjusted OR 1.12) did not have a statistically significant (all p > .05) increased risk of recurrent falls.
Sensitivity analyses showed similar results when the sample was restricted to those with hypertension and stratified by blood pressure control (Supplementary Table 1). Similar results were also seen when the sample was stratified by those with a previous falls history at baseline (Supplementary Table 2). Multivariable analyses stratified by the median gait speed at baseline (1.17 m/s) found that any antihypertensive use did not increase the risk of recurrent falls in those with slower or faster gait speeds (AOR [95%CI] 0.61 [0.35–1.06] and AOR [95%CI] 0.99 [0.56–1.74]), respectively.
Discussion
This study assessed the association between antihypertensive use and recurrent falls in well functioning (at baseline) community-dwelling older adults. We found a nonsignificant 13% increase in risk of recurrent falls with any antihypertensive use. This is consistent with the non-significant findings from three other studies of antihypertensive drugs and recurrent falls in community dwelling elders (8–10).
This study also explored both dose– and duration–response relationships with any antihypertensive use and recurrent falls and found no increased risk. It is important to note our findings are robust in that our sensitivity analyses designed to address potential confounding by prior fall history and indication by restricting the sample to only those with hypertension revealed similar results.
In examining whether recurrent falls risk was elevated with use of any of seven antihypertensive subclasses, we found a statistically significant 54% increase in risk of recurrent falls associated with loop diuretic use, but not with the remaining six. A small, case–control study similarly found over a twofold increase in risk of falling in loop diuretic users versus the comparison group (31.1% vs 13.3%; p < .05) (26). A nursing home study using a novel case-crossover design reported that nine persons with a loop diuretic change had a nearly 2.5-fold increased risk of a fall (27). Few other studies have found this relationship, however. A prospective cohort study of women found no association between loop diuretic use and recurrent falls (28). One explanation for the different findings may relate to duration of use, which we found to be an important factor. Further study is needed on the risk of falls with long-term loop diuretic use. One plausible explanation for this potential association is that loop diuretics can cause dehydration and/or orthostatic hypotension, potentially causing presyncope and leading to a fall.
So what are some of the clinical implications of our findings? First, clinicians should feel comfortable prescribing chronically antihypertensives for the treatment of hypertension in older adults as recommended by recent guidelines and without fear that the risk of recurrent falls is increased (1). This is important since the use of these drugs results in a reduction of adverse cardiovascular events in patients ≥80 years of age. That said, rarely the initiation of some antihypertensives in high doses may cause orthostatic hypotension and syncope that could result in a serious fall (29–31). Because of this, the 2012 American Geriatric Society Beers Criteria suggest that the use of both peripheral and central alpha-blockers for hypertension be avoided in older adults (32). Clinicians should also consider not prescribing loop diuretics for hypertension because there may be an increased risk of recurrent falls as documented by the current study. Moreover, some evidence suggests that thiazide diuretics are more effective in reducing elevated blood pressure than loop diuretics (33).
There are important potential limitations to our study. First, potential misclassification of the main outcome of recurrent falls has to be considered as it was retrospectively collected via self-report. However, it is a highly specific method in comparison to self-reporting of falls via diary (18). Moreover, there are limitations to the timing of the collection of falls data across the years that may have prevented some participants from being classified as a recurrent faller. In this respect, our recurrent fall rates are probably slight under estimates of the true rates. Second, medication data were collected at fixed annual assessments, preventing us from documenting the exact date in which antihypertensive medications were initiated, changed, or discontinued. Third, while it is possible that there is residual confounding, we were mindful of potential confounding bias and adjusted for common indications for antihypertensive therapy. Moreover, our sensitivity analyses showed similar results when restricted to only those with hypertension and when stratified by those with a baseline falls history. Fourth, we did not have information about the time of onset of hypertension. Fifth, because the majority of ENaC users were also taking a thiazide diuretic, it is difficult to say with confidence what the independent effect of their use was and thus it was not examined separately from the other seven antihypertensive classes. Sixth, we were unable to assess total duration of antihypertensive treatment. Finally, our analysis focused on older, well-functioning community-dwelling black and white men and women from the Memphis and Pittsburgh areas, and the generalizability to older adults in other regions or different care settings is unknown.
In conclusion, antihypertensive use overall was not statistically significantly associated with recurrent falls after adjusting for important confounders. Analyzed separately, however, loop diuretics were significantly associated with recurrent falls. Future research is needed with larger sample sizes in both community and long-term care facility settings with older adults to improve the evidence base on about medication-related fall risk.
Supplementary Material
Supplementary material can be found at: http://biomedgerontology.oxfordjournals.org/
Funding
This work was supported by National Institute on Aging (NIA) grants (R01AG027017, P30AG024827, T32AG021885, and K07AG033174). This work was also supported in part by the Intramural Research program of the National Institutes of Health (N01AG62101, N01AG62103, N01AG62106, and R01AG028050) and a National Institute of Nursing Research grant (R01NR012459).
Supplementary Material
Acknowledgments
We would like to thank Yihuang Kang, PhD and Robert M. Boudreau, PhD for their advice regarding Health ABC data and data analysis.
References
- 1. Aronow WS, Fleg JL, Pepine CJ, et al. ACCF/AHA 2011 expert consensus document on hypertension in the elderly: a report of the American College of Cardiology Foundation Task Force on Clinical Expert Consensus documents developed in collaboration with the American Academy of Neurology, American Geriatrics Society, American Society for Preventive Cardiology, American Society of Hypertension, American Society of Nephrology, Association of Black Cardiologists, and European Society of Hypertension. J Am Coll Cardiol. 2011;57:2037–2114. 10.1016/j.jacc.2011.01.008 [DOI] [PubMed] [Google Scholar]
- 2. Ostchega Y, Dillon CF, Hughes JP, Carroll M, Yoon S. Trends in hypertension prevalence, awareness, treatment, and control in older U.S. adults: data from the National Health and Nutrition Examination Survey 1988 to 2004. J Am Geriatr Soc. 2007;55:1056–1065. [DOI] [PubMed] [Google Scholar]
- 3. Rooks RN, Simonsick EM, Klesges LM, Newman AB, Ayonayon HN, Harris TB. Racial disparities in health care access and cardiovascular disease indicators in Black and White older adults in the Health ABC Study. J Aging Health. 2008;20:599–614. 10.1177/0898264308321023 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Hanlon JT, Landerman LR, Fillenbaum GG, Studenski S. Falls in African American and white community-dwelling elderly residents. J Gerontol A Biol Sci Med Sci. 2002;57:M473–M478. [DOI] [PubMed] [Google Scholar]
- 5. de Rekeneire N, Visser M, Peila R, et al. Is a fall just a fall: correlates of falling in healthy older persons. The Health, Aging and Body Composition Study. J Am Geriatr Soc. 2003;51:841–846. [DOI] [PubMed] [Google Scholar]
- 6. King MB. Falls. In: Halter J, Ouslander J, Tinetti M, Studenski S, High K, Asthana S, eds. Hazzard’s Geriatric Medicine and Gerontology. 6th ed. New York: McGraw-Hill; 2009:659–670. [Google Scholar]
- 7. Hanlon JT, Boudreau RM, Roumani YF, et al. Number and dosage of central nervous system medications on recurrent falls in community elders: the Health, Aging and Body Composition study. J Gerontol A Biol Sci Med Sci. 2009;64:492–498. 10.1093/gerona/gln043 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Tromp AM, Smit JH, Deeg DJ, Bouter LM, Lips P. Predictors for falls and fractures in the Longitudinal Aging Study Amsterdam. J Bone Miner Res. 1998;13:1932–1939. [DOI] [PubMed] [Google Scholar]
- 9. Chu LW, Chi I, Chiu AY. Incidence and predictors of falls in the Chinese elderly. Ann Acad Med Singapore. 2005;34:60–72. [PubMed] [Google Scholar]
- 10. Gerdhem P, Ringsberg KA, Akesson K, Obrant KJ. Clinical history and biologic age predicted falls better than objective functional tests. J Clin Epidemiol. 2005;58:226–232. [DOI] [PubMed] [Google Scholar]
- 11. Newman AB, Haggerty CL, Kritchevsky SB, Nevitt MC, Simonsick EM; Health ABC Collaborative Research Group. Walking performance and cardiovascular response: associations with age and morbidity—the Health, Aging and Body Composition Study. J Gerontol A Biol Sci Med Sci. 2003;58:715–720. [DOI] [PubMed] [Google Scholar]
- 12. Psaty BM, Lee M, Savage PJ, Rutan GH, German PS, Lyles M. Assessing the use of medications in the elderly: methods and initial experience in the Cardiovascular Health Study. The Cardiovascular Health Study Collaborative Research Group. J Clin Epidemiol. 1992;45:683–692. [DOI] [PubMed] [Google Scholar]
- 13. Marcum ZA, Perera S, Donohue JM, et al. ; Health, Aging and Body Composition Study. Analgesic use for knee and hip osteoarthritis in community-dwelling elders. Pain Med. 2011;12:1628–1636. 10.1111/j.1526-4637.2011.01249.x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Pit SW, Byles JE, Cockburn J. Accuracy of telephone self-report of drug use in older people and agreement with pharmaceutical claims data. Drugs Aging. 2008;25:71–80. [DOI] [PubMed] [Google Scholar]
- 15. Caskie GI, Willis SL, Warner Schaie K, Zanjani FA. Congruence of medication information from a brown bag data collection and pharmacy records: findings from the Seattle longitudinal study. Exp Aging Res. 2006;32:79–103. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Pahor M, Chrischilles EA, Guralnik JM, Brown SL, Wallace RB, Carbonin P. Drug data coding and analysis in epidemiologic studies. Eur J Epidemiol. 1994;10:405–411. [DOI] [PubMed] [Google Scholar]
- 17. McEvoy GK, ed. AHFS Drug Information 2004. Bethesda, MD: American Society of Hospital Pharmacists; 2011. [Google Scholar]
- 18. Ganz DA, Higashi T, Rubenstein LZ. Monitoring falls in cohort studies of community-dwelling older people: effect of the recall interval. J Am Geriatr Soc. 2005;53:2190–2194. [DOI] [PubMed] [Google Scholar]
- 19. National Institutes of Health; National Heart, Lung, and Blood Institute. The Sixth Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure http://www.nhlbi.nih.gov/guidelines/archives/jnc6/jnc6_archive.pdf Accessed July 7, 2014. [PubMed]
- 20. Radloff LS. The CES-D scale: a self-report depression scale for research use in the general population. Appl Psych Measur. 1977;1:385–401. [Google Scholar]
- 21. Teng EL, Chui HC. The Modified Mini-Mental State (3MS) examination. J Clin Psychiatry. 1987;48:314–318. [PubMed] [Google Scholar]
- 22. Marcum ZA, Fried LF. Aging and antihypertensive medication-related complications in the chronic kidney disease patient. Curr Opin Nephrol Hypertens. 2011;20:449–456. 10.1097/MNH.0b013e32834902ad [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Liang K, Zeger S. Longitudinal data analysis using generalized linear models. Biometrika. 1986;73:13–22. [Google Scholar]
- 24. Diggle PJ. An approach to the analysis of repeated measurements. Biometrics. 1988;44:959–971. [PubMed] [Google Scholar]
- 25. Diggle PJ, Heagerty P, Liang K, Zeger S, eds. Analysis of Longitudinal Data. 2nd ed. New York: Oxford University Press; 2002. [Google Scholar]
- 26. Sobel KG, McCart GM. Drug use and accidental falls in an intermediate care facility. Drug Intell Clin Pharm. 1983;17:539–542. [DOI] [PubMed] [Google Scholar]
- 27. Berry SD, Mittleman MA, Zhang Y, et al. New loop diuretic prescriptions may be an acute risk factor for falls in the nursing home. Pharmacoepidemiol Drug Saf. 2012;21:560–563. 10.1002/pds.3256 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Lim LS, Fink HA, Blackwell T, Taylor BC, Ensrud KE. Loop diuretic use and rates of hip bone loss and risk of falls and fractures in older women. J Am Geriatr Soc. 2009;57:855–862. 10.1111/j.1532-5415.2009.02195.x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Hanlon JT, Linzer M, MacMillan JP, Lewis IK, Felder AA. Syncope and presyncope associated with probable adverse drug reactions. Arch Intern Med. 1990;150:2309–2312. [PubMed] [Google Scholar]
- 30. Butt DA, Mamdani M, Austin PC, Tu K, Gomes T, Glazier RH. The risk of falls on initiation of antihypertensive drugs in the elderly. Osteoporos Int. 2013;24:2649–2657. 10.1007/s00198-013- 2369-7 [DOI] [PubMed] [Google Scholar]
- 31. Tinetti ME, Han L, Lee DS, et al. Antihypertensive medications and serious fall injuries in a nationally representative sample of older adults. JAMA Intern Med. 2014;174:588–595. 10.1001/jamainternmed.2013.14764 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. American Geriatrics Society 2012 Beers Criteria Update Expert Panel. American Geriatrics Society updated Beers Criteria for potentially inappropriate medication use in older adults. J Am Geriatr Soc. 2012;60:616–631. 10.1111/j.1532-5415.2012.03923.x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Araoye MA, Chang MY, Khatri IM, Freis ED. Furosemide compared with hydrochlorothiazide. Long-term treatment of hypertension. JAMA. 1978;240:1863–1866. [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.