Deprescribing Electronic Case Reviews for Older Veterans at Risk for Falls: Effects on Drug Burden and Falls

Abstract

Background:

Falls are the most common medication-related safety event in older adults. Deprescribing fall risk-increasing drugs (FRIDs) may mitigate fall risk. This study assesses the effects of an innovative deprescribing program in reducing FRID burden and falls-related acute visits over 1 year.

Methods:

The Falls Assessment of Medications in the Elderly (FAME) Program is a pilot deprescribing program designed to improve medication safety in Veterans aged ≥ 65, screening positive for high fall risk at the Durham Veterans Affairs Health Care System. Central case finding and electronic case reviews with deprescribing recommendations were completed by an interdisciplinary team, forwarded to prescribers for approval, then implemented during follow-up telephone visits by FAME team. Primary outcome was change in FRID burden calculated by modified Drug Burden Index (DBI) at 1 year and an exploratory outcome was 1-year fall-related acute visits.

Results:

Overall, 472 patients (236 intervention cases, 236 matched controls) were included in the study. Of the 236 patients receiving a FAME deprescribing plan, 147 had recommendations approved by prescriber and patient. In the intention-to-treat analysis, the 1-year change in modified DBI was −0.15 (95% CI −0.23, −0.08) in the intervention cohort and −0.11 (−0.21, −0.00) in the matched control cohort (P= 0.47). The odds of increasing DBI by a clinically important threshold of 0.5 was significantly lower in the FAME cohort (OR 0.37, 0.21, 0.66). Fall-related acute events occurred in 6.3% of patients in the intervention group vs. 11.0% in control patients over a one-year period (P = 0.10).

Conclusions:

The program was associated with a significantly lower odds of further increasing FRID burden at 1 year compared to matched controls. An electronic case review and telephone counseling program has the potential to reduce drug-related falls in high-risk older adults.

INTRODUCTION

Falls are the most common and costly medication-related safety event in older adults, with a fall occurring in the U.S. every second.^1,2 Falls in older adults result in approximately 3 million emergency department visits, over 300,000 hip fractures, and are associated with an estimated cost of $50 billion each year.^1,3,4 In addition to fractures and other fall-related injuries, falls lead to disability, risk of institutionalization, activity restriction and reduction in quality of life. Older Veterans have a high prevalence of fall risk factors, of which fall risk-increasing drug (FRID) use is a major factor and a key target for fall prevention.^5,6

A robust body of evidence supports medications as a major causal factor for falls in older adults. Psychoactive medications including benzodiazepines, antidepressants, antipsychotics, and anticholinergic medications have been consistently associated with a 2-fold or higher risk of falls and fractures in older populations.⁷ While less frequently studied, evidence also links antihypertensive medications (especially diuretics and alpha-adrenergic antagonists)^7,8, anti-epileptic drugs⁹, and hypoglycemic drugs (especially insulin and sulfonylureas and in the setting of “tight” glycemic control) to falls in community dwelling older adults.^10,11

Deprescribing fall-related medications, defined as intentionally stopping or reducing the dose of a medication to improve health or reduce the risk of adverse effects, is therefore a key step in reducing the burden of falls in older Veterans. Deprescribing has been a part of almost all multi-component fall prevention strategies tested, with overall reductions in fall rates of 26% in a recent meta-analysis.^12,13 Two studies have tested deprescribing as a single intervention to prevent falls; a pharmacist-led intervention targeted to patients on psychoactive medications reduced fall rates by 66%, and a prescribing modification program for primary care physicians reduced falls by 39%.^14,15 With 75% of older adult Veterans at risk for falls taking 2 or more FRIDs⁶, identifying scalable interventions to facilitate deprescribing of fall-related medications is a priority. We designed a pilot study of a remote, Electronic Health Record (EHR)-based population deprescribing program, Falls Assessment of Medications in the Elderly (FAME) Deprescribing Program, to reduce fall-related medication use in high-risk Veterans. The purpose of this study is to evaluate the effectiveness of the program in reducing FRID burden and falls-related acute visits over one year.

METHODS

Centralized Case-finding

Veterans aged 65 or older, who were being followed by a primary care provider at the Durham Veterans Affairs (VA) Health Care System and who met one of the following criteria were eligible for electronic deprescribing case reviews: (1) a history of falls on their active problem list, (2) an outpatient clinical encounter for a fall between September 2016 and September 2018 (ICD-10 code), or (3) a positive result from an annual primary care falls risk screening between September 2016 and September 2018. Additionally, the Veterans had to have an active prescription for ≥1 target FRID at the time of screening.

To identify eligible Veterans, a patient list was generated on a yearly basis by querying the Veterans Affairs Corporate Data Warehouse, which is an EHR data repository. In order to have a representative sample across different FRID classes to assess feasibility and acceptability of deprescribing recommendations, Veterans were randomly selected from each medication class for the case reviews. The Durham VA Health Care System institutional review board approved this study.

Description of FAME Deprescribing Program

The program involved bi-weekly in-person or video conference team meetings for electronic case reviews. The interdisciplinary FAME team consisted of two geriatricians, three geriatric clinical pharmacists, one to two geriatric pharmacy residents, one rotating geriatric fellow, and a registered nurse. Before the conference, geriatrics trainees reviewed and prepared three to five cases each by reviewing EHRs, which included outpatient and inpatient progress notes, medication data, and lab data. The case review meetings lasted approximately one hour, during which six to ten cases were reviewed and discussed. The average time spent reviewing each patient chart was 10-30 minutes. The targeted falls-related medications included alpha-adrenergic antagonists, anticholinergics, antidepressants, gabapentinoids, antihyperglycemics, antipsychotics, benzodiazepines, diuretics, and sedative/hypnotics. Opioids were not part of the FAME program’s focus as there were already various VA programs aimed at optimizing opioid prescribing, but were considered as covariates. Deprescribing recommendations, including indications/contraindications for a deprescribing trial and suggested tapering schedule, were developed using evidence-based algorithms and clinical judgement [Supplemental Table S1].^16–21

The FAME team made a maximum of three deprescribing recommendations after team consensus. The recommendations were then sent electronically to the primary care provider (PCP) and mental health providers for approval via a templated FAME progress note in the EHR. The providers could accept the recommendations by co-signing the note, make changes, or opt out of the program for their patient. A FAME team pharmacist or nurse would then call the Veteran to review the recommendations, and the Veteran could choose to accept or reject them. If either the provider or Veteran declined all recommendations, the FAME team would not follow that Veteran. If the Veteran agreed with at least one of the recommendations, a member of the FAME team would place orders for the PCP to co-sign and mailed educational materials to support adherence to the changes.^22–25 The FAME team monitored the deprescribing plan through follow-up phone calls at 1 month and 3 months, during which they assessed adherence, withdrawal adverse effects, addressed barriers, and adjusted the plan as needed. Initial phone calls for counseling Veterans averaged 15 minutes, while follow-up calls took about 5 to 10 minutes. All deprescribing recommendations, provider and patient acceptance, and withdrawal adverse effects were recorded in a secured database.

Patient Characteristics

Demographics, including age, gender, race, as well as comorbidities associated with falls-related medication use were examined. Demographic data and comorbidities were manually abstracted from the EHR. Comorbidity data came from active patient problem lists.

Program Measures

Acceptability

We categorized the acceptability of deprescribing recommendations at both the provider and patient levels. Acceptability of recommendations was examined at the individual medication level and then grouped by medication class.

Tolerability and Safety

Intolerance to deprescribing was defined as a return to the initial dose of the target medication within 1 month, as assessed during program telephone calls and/or EHR notes. Reasons for not tolerating the medication change were qualitatively examined. Safety was measured as adverse drug withdrawal events, defined as an acute outpatient /emergency department/hospitalization visit for uncontrolled hypertension, hyperglycemia, urinary retention, worsening pain, anxiety, or depression within 30 days of case review date, using EHR chart review and two reviewer concurrence for relatedness to deprescribing.^26,27

Clinical Outcomes

A modified Drug Burden Index (DBI) was calculated to determine total exposure to anticholinergics/sedative medications and was selected to potentially capture dose changes and individual medication discontinuations.²⁸ Medication classes included in the DBI were anticholinergics, antidepressants, gabapentinoids, antipsychotics, benzodiazepines, and sedative/hypnotics. DBI was calculated as Daily Dose / (Daily Dose + Minimum Daily Dose). Range of scores for each drug is between 0 and 1, and these are summed across drugs. The primary outcome was change in the modified DBI, measured at baseline (date of case review) and at 1-year follow-up.^29,30 A clinically meaningful change in DBI was set at 0.5, which reflects dose tapering or discontinuation of at least 1 DBI-related medication. Change in DBI was dichotomized as either having or not having a change of ≥ 0.5.

Injurious falls were an exploratory outcome, measured as an acute outpatient visit, emergency department visit, or hospitalization for fall within 1 year from the FAME case review date. Falls data was captured using a Voogle search (a search engine for rapid query of VA EHR notes), using the search term “fall” or “falls”. Identified falls events were then verified using two reviewer concurrence.

Statistical Methods

Sample:

The initial EHR data abstraction included all eligible patients aged ≥65 years old followed by Durham VA Health Care System primary care between September 1, 2016 and September 1, 2018 (N = 18,727), and at high risk for falls (n = 9228). Only patients prescribed ≥1 target falls-related medication were included in the screening sample (n = 6,884), and from here patients were randomly selected for electronic case reviews (Figure 1). We selected a sample of 250 case reviews for feasibility and power reasons. With this sample size, an alpha level of p < 0.05, and a clinically meaningful change in DBI score of 0.5, we anticipated sufficient power to detect moderate to large effects but less power for detecting small effect sizes. The final sample for program evaluation was 294, and consisted of patients who had a completed case review during the evaluation period from 08/17/2018 to 03/20/2020. Veterans were excluded if they had a life expectancy of < 6 months, were residing in a long-term care facility, had documented medical or psychiatric instability, or had their target medications recently discontinued. If we discussed their case but excluded them based on exclusion criteria, they were considered ineligible (n = 58). Thus, there were 236 Veterans included in the evaluation (Figure 1). To derive a comparison cohort, control patients were matched based on age, falls-related medication class use, and date of matched FAME review. Overall, 472 patients (236 intervention cases, 236 matched controls) were included in the study. The intention-to-treat group were all Veterans who had at least 1 FAME recommendation sent to PCP or Mental Health (n = 236). The per protocol group includes those who had the FAME recommendations sent out, did not become ineligible or die during the 1-year follow-up, had the PCP and/or Mental Health accept at least recommendation, and patient agreed to trial of deprescribing (n = 147). Model-specific exclusions were made to ensure at least a 6-month follow-up was available for clinical outcomes.

Figure 1.

Patient screening and inclusion for FAME team case reviews

DVAHS=Durham VA Health Care System; FAME = Falls Assessment of Medications in the Elderly; FRID = Fall-risk increasing drug

Analysis:

Descriptive statistics were calculated for patient characteristics, deprescribing recommendation acceptability, tolerability, safety, and clinical outcomes. Chi-square and Wilcoxon Rank Sum tests were performed where appropriate. We set our level for statistical significance at 0.05. Logistic regression models were fit both as intention-to- treat and per protocol to determine the relationship between FAME deprescribing electronic case reviews and change in DBI score by ≥0.5 at 1 year and falls within 1 year of the case review completion date. The statistics were generated using SAS/STAT software (SAS Institute Inc. 2020. SAS/STAT^® 15.2 User’s Guide. Cary, NC: SAS Institute Inc.).

Results:

Characteristics of Veterans in the FAME intervention and control groups are described in Table 1. Overall, Veterans were predominantly male 95%, white 65%, and had high prevalence of comorbidities associated with falls-related medication use, including hypertension (73%), depression (47%), chronic pain (47%), diabetes (44%), or post-traumatic stress disorder (42%); mean (standard deviation) age was 73.2 (6.0) years. There were significant differences between the intervention and control Veterans in terms of cardiovascular disease, hypertension, insomnia (which was more prevalent in the control group), and chronic pain (which was more prevalent in the FAME group). The medication classes were generally balanced between the two groups, except for anticholinergics, which were more common in the FAME group than in the control group (Table 1).

Table 1.

Patient characteristics of FAME intervention and control groups at baseline

Patient Group	FAME Group (N = 236)	Control Group (N = 236)	P value
Age, mean (SD)	73.2 (6.1)	73.3 (6.0)	0.79
Sex=Male	94.9	96.6	0.38
Race, White	65.3	68.8	0.72
Anxiety	14.4	16.2	0.58
Arrhythmia	16.5	21.4	0.18
Congestive Health Failure	5.5	8.6	0.20
Chronic pain	46.6	33.8	0.005
Cognitive Impairment	15.7	11.1	0.15
Cardiovascular Disease	14.4	29.5	<0.0001
Depression	46.6	50.0	0.46
Diabetes	44.1	43.2	0.84
Hypertension	72.5	83.3	0.005
Insomnia	17.0	37.2	<0.0001
Other Psychiatric Conditions	14.0	38.9	<0.0001
Post-traumatic Stress Disorder	42.8	38.5	0.34
Falls in last 3 months	24.2
% with 1 or more meds in each medication class
Alpha-adrenergic Antagonists	40.7	45.3	0.31
Anticholinergic	61.4	50.0	0.012
Antidepressants	69.1	79.8	0.69
Antipsychotic	17.0	13.6	0.31
Benzodiazepine	22.0	16.5	0.13
Diuretics	26.3	24.6	0.67
Gabapentinoids	52.5	48.3	0.36
Hypoglycemic	34.8	28.8	0.17
Sedative-Hypnotic	14	14.8	0.79

The frequency of FRID use in our population has been previously documented in a paper by Elias et al. The study found that among Veterans aged 65 years or older who were enrolled in Primary Care and prescribed at least 1 fall-related medication, 27% were prescribed 4 or more FRIDs. Additionally, nearly half of the Veterans were prescribed antidepressants or diuretics.⁶ In the FAME intervention group, antidepressants (69%) and anticholinergics (61%) were the most frequently prescribed FRIDs (Table 1).

Acceptability

PCPs accepted at least 1 or more recommendations in over 90% of FAME case reviews (213/228), and Mental Health providers accepted at least 1 or more recommendations in 70% of FAME case reviews (58/83). There is some overlap as either PCP and/or Mental Health may have received recommendations from a single Veteran. There were 45 Veterans excluded prior to contact either due to provider opting out of the FAME recommendations (n = 24), FAME team decided not to contact the patient because their health status acutely changed (n = 11), or the patient was unable to be reached (n = 10). Of the patients contacted, 71% opted into the FAME program to taper or discontinue at least one fall-related medication. Among those that opted out, the most frequent reasons for declining were recent health status change, or hospice initiation, and in few cases, Veteran did not tolerate previous dose reductions.

Figure 2 shows the number of accepted and declined deprescribing recommendations for each medication class by provider type and patients. Anticholinergics and antidepressants were the most recommended classes. PCPs were more likely than Mental Health providers and patients to accept deprescribing recommendations across all medication classes. Less than half of the patients accepted deprescribing recommendations for antidepressants and benzodiazepines.

Figure 2.

Acceptance of FAME deprescribing recommendations by providers and patients.

FAME = Falls Assessment of Medications in the Elderly; PCP= Primary Care Provider; MH = Mental health provider; Pt = patient

Tolerability and Safety

Out of the Veterans who agreed to deprescribe at least one fall-related medication, 31% (37 patients) did not tolerate at least one of the medication changes and had to go back to the initial dose due to worsened symptoms. The majority did not tolerate the change within the first month (26 out of 37), mainly due to worsened symptoms, and some were restarted on the medication for a new indication. Five patients who did not tolerate the taper due to worsened symptoms were prescribed the fall-related medication for an off-label indication. Anticholinergics, gabapentinoids, and benzodiazepines were the most common medication classes not tolerated for taper or discontinuation. One adverse drug withdrawal event occurred, which required an unplanned medical visit within 30 days. It was a patient who presented to the emergency department for worsened nasal congestion due to seasonal allergies and was discharged home with cetirizine, the medication recommended for deprescribing by FAME.

Clinical Outcomes

In the intention-to-treat analysis, the 1-year mean overall change in modified DBI was −0.15 (95% CI −0.23, −0.08) in the intervention cohort and −0.11 (−0.21, −0.00) in the matched control cohort (p=0.47) (Table 2). In the per protocol analysis, the changes were −0.25 (−0.36, −0.13) and −0.11 (−.21, −0.00), respectively (p=0.08). In the intention-to-treat analysis, the odds of increasing DBI by a clinically important threshold of 0.5 was significantly lower in the FAME cohort (OR 0.37 (0.21, 0.66), p <0.001). Fall-related acute events occurred in 6.3% of patients in the intervention group vs. 11% in control patients over a one-year period (p=0.10) (Table 2).

Table 2.

Descriptive statistics of outcomes in FAME intervention and control groups.

	Baseline		P-value	1 Year Post Consult:		P-value
ITT Analysis	FAME Cases (N = 236)	Control Group (N = 236)		FAME Cases (N = 236)	Control Group (N = 236)
Drug Burden Index (ITT)	2.04 (1.10)	1.65 (0.99)	0.11	1.89 (1.14)	1.55 (1.11)	0.68
Overall Mean DBI Change				−0.15 (0.57) 95% CI (−0.23,−0.08)	−0.11 (0.79) 95% CI (−0.21, −0.00)	0.47
Acute Falls Events, Any				15/236 (6.3%)	26/236 (11%)	0.10
DBI Increase by ≥ 0.5				18 (7.6%)	43 (18.2%)
Odds of DBIa increase by ≥ 0.5				OR 0.37 (0.21, 0.67)		<0.001
Per Protocol Analysis	FAME Cases (N = 236)	Control Group (N = 236)		FAME Cases (N = 147)	Control Group (N = 236)
Drug Burden Index (DBI)	2.04 (1.10)	1.65 (0.99)	0.11	1.82 (1.13)	1.55 (1.11)	0.76
Overall Mean DBI Change				−0.25 (0.71) 95% CI (−0.36,−0.13)	−0.11 (0.79) 95% CI (−0.20,−0.00)	0.08
Acute Falls Events, Any				15/236 (6.3%)	26/236 (11%)	0.10
DBI Increase by ≥ 0.5				18 (12.2%)	43 (18.2%)
Odds of DBIa increase by ≥ 0.5				OR 0.63 (0.35, 1.1)		0.12

DBI = Drug Burden Index; FAME = Falls Assessment of Medications in the Elderly; ITT = intention to treat

^aOdds of DBI increase vs. no change or decrease by same threshold

DISCUSSION

Interdisciplinary electronic case review with telephone counseling and intervention led to a modest decreased burden of FRIDs in community-dwelling older Veterans at risk for falls. Patients who received FAME review and intervention experienced slightly greater average reductions in FRID burden, but overall had significantly lower odds of clinically important increases in FRID use at one year, suggesting that the intervention may have more impact in preventing additional FRID prescriptions than in decreasing existing use. Although the outcome was exploratory and underpowered to detect a significant difference, there were fewer acute fall-related visits among FAME patients, suggesting a potential benefit of the intervention.

Our fall risk reduction process targeted drug burden, a measure associated with decreased physical, cognitive, and functional performance.³¹ In the Health ABC study, a one-unit increase in drug burden negatively affected physical and cognitive function, equivalent to 3-4 additional physical comorbidities.²⁸. A one-unit increase in drug burden was also associated with an increased risk for hospitalizations and mortality, informing our threshold of 0.5 units for a clinically significant change in drug burden.³¹ While the risks of drug burden are well documented, the effectiveness of interventions to reduce it and mitigate these risks is less studied.

Current evidence supports the effectiveness of FRID deprescribing in reducing the burden of FRIDs, but it is not yet clear whether this intervention can effectively prevent fall-related events.^32–34 Two older trials have showcased a direct benefit on fall-related events for collaborative deprescribing in community-dwelling older adults, although in small populations with short trial durations (≤ 6 months).^14,15 Although our analysis was not designed to detect differences in fall-related events, our exploratory analysis suggests that a larger study may reveal a potential benefit of the intervention. Overall, making direct comparisons with other deprescribing trials is challenging due to the considerable heterogeneity in settings, targeted medications, and studied outcomes.^32–34 Unlike some deprescribing efforts, our approach addressed a wide range of medication classes, facilitating an efficient process with multiple recommendations per patient.^14,17

Implementation of FAME deprescribing recommendations was well tolerated, safe, and similar to rates seen with other deprescribing initiatives that assessed tolerability/safety.^17,35 Only one patient experienced an adverse effect attributable to medication withdrawal (worsening allergies). While these findings further reinforce the known safety of evidence-based deprescribing initiatives, it is possible that withdrawal effects were under appreciated given limited follow-up frequency and the lack of integration with prescriber teams.³⁶

Patients and prescribers both demonstrated high acceptance of medication deprescribing recommendations, as evidenced by mutual agreement. The use of an interdisciplinary approach, along with telephone visits to formulate and communicate deprescribing plans, likely contributed to the observed high acceptability of recommendations. This is especially important because many older adults are not aware of which medications increase the risk for falls but are receptive to changes when provided with this information.^37,38 Providing education to both prescribers and patients may also contribute to benefits in FRID burden and fall risk beyond our period of observation. However, some exceptions should be noted. While PCPs and Mental Health providers had high acceptance rates for deprescribing antidepressants and benzodiazepines, fewer patients—less than half—adopted similar recommendations. The contrasting acceptance rate warrants further investigation into contributing factors. Moreover, 31% (37 patients) encountered difficulties tolerating recommended medication changes and reverted to initial doses due to worsened symptoms. Notably, our study highlighted anticholinergics, gabapentinoids, and benzodiazepines as particularly challenging to taper or discontinue, suggesting the necessity for cautious strategies when deprescribing these medications.³⁶

Program sustainability is crucial. Engaging learners like geriatrics fellows and pharmacy residents integrates the FAME program into their curriculum, enriching their education and enhancing sustainability. However, assessing the value of follow-up phone calls remains a persistent concern. The program’s scalability is feasible. Despite varying resources across settings, standardized protocols and note templates ease dissemination. The intervention could involve primary care, mental health, or other health system facility PharmDs or Geriatricians. Involving learners like pharmacy students and housestaff also enhances education. Furthermore, training clinic nurses or allied health members using standardized protocols for telephone visits could replace dedicated nurses and facilitate implementation. Additionally, one deprescribing team could serve multiple health system sites via remote EHR access and clinical privileging. Cost-effectiveness is also a significant consideration. Diverse deprescribing strategies may be necessary to address varying clinical contexts. While lower-cost programs like EMPOWER can be effective for specific classes in some populations,³⁹ this VA cohort with central nervous system polypharmacy and a high prevalence of high-risk comorbidities may require more personalized plans and increased clinical support. Future research should focus on optimizing cost-effective deprescribing strategies by matching them to specific populations, potentially employing them sequentially.

Our study has several limitations. The single-center design may limit the generalizability of our findings, but our structure as an interdisciplinary teaching team within an academic medical center can be applied to similar practice sites. Additionally, the FAME program’s reliance on the VA formulary system for medication recommendations may limit generalizability to settings with diverse medication formularies and costs. While the VA system significantly reduces medication costs for Veterans, this might not entirely mirror the financial constraints experienced by patients in other healthcare settings, potentially impacting deprescribing decisions. Fall outcomes were abstracted through EHR review, which will not fully capture all fall events. Future work can consider including self-reported falls measures. Furthermore, we were unable to track medications prescribed outside of the health system, which may lead to an underestimation of the true burden of FRIDs in our population. Deprescribing plans were communicated electronically to prescribers and by unscheduled phone calls to patients, without a previously existing relationship with the FAME team. While the acceptance rate of recommendations was high, it is possible that acceptance may have been even higher in the context of preexisting clinical relationships. Future work could explore differences in acceptance rates between primary care and mental health. Furthermore, it is possible that the guidance provided by FAME recommendations could have prompted primary care and mental health providers to modify their approach with other patients, leading to less divergence in deprescribing results between the groups than anticipated. Our study also did not survey patients and providers to gather their perceptions of medication recommendations. Using a validated survey to analyze the experience of both patients and providers receiving deprescribing recommendations with varying outreach methods may provide valuable insights for future studies.⁴⁰

CONCLUSION

The FAME deprescribing program effectively reduced FRID burden and potentially lowered fall-related events. The high acceptance rate of deprescribing recommendations by both patients and providers further supports the potential for success of this intervention. Future efforts, including surveying patient and provider perceptions of deprescribing recommendations, and analyzing a larger intervention population, can refine and improve program effectiveness. Exploring different electronic communication methods and expanding the intervention to a broader population can inform the development of effective medication deprescribing strategies.

Key Points:

• Our program’s implementation reduced the chances of raising fall risk-increasing drug burden further, indicating its potential to improve medication management and reduce polypharmacy in older adults.

• This electronic deprescribing program was feasible to implement across the health system and acceptable to patients and providers, and has the potential to be widely adopted in healthcare settings.

Why does this matter:

By utilizing centralized case finding, electronic case review, and telephone counseling, our program demonstrated the potential to significantly reduce drug-related falls in high-risk older adults, promoting safer medication use in this vulnerable population.