Get started
Work

Patient adoption of pharmacist recommendations to older adults presenting to emergency department with falls: A secondary analysis of GAPcare

Year: 2021
Type: article
Abstract: Among adults aged 65 and older, falls are the leading cause of both fatal and nonfatal injuries; annually, they result in 3 million emergency department (ED) visits, 950,000 subsequent hospitalizations, and 32,000 deaths.1 To address this crisis, the 2013 Geriatric Emergency Medicine Guidelines state that ED clinicians should initiate fall prevention efforts by screening for polypharmacy and fall risk–increasing medication.2, 3 Organizations such as the National Quality Forum have also endorsed quality metrics related to the assessment of medications that increase the risk of falling.4 Despite these recommendations, there is limited research on ED-based interventions to address medication risk factors. Interventions targeting medications outside the ED have relied primarily on pharmacist review alone and have not included direct consultation with patients, despite the fact that patient engagement is necessary to modify medication-related choices.5-7 Here we present the results of a pharmacist-led motivational interviewing–based intervention aimed at reducing subsequent falls and health care utilization among older adults who present to the ED after a fall. This pharmacy intervention was part of GAPcare, the Geriatric Acute and Post-Acute Fall Prevention Intervention in the Emergency Department, a randomized controlled trial developed and tested at two urban academic EDs.8, 9 As part of the intervention, pharmacists reviewed the participants' medications and asked open-ended questions to gain an understanding of participants' knowledge of their medications. Pharmacists then assessed participants' willingness to change medications to reduce fall risk. Together, participants and pharmacists formulated a medication-related action plan, which identified one to three medication recommendations to reduce fall risk. Plans were shared with primary care physicians (PCPs). We evaluated the medication changes suggested by pharmacists and examined factors associated with participant uptake of these recommendations at the 6-month follow-up visit. Research staff reviewed the ED electronic health record (EHR) to screen patients 65 years old and older who presented to the ED within 7 days of a fall and were likely to be discharged from the ED. Participants were consented and randomly assigned to the usual care or intervention arm and followed for 180 days. A total of 110 participants were recruited (n = 55 per arm). Participants in the intervention arm received pharmacy and physical therapy consultations during their ED visits, while participants in the usual care arm received brochures on home safety. Licensed pharmacists performed a 20-minute medication therapy management session using elements of motivational interviewing. They asked participants open-ended questions to determine their knowledge of their medications and assessed their willingness to make changes to these medications. Pharmacists then made one or more recommendations about one to three medications as needed to decrease fall risk. The pharmacist drafted a consultation note in nontechnical language and gave it to the participant or caregiver. The note was electronically transmitted to their PCP and saved within the EHR. Research staff abstracted recommendations following best practices in chart review and classified the type of recommendation made in the pharmacy note. At 180 days postenrollment, research staff conducted a home visit with study participants and asked participants to self-report their uptake of each recommendation on a Likert scale as "none," "partial," or "full." After study completion, the specific medications were classified into categories according to the Beers criteria.3 Drugs not specifically listed in the Beers criteria were further analyzed and grouped into seven broad categories. Descriptive statistics about study participants were summarized by medians and interquartile ranges for nonnormally distributed continuous variables and counts and percentages for categorical variables. The associations between uptake of recommendations by type of recommendation and class of medication were examined using cross-tabulation and confidence intervals (CIs). Exact CIs were calculated; R (4.0.2) and SAS (9.4) were used for all analyses. Of the 55 participants in the intervention arm, 54 participants received the pharmacy intervention. Recommendations were made for 53 participants, and at least partial follow-up data were obtained from 39 participants. The median (IQR) age of participants was 83 (76–90) years, of whom 92% were White, and two-thirds were women. Participants with and without uptake data did not differ significantly in terms of demographics. Pharmacists made 219 medication recommendations with a median of four total specific recommendations concerning two medications per participant. Both the total number of recommendations and the type of recommendations made varied by drug class (Fisher's exact test p < 0.001). Pharmacists most frequently addressed medications to treat pain (n = 70); among recommendations in this drug class, the most common recommendation was to use an over-the-counter (OTC) alternative (n = 28). Gastrointestinal medications (n = 41) were the second most frequently mentioned drugs. We found a relationship between the type of recommendation and drug class: over 40% of recommendations for increased precautions were for cardiovascular drugs, and the majority of recommendations for a decrease in dosage (n = 15) were for neurological medications and antidepressants (n = 8). We also examined uptake of recommendations by recommendation type and medication class (Table 1). Participants were more likely to report uptake of advisory recommendations (full uptake 66% [95% CI = 55% to 81%] and partial uptake 15% [95% CI = 4% to 30%]), such as "discuss further with your prescriber" or "use increased precautions with a medication." Uptake was lower for specific actions (full uptake 39% [95% CI = 28% to 51%]) and partial uptake 25% [95% CI = 14% to 37%]). Within the specific action category, participants had higher uptake of "changing timing" and "stopping medication" than "decreasing dose," "starting a new medication," or "using an OTC medication." Uptake was greatest for anticholinergic and cardiovascular medications, with over 70% of participants reporting either partial or full uptake of recommendations in these drug classes. Self-reported uptakes of neurological and antidepressant medication recommendations were lower. Reducing the risk of falls among older adults is an important health care priority. Our evaluation of an RCT motivational interviewing intervention delivered by pharmacists found that participants partially or fully adopted over two-thirds of recommendations to address polypharmacy and fall-risk increasing medications, but recommendation uptake depended on the type of recommendation and drug class. This has important implications for the design of future ED-based interventions, because it suggests that while a brief motivational interview may successfully change medication-related behaviors, different approaches may be needed for actions such as decreasing drug doses and modifying the use of psychoactive medications. These changes often require closer follow-up and more incremental changes over time and therefore could be more challenging for patients and clinicians. Larger future studies may provide more information on specific recommendation types and drug classes that are more amenable to brief ED based interventions. Given GAPcare's more general success in reducing repeat fall-related (adjusted incidence rate ratio = 0.34 [95% CI = 0.15 to 0.76]) and all-cause ED visits (adjusted incidence rate ratio = 0.47 [95% CI = 0.29 to 0.74]),8 and our findings of adoption of the majority of recommended medication changes, this suggests that medication modification might be an important factor in reducing fall risk. The risk/benefit profile of medication use often changes after a fall and medications that were once considered advantageous to health may no longer be considered safe. However, because older adults underreport falls to PCPs, opportunities to address medications are often missed. In GAPcare, pharmacists uncovered several prescribing problems that helped reduce fall risk by increasing medication safety and reducing or stopping medication use. These results also suggest that our approach needs to be refined; participants in our study were far more likely to adopt recommendations that were advisory, such as discussing recommendations from the ED with their prescribers or taking increased precautions, compared to more specific actions such as decreasing medication dosage, stopping medications, or starting new medications. Although pharmacist recommendations were provided in writing to PCPs, these findings may suggest that patients prefer to have the approval of PCPs or other specialist prescribers before making changes. Enhancing communication with these clinicians could encourage greater uptake of recommendations. While previous ED-based fall prevention programs have included education about the association between falls and certain medications and medication review with a physician as part of a multifactorial intervention, these studies did not incorporate a licensed pharmacist and recommended changes in medication and uptake to these changes were not measured.10, 11 In the United States, prior pharmacist-led interventions have shown limited efficacy in reducing health care utilization.7 However, these studies have not directly evaluated whether patients or their physicians adhered to these recommendations, so poor uptake may be a reason for the limited efficacy. In contrast, previous studies internationally have shown the efficacy of ED interventions in reducing polypharmacy.5, 7 There are several limitations to our study. Our data on uptake were self-reported by participants and may not accurately reflect the participants' actual levels of uptake. Furthermore, we did not systematically define the meaning of "full" or "partial" uptake, so there may have been differences in both participant and research staff interpretation of these terms. Fourteen of the 53 participants did not have a follow-up visit, and their responses may have varied from those who completed the study. More generally, pharmacists were instructed to limit their recommendations to what they viewed as the three most important recommendations; there may have been variation between pharmacists. Furthermore, recommendations were tailored to participants' current medications, health history, and willingness to consider medication changes; therefore, these results may not be generalizable to other patients on the same medications or to different pharmacists interacting with the same patient. We also do not know the role that the PCPs or specialists played in the adoption of these recommendations. More research is needed to examine interventions in the ED setting to reduce fall risk by motivating patients to feel empowered to change their medications under the supervision of their medical care team. While GAPcare provides a model for future interventions that is brief and feasible in the ED setting, replication of our work in a larger study and a refinement of our approach to better target more difficult to adopt medication changes is needed. The authors acknowledge the assistance of Jessica Nagy and Joseph Aquilante in leading the pharmacy intervention for the study.
DOI: 10.1111/acem.14302
Language: English
Field: Medicine
Open Access status: green
OpenAlex ID: W3164864201
FWCI: 0.138
Cites: 11
Cited by: 5
Related to: 10