The US population is aging rapidly; the population aged 65 years and above is projected to be doubled by 2050. The most rapidly growing segment is the population of older adults aged 85 years and above, which will more than triple from 5.9 million in 2012 to 18 million in 2050 [1,2]. Older adults aged 65 years and above often have multiple chronic diseases—more than 60% have 2 or more chronic diseases [3], and 17% have 4 or more chronic diseases [4]. Older adults with multiple chronic comorbidities often require multiple medications for management, particularly with guideline-based management of chronic diseases [5,6]. The use of multiple medications in older adults is common, with almost 20% of older adults aged 65 years and above taking 10 or more medications [7,8]. Multiple-medication use in older adults is associated with a lower adherence rate and increased use of inappropriate medications [9,10]. The use of multiple medications increases the risk for potential drug interactions, leading to undesirable adverse drug reactions (ADRs), which could also contribute to lower medication adherence. In particular, older adults are more susceptible to ADRs due to the changes in their physiology, clearance, and reserves [11], particularly with polypharmacy [9,12] Based on the phycological changes in the older population, knowing the drug regimens with older adults is an important feature in order to provide medication safety and make adjustment to their regimen accordingly to ensure appropriate medication use. The START/STOPP (Screening Tool to Alert to Right Treatment/Screening Tool of Older Persons’ Prescriptions) criteria [13-15] and American Geriatrics Society (AGS) Beers criteria [16,17] for potentially inappropriate medications (PIMs) were developed to provide evidence-based guides to signal clinicians about medications that are potentially inappropriate and to enhance medication appropriateness. These medications include those with strong anticholinergic properties, which may disproportionately affect older adults and are linked to adverse outcomes of ADRs, and other medications that are demonstrated to have significant side effects in older adults.

Telemedicine is a modern visit option enabled by advances in telecommunications technology [18]. The use of telemedicine in older adults has been examined in prior small studies that have demonstrated feasibility, acceptability, and user satisfaction [19-22]. Extending televisits to patients’ homes has the potential to impact many aspects of care that rely on patient self-management, such as medication use. Although telephone-based pharmacist interventions have the potential to impact medication use at home [23,24], they still rely on accurate information reported by older adults over the phone without a mechanism for confirmation. The addition of video has the potential to further enhance the visit [25,26] by (1) visual ascertainment of actual medications taken by patients, (2) visual demonstration of the patients’ actual use of medications, and (3) education of patients on proper use. Although in-person home-based reconciliation has the potential to improve accurate appraisal of medication use and reconciliation, it is not feasible for wide adoption as it is resource intensive. The study proposed here will examine the impact of home televisits by pharmacists on patients at high risk for ADRs (ie, with polypharmacy and multiple chronic conditions).

This study is designed as a mixed methods hybrid type 1 effectiveness implementation study [27], where we will test the health impact of pharmacy televisits, while also collecting data on the implementation process to facilitate subsequent scale-up efforts [28]. As older adults have higher rates of chronic conditions and polypharmacy, we included veterans aged 65 years and above in this study. The inclusion criterion of 5 or more medications is based on prior literature on polypharmacy and findings of increased risk for drug interactions and ADRs. The enrollment criteria are consistent with and exceed the criteria set in fiscal year 2019 by the Center for Medicare Services for provision of medication therapy management in Medicare Part D [29]. Among patient characteristics, polypharmacy (≥5 medications) and multiple chronic conditions are considered important risk factors for ADRs and are highly prevalent among frail older adults at risk for ADRs [30].

The intervention for the cluster randomized controlled trial (RCT) was iteratively adapted after a pilot test with 20 veterans [31]. Home televisits, where pharmacists conducted medication reconciliation and management, were refined, after which a formative evaluation [32] was conducted, guided by the Consolidated Framework for Implementation Research (CFIR) [33]. Data were gathered in several ways: (1) enrolled veterans (n=20) were observed by study staff present within the veterans’ homes during the clinical pharmacist home televisit, and (2) study staff administered a postencounter questionnaire with the goal of improving the veteran-based technical experience and clinical encounter. The data points were used to adapt and make changes to the intervention for the cluster RCT.

The cluster RCT is registered on ClinicalTrials (NCT04340570). The intervention will be a pharmacist-led home televisit including medication reconciliation and assessment of actual medication use. Pharmacists will review medication appropriateness using evidence-based criteria and provide recommendations for change of medication use to the patients’ primary patient-aligned care team (PACT). PACTs provide team-based care and consist of a primary care provider (PCP), a registered nurse, a medical assistant, and, often, either a social worker or a pharmacist. In the control group, participants will receive usual care in which medication reconciliation and review will be conducted in clinics by primary care teams. As a primary outcome, we will examine the effect of the intervention on the number of veterans with PIMs, as determined using evidence-based criteria. As secondary outcomes, we will determine the number of PIMs, medication discrepancy using data from record reviews and interviews, and medication appropriateness using validated instruments at 6 months after the intervention. We will also compare the intervention’s effects on the incidence of ADRs using data from record reviews and interviews. To assess the potential for future implementation of the intervention, we will administer postintervention questionnaires to key stakeholders, including veterans and PACT clinicians, to examine implementation barriers to and facilitators of the intervention.

This study was approved by the Veterans Affairs (VA) Central Institutional Review Board (CIRB; IRBNet ID 1612635). To obtain participants’ informed consent, a research assistant (RA) will determine whether eligible patients have the capacity to provide informed consent to participate in the study using a screening questionnaire that assesses the 4 elements required for capacity understanding of study procedures, appreciation of what will happen if enrolled, communication of a choice to enroll or not, and demonstration of a rationale for that choice. Recruitment and written informed consent will take place in a location that ensures privacy and convenience for the patient.

Our conceptual framework is informed by the CFIR [34] and Steinman et al [35] on prescribing and prescription monitoring processes. The pilot phase was primarily informed by the Steinman model of the prescribing process adapted to include system- and provider-level factors. The system-level factors include the Veterans Health Administration (VHA) policy on medication management (including non-VA care use), facility-level prescribing environment and practices, and decision support systems. In designing our proposed intervention, we considered the potential effect of home televisits by pharmacists to ascertain the medication regimen and use, identify discrepancies and educate patients on use, and review medication appropriateness, thereby enhancing steps in the monitoring phase of the prescribing process (Figure 1).

The inclusion criteria include the following: (1) the veteran must be a PACT patient in a Bronx VA or Bedford VA geriatrics or primary care clinic; (2) be 65 years or older; (3) have 2 or more chronic conditions; and (4) have 5 or more medications listed on the VA medication record continuously in the previous 6 month. Patients fulfilling these criteria will be approached for study enrollment through letters of invitation after approval by human subjects’ committees.

We will randomize patients by PACT to prevent a team from having participants in both treatment and control groups and to reduce contamination. Prior to initiating enrollment, the project coordinator will assign PACTs to intervention and control groups using lists of computer-generated random numbers in a 1:1 ratio, with separate lists for the Bronx VA and the Bedford VA, in order to maintain balance in both groups within each study site. There are 69 primary care and geriatric PACTs in the Bronx VA and 14 in the Bedford VA, and each PACT will keep the group assignment for the duration of the study.

After enrollment, veterans who agree to participate in the study will be interviewed by a trained research coordinator using survey instruments that will include data elements listed in Table 1. Veterans will receive the televisit intervention, or not, based on the PACT team they belong to, as discussed earlier. The detailed procedure for control and intervention groups is described later. To observe what occurs at follow-up appointments with the PACT provider, after the first follow-up appointment postintervention, participants will be asked about whether medication use was discussed during the appointment, whether medication reconciliation was performed and how, and what, if any, medication changes were made. Electronic charts will be reviewed for documentation of medication reconciliation and management during the visit. Primary outcomes will be ascertained at 6 months after study enrollment and are described in Table 1, including the subject survey and chart review. The subject survey for outcome assessment will be conducted by a research coordinator trained to take the subject’s medication history, and the chart review will be conducted by a clinician blinded to the study assignment. The subject survey will include confirmation of medications participants are currently taking, including names, dosages, and frequency. Participants will be asked to list medications that they take, including over-the-counter drugs and supplements. A clinician rater blinded to the study assignment will review medication data from the subject survey and chart review to determine medication discrepancies and appropriate use of medications.

Patients assigned to the intervention will have a pharmacist televisit appointment made and coordinated. Participants will be asked whether they have a device at home capable of supporting televisits, including home computers with cameras, tablets, smartphones, and a broadband or 4G connection. Participants who do not have appropriate devices will be provided with VA-issued internet-enabled tablets on which to conduct the visits. On the scheduled day of the visit, the research team will coordinate with the participants over the phone to facilitate the initialization of the televisit by the pharmacist. The content of the televisit is described next, in the Design of the Televisit Intervention section, and has been adapted through the pilot phase with direct observation of the televisit at home. Subsequent to the televisit, the pharmacist will document the visit content in the electronic medical record (EMR) to note discrepancies noted during the televisit, a review of medications with START/STOPP criteria, and recommendations based on the criteria-based review for PCP review and concurrence for adjustments of medications. Recommendations will be communicated to the PCP electronically through secure email or an electronic note or over the phone or through a face-to-face discussion if preferred by the PCP. Adjustments in the medication regimen will be noted at 7 days in the EMR after recommendations are made. If recommendations are not adopted, the pharmacist will communicate with the PCP electronically to request reasons for not adopting recommendations. Participants will then continue follow-up primary care with their PCP.

We plan to use the VA video connect (VVC) capability introduced in fiscal year 2018 to conduct televisits between clinical pharmacists located at the VA medical centers and participants at home. An appointment will be scheduled with each participant for time to conduct the home televisit. The devices, EX90s or web cameras, used at the clinician side will be located at the Bronx VA or the Bedford VA and at the participants’ side will be their choice of computer, tablet, or smartphone with an internet connection and webcam capability or a VA-provided tablet if they do not have a device or home internet capabilities. Because the experience of the televisit may differ with different participant devices, various devices were tested in phase I of the study, and their strengths and pitfalls and modes of connecting were identified.

After establishing a video connection, the pharmacist will conduct home-based medication reconciliation by asking the participants to show and explain the use of each of their medications. The pharmacist will also include a brief medication-focused functional assessment, including asking the participants to read aloud and interpret 1 or more of their prescription bottles and to demonstrate the ability to open the bottles. Before ending the visit, the pharmacist will educate the participants regarding possible side effects or interactions between their medications. The pharmacist will also solicit and answer any questions the participants have about the medications. The pharmacist will note the medications used and compare them against the list of medications on each participant’s EMR to note discrepancies and then generate a note in the EMR to notify the primary care team of the review and information obtained. If the pharmacist finds that patients are taking medications differently, they will clarify the correct usage of the medications.

After baseline MoCA and technology comfort assessment, medication reconciliation and management will be conducted by the control primary care teams in the manner they usually do. PACTs are guided by the VHA Directive on Medication reconciliation [46], which includes in-clinic assessment of medication information provided by patients via lists, recall or actual medication reviews, comparing information obtained to the VA EMR to note discrepancies, and educating the patient on updated medications. PACT PCPs may also request assessment of medication regimens by pharmacists embedded in their clinic based on clinical decisions regarding the needs of the individual patient and the availability of such services. The prescription of a new medication may trigger a review by a VA pharmacist guided by EMR-based interaction alerts. We considered using an active control group with pharmacist review (in-clinic or chart review); however, the intervention with pharmacist-led home televisits informing medication review represents a bundled intervention, and the study will examine the effectiveness of the intervention as an enhancement of care to what currently occurs (usual care). This design involving participant interviews and chart reviews will also allow us to characterize current usual medication management for older adults in PACTs.

Postintervention questionnaires will be administered to key stakeholders, including veteran participants and PACT clinicians, to assess the potential for future implementation of the intervention. This will allow us to examine implementation barriers and facilitators. An invitation to complete the postintervention questionnaires will be emailed to PACT providers in the intervention arm. To ensure we understand PCP decisions on pharmacist recommendations, we will track PCP adoption of recommendations provided by pharmacist reviews and examine PCP reasons for adoption or nonadoption of recommendations, as well as provider factors that can influence those decisions. Patient and clinician interviews and questionnaires will identify perceived barriers and facilitators on a Likert scale. The knowledge and attitude on prescribing for older adults will also be obtained using validated Likert questions, although the attitude and behavior in clinician communication with pharmacists will be gathered using the Home Medicines Review Inventory (HMRI).

The complex medication regimens of many older adults contribute to increased risk for drug interactions, ADRs, and other poor health outcomes. A number of studies have found that the use of PIMs is associated with adverse outcomes, such as falls, acute care use, and other negative outcomes [49,50]. Improving medication prescribing using START/STOPP criteria has been found to reduce adverse effects and improve patient outcomes in a number of studies, although mostly in institution-based settings (acute care hospitalization, nursing homes) [50]. Pharmacist-led interventions have been demonstrated to have beneficial effects [51,52] but often do not reach the majority of those who would most benefit from them. Leveraging the advances of telemedicine, clinicians can provide medication management services to older veterans at home. The televisit intervention with pharmacists in medication management has the potential to bridge the current gaps in older adult care and provide a scalable solution to improve medication use in older adults.

The study’s pilot phase refined the procedure for pharmacists to conduct home televisits. It is possible that veterans will consider televisits to home cumbersome or invasive and that VA primary care staff will consider tailoring medications for older adults an additional task burden that competes with other mandated tasks. Our prior experiences of home televisits suggest that they are well accepted, and our hybrid effectiveness implementation study will enable us to identify potential implementation challenges and thus can inform future implementation. To ensure we understand PCP decisions on pharmacist recommendations, we will track PCP adoption of recommendations provided by pharmacist reviews to identify PCP reasons for adoption or nonadoption of recommendations, as well as provider factors that can influence those decisions.

The older adult population has increased the use of technology, allowing the VA to use technological advances for VA televisits, overcoming previous barriers, and we expect future changes in technology will further facilitate use. Our study design will allow us to observe older adults using technology at home to enhance our understanding of the barriers to use, and qualitative interviews with clinicians and patients will further identify factors to guide future implementation. We have elected to use 2 separate sites in the RCT phase in order to enhance our understanding of contextual issues and external validity.

This study design includes both a pilot phase trial and a hybrid effectiveness implementation trial, which ensures that we would observe and be able to address challenges prior to the RCT, although the results of the hybrid implementation effectiveness trial would allow us to determine its effectiveness, while preparing us to spread the intervention, if effective, in the health system. We aim to limit crossover contamination for providers with cluster randomization. Lastly, the standardized intervention and structured pharmacist protocol will ensure consistency and reliability.

The study design also has a few limitations. One is that the study will be conducted in urban and suburban settings, which may limit generalizability Another limitation is that the pilot phase was conducted during the COVID-19 pandemic, which introduced unique challenges in recruitment due to factors such as travel restrictions and increased risk for participants; however, the study team tackled those challenges and completed the pilot phase of the study despite limitations introduced by COVID-19. This study will contribute toward the literature on pharmacist-led medication management through home televisit procedures, as well as interventions aimed at improving medication use in older veterans.

This study design will demonstrate how medication management will be beneficial through televisits. The implementation effectiveness trial will show how it can improve the use of medications and interventions related to the potential impact on older adults with multiple chronic conditions and polypharmacy. Clinicians can offer older veterans medication management services at home by using telemedicine’s advancements. The use of pharmacists in televisit medication management interventions has the potential to close existing gaps in older adult care and offer a scalable way to enhance older adults’ medication use.