Abstract
Keywords
Introduction
The rising complexity of care needs among patients receiving home care services, driven by population ageing, multimorbidity, and polypharmacy, demands a coordinated, multidisciplinary approach to ensure safe and effective medication management. Interprofessional collaboration, particularly involving pharmacists with prescribing privilege, has emerged as a strategy to optimise medication use, and improve clinical outcomes.1,2 In United States, Canada, United Kingdom and Ireland, pharmacist-led medication reviews and prescribing within interprofessional teams have led to improvements in medication appropriateness, reductions in drug-related problems (DRPs), and enhanced patient satisfaction.3,4 Pharmacist prescribers are prevalent in the healthcare systems of Canada, United States and the United Kingdom, where they function as collaborative prescribers in the Canada and United States,5,6 and as independent and supplementary prescribers in the United Kingdom.7,8
In Singapore, legislative changes in 2018 enabled accredited pharmacists to prescribe medications to patients under collaborative practice agreement (CPA) with physicians. 9 These agreements outline the specific conditions under which a pharmacist can prescribe and manage medications. Recent local initiatives, including collaborative prescribing practices in outpatient clinics, have shown promising results in enhancing chronic disease management and reducing DRPs.10,11 However, the role of prescribing pharmacists within the home care setting remains underexplored. While some home care teams may involve pharmacists in medication reconciliation or advisory roles, structured interprofessional collaboration with pharmacist prescribers is not yet systematically implemented or evaluated.
Changi General Hospital (CGH) piloted the integration of pharmacist collaborative prescribers as part of the interprofessional team in home care service since December 2021 and this initiative aims to reduce DRPs at home care setting. Pharmacists conduct home visits, provide medication reviews, collaborate with physicians to optimise medication and treatment plans, and coordinate with nurses to ensure patients receive comprehensive support including medication refills.
This study aims to evaluate the impact and feasibility of interprofessional collaboration involving prescribing pharmacists in the home care setting in Singapore. By generating local evidence, we hope to inform service design, CPAs and workforce planning to better support the integration of pharmacists in home care teams.
Method
Setting
In Singapore, public healthcare is delivered through a network of institutions under three integrated Regional Health Systems: SingHealth, National Healthcare Group, and National University Health System. These systems coordinate care across primary, acute, and community settings within their regions. Primary care is delivered through government-funded polyclinics and private general practitioners (GPs). The Medifund supports patients who require assistance with bills for inpatient, outpatient, and long-term care services. The Community Health Assist Scheme (CHAS) supports patients from a lower socioeconomic background, to receive government subsidies when seeking care from participating private GPs. This integrated model supports continuity of care across different settings.
This study focused on patients participating in the transitional care program, a 3-month healthcare initiative designed to support patients as they move from hospital to home. The primary goal of the program is to ensure continuity, reduce readmissions, and improve health outcomes by addressing the needs of patients during the transition. Care team included community nurses and/or transitional care physicians. Pharmacist collaborative prescribers have been involved since Dec 2021, to help improve medication use among those patients enrolled into the program.
CGH has a total of four pharmacist collaborative prescribers trained to conduct medication reviews in a home care setting. Pharmacists providing home-based medication reviews must have at least 3 years of patient care experience as a pharmacist. They are required to complete a nationwide Collaborative Prescribing Programme (CPP) training, developed by Singapore’s Ministry of Health (MOH) and National University of Singapore. This program is a 3-month, part-time course that includes 80 clinical hours and 12 full days of didactic lectures covering various clinical topics. Upon passing formative and summative assessments, including an Objective Structured Clinical Examination (OSCE) and a final written examination, pharmacists become accredited as Collaborative Prescribing Practitioners (CPPs). After registration with MOH as a CPP, they are required to complete an in-house clerkship of 30 home-based medication reviews under the direct supervision of a transitional care physician before they can provide the service independently.
The pharmacists have also demonstrated their competence at clinical prescribing practices, and have CPAs with multiple specialties, such as cardiology, endocrinology, renal medicine, chronic pain service, rheumatology and gout. Through the agreements, pharmacists can initiate, adjust, or deprescribe medications as appropriate, improving medication safety and efficacy.
Population
Patients over 21 years old who were enrolled in the transitional care program and met the criteria for polypharmacy, defined as the use of five or more long term medications during the assessment period from December 2021 to February 2023 were included in the study. Patients who were not enrolled in the transitional care program were excluded.
Study design
This was a retrospective matched cohort review of patient case notes and medication records from December 2021 to February 2023. This study was designed to evaluate the impact and feasibility of interprofessional collaboration with prescribing pharmacists as part of the home care team.
Effectiveness of the interprofessional collaboration in addressing DRPs was measured using the number of DRPs identified and resolved post index visit. All patients were assessed for 6 months. Safety of a pharmacist collaborative prescribing initiative in home care services was evaluated based on incidence of all-cause hospitalisation within 6 months post index visit comparing between intervention arm and matched control arm. All-cause hospitalisation is defined as any admission to the hospital for any reason during the study period.
Other outcomes included types of DRPs identified and resolved.
The DRPs were classified using a modified Hepler and Strand classification scheme, 12 with additional categories for “optimised drug therapy’, “expired drug”, “deprescribing”, “inadequate monitoring”, “inappropriate dosing” and “inappropriate drug use”. The “optimised drug therapy” category focused on improving a patient’s medication regimen, such as by adjusting dosages or adding another medication to achieve the intended therapeutic effect. The “expired drug” category addressed instances where a patient was using a medication past its expiration date. The “deprescribing” category involved systematically discontinuing medications that were no longer necessary or lack a clear therapeutic benefit. The “inadequate monitoring” category included pharmacist identified and recommended clinical or laboratory parameters to monitor the effectiveness and safety of a patient’s medication therapy. The “inappropriate dosing” referred to situations where the dose of a medication did not align with a patient’s needs or clinical guidelines, potentially leading to suboptimal therapeutic outcomes or harm. This could involve both over- and under-dosing, or failure to adjust the dose for renal impairment. The “inappropriate drug use” referred to situation where a medication was prescribed not aligning with patient’s clinical needs. This could include not prescribing a necessary drug when one was indicated for the patient’s condition or continuing a medication that had been held off.
The review panel, consisting of at least one physician and three pharmacist collaborative prescribers, reviewed the cases, confirmed that the appropriate DRPs were identified and categorised. The review panel was independent of the care team. They were not blinded during the evaluation of the DRPs.
Index home visit was defined as the first home visit with pharmacist involvement within the study period for the intervention arm. At index home visit, patients had their medication regimens assessed by a prescribing pharmacist and any DRPs identified would be resolved in collaboration with the physician. Index visit for cases in the control arm was the discharge date for patients whose discharge occurred within the same month as the enrolment of cases in the intervention arm. At index visit, cases in the control arm had their medication regimens assessed by dispensing pharmacists when they collected medications at the pharmacy on discharge.
Deceased patients at the point of data collection were not included into this study.
Ethical approval for this study was obtained from the SingHealth Centralised Institutional Review Board (CIRB ref: 2022/2596). Informed consent was waived based on ethical consideration.
Details of intervention
The transitional care program relies on strong interprofessional collaboration between physicians, pharmacists, and nurses to ensure safe, and effective medication use in home care setting. Each profession has a distinct yet interconnected role in identifying and resolving DRPs and improving patient outcomes.
Physicians serve as primary decision-makers in diagnosing conditions and initiating treatment plans, ensuring that prescribed medications align with the patient’s overall health needs. Primary team physicians do not conduct home visits, with the exception of the transitional care physicians who attend to selected complex cases. Nurses are essential in patient education, monitoring medication adherence, and providing continuity of care during home visits. Pharmacists, particularly those with prescribing privilege, contribute by conducting comprehensive medication reviews, identifying and resolving DRPs, optimising pharmacotherapy in collaboration with physicians and reducing the risk of adverse effects for enrolled patients.
Pharmacists work within CPAs that define their prescribing privileges and scope of practice. If a medication adjustment falls beyond their scope, pharmacists communicate recommendations to physicians through direct messaging or phone calls. Pharmacists document changes in the electronic medical records (EMR), ensuring that all healthcare providers across the system have updated information. Additionally, medication changes are updated in the National Electronic Health Records (NEHR) to prevent errors and improve future care. Robin et al. highlighted the benefits of electronic health record, noting that a medication list updated at each patient visit provided essential information about a patient’s past and current medications, as well as other relevant details needed for an effective treatment. 13 Through this team-based approach, the transitional care program ensures that physicians oversee and adjust treatment plans in response to pharmacist recommendations. Nurses monitor patients, reinforce education, and escalate concerns for timely intervention.
Nurses refer patients to prescribing pharmacists when they suspect DRPs. The collaborative prescribing pharmacist then conducts a home visit to assess the patient’s condition. Additionally, the pharmacist reviews the transitional care program’s patient list to identify individuals with polypharmacy. When multiple cases of polypharmacy are detected, priority is given to patients from lower socioeconomic backgrounds, such as those receiving MediFund assistance or with a documented history of poor medication adherence. This prioritisation is based on findings by Sarah et al. and Ekenberg et al., who highlighted a correlation between lower socioeconomic status and medication non-adherence.14,15
Before the visit, the pharmacist reviews the patient’s clinical history and medical records to reconcile the medication list. During the home visit, all medications present in the home are physically verified by the pharmacist. Interventions may target both acute and chronic conditions. When necessary, the pharmacist conducts a physical examination and collaborates with physicians to ensure timely management. For patients with unresolved DRPs, follow-up visits or phone calls are arranged. These follow-ups typically occur within 1 month and continue until the DRPs are resolved.
Patients in the control group were those enrolled in a conventional home care service but were neither identified nor referred to the prescribing pharmacist for a home visit. These patients received a standard medication review from dispensing pharmacists at the pharmacy when they collected their discharge medications. They receive conventional home care service provided by nurses and/or transitional care physician, but did not receive a comprehensive medication review by a prescribing pharmacist in the home care setting. Patients in the control arm were selected using the propensity score matching method.
Data collection
Data on patient demographics and clinical characteristics were obtained from CGH electronic medical records and inpatient discharge summaries. Dispensing pharmacists identified and resolved DRPs for patients in both the intervention and control arms at the point of discharge. In contrast, collaborative prescribing pharmacists were involved in identifying and resolving DRPs only in the intervention arm during home visits. DRPs identified by dispensing pharmacists were recorded using the Medication Intervention Note, while those identified by collaborative prescribing pharmacists were documented in the Pharmacist Consult Note. This database was integrated with the electronic medical records, allowing pharmacists to track DRPs and monitor their resolution. The review panel assessed DRPs documentation generated from this database for further analysis.
Statistical analysis
We conducted nearest-neighbour propensity score matching with a target ratio of 1:5 using age, sex, race, religion, Charlson Comorbidity Index (CCI), and prior 6-month hospitalisation as matching variables. As nearest-neighbour matching is probability-based, not all intervention patients were matched to exactly five unique controls, and some control patients were matched more than once.
Descriptive statistics of patient demographic and clinical characteristics were reported as number and percent for nominal data, mean ± standard deviation (SD) for normally distributed data, and median and interquartile range (IQR) for non-normally distributed data.
To assess hospitalisation outcomes, we used mixed-effects negative binomial regression. This model was selected because hospitalisation counts were overdispersed and not normally distributed. The mixed-effects framework allowed for repeated measures within patients across pre- and post-6-month periods. Covariates included in the adjusted model were age, gender, race, and religion. Crude and adjusted incidence rate ratios (IRRs) with 95% confidence intervals (CIs) and exact p-values were reported.” All statistical analyses were conducted using Stata 17 (College Station, TX: StataCorp LLC).
There were no missing values for any of the variables included in the analysis.
Results
Demographic and clinical characteristics
A total of 70 patients were enrolled in the intervention group, which received interprofessional care with collaborative prescribing pharmacist as part of the team. The control group consisted of 981 patients who received conventional home care service without a prescribing pharmacist. The resulting matched cohort included 69 intervention patients and 282 controls. One patient from the intervention group did not have suitable match and was excluded from the analysis. Although a 1:5 ratio was specified, some intervention cases had fewer than five unique matched controls, and certain controls were matched to more than one case. This reflects the nature of nearest-neighbour matching (Figures 1 and 2). A flowchart for the propensity score matching analysis. Overview of control group (conventional home care service) and intervention group (with newly added collaborative prescribing pharmacist as part of the interprofessional home care team).
Demographic and clinical characteristics of propensity score-matched patients enrolled to study.
Abbreviation: SD: Standard Deviation, IQR: Interquartile Range, %: Percentage.
p-value: statistical significance level of the difference between groups.
Charlson Comorbidity Index: A measure of comorbidity that predicts the 10-year mortality for a patient who may have a range of health conditions.
*Mortality reflects deaths occurring after 6 months of follow-up. Patients who died within 6 months were excluded from the analytic cohort.
Outcomes
As shown in Table 1, The intervention group had significantly more DRPs identified and resolved (214 vs 18) compared to the control group. All DRPs identified by the prescribing pharmacist in the intervention group were resolved before the patients were discharged from the transitional care program. The DRPs identified by dispensing pharmacists at discharge in both control and intervention group were resolved at medications provision to patients. A DRP was defined as resolved if it was communicated to the primary team physician and/or the medication list was updated by physician or pharmacist, and patient no longer required further follow up.
Number and types of drug-related problems identified and resolved.
Mixed-effect negative binomial regression model of all-cause hospitalisations pre- and post-sixth month index visit on groups.
IRR = Incidence Rate Ratio; CI = Confidence Interval. *Adjusted for age, gender, race, and religion. Patients who died within 6 months were excluded from the analysis.
Values are marginal effects (change in log hospitalisation rate) estimated from mixed-effects negative binomial regression models with patient-level random effects. Crude models are unadjusted; adjusted models include age, sex, Charlson comorbidity index, and baseline hospitalisations as covariates. “ref” = reference category (pre-6 months). Negative values indicate a reduction in hospitalisation rates compared to the pre-6 month period. P-values correspond to between-group differences at each time point.
Selected details on interprofessional collaboration with prescribing pharmacist as part of the home care team to resolve drug-related problems.
Discussion
This study found that interprofessional collaboration with pharmacist as part of the home care team was associated with significantly more DRPs being identified and resolved compared to the control group (214 vs 18). While this outcome may seem intuitive, as an addition of prescribing pharmacist to the interprofessional team can more effectively detect and address DRPs, the findings offer important insights into the pharmacy practice redesign to better support the integration into home care teams. Early post-discharge review by prescribing pharmacists can help detect clinical deterioration at an early stage. As part of a team-based care approach, pharmacists do physical examination as needed and regularly update physicians on clinical parameters, recommend treatment plans, and titrate medication doses. Physicians review these care plans, trusting pharmacists to manage follow-up care. Nurses reinforce medication adherence, educate patients on proper medication use, and coordinate care by ensuring that pharmacist and physician recommendations are understood and followed. These findings align with Helgesen et al. who showed that interprofessional collaboration between nurses, pharmacists and physician contributed to increase quality and safe pharmaceutical care in community health services. 16
Among the different categories of DRPs, non-adherence to medications was the most commonly identified DRPs in the intervention group at home care setting, accounting for 43% of cases. This was aligned with the findings by Cheen et al. which identified non-adherence as the most common DRP among care transition program patients in Singapore. 17 Home visits allowed pharmacists to assess patients’ living conditions and identify potential issues. The home environment can significantly influence patient behaviours. 18 During home visits, pharmacists were able to review all medications based on the patient’s physical stock and facilitate the establishment of a structured daily routine to reinforce adherence, an approach not typically feasible in hospital settings.18–20
In contrast, DRPs in the control group were identified solely by dispensing pharmacists during routine medication provision, limiting their ability to observe patient behaviour or verify existing physical medication stock. This is particularly important, as non-adherence is responsible for a significant proportion of avoidable hospitalisations and treatment failures. 21 In the control group, the most prevalent issue was inappropriate dosing, occurring in 50% of cases, followed by inappropriate drug use at 22%.
Importantly, pharmacist collaborative prescribing was safe, with no increase in hospitalisations over a 6-month follow-up. This extended timeframe, chosen over the standard 30-day readmission window, allowed for a more accurate assessment of the service’s safety, as the effects of pharmacotherapy optimisation typically emerge over a longer duration. Pharmacists worked under CPAs, which outlined specific conditions, medications, and treatment guidelines they could manage independently. However, for complex cases or significant medication changes, pharmacists consulted physicians, who reviewed and approved recommendations before implementation. These findings supported that, with an appropriate framework in place, pharmacist collaborative prescribing was a safe model of care.
Case examples further illustrate the added value of pharmacist interventions in the home care setting, including improvements in clinical indicators and the prevention of medication errors. For instance, in one case, a pharmacist identified that a patient with diabetic neuropathy was experiencing hallucinations likely due to pregabalin and, after consulting the relevant specialist, advised discontinuation. In another case, a patient on quetiapine experienced significant low blood pressure, prompting the pharmacist to work with the physician to discontinue it. Such timely interventions helped prevent potential adverse outcomes and underscore the importance of interprofessional collaboration in home care setting. Selected details on interprofessional collaboration in resolving DRPs in home care setting were summarised in Table 4. Research by Supper et al. supports this model, showing that interprofessional collaboration, adapted to the setting and stakeholders, improved patient outcome. 22
Integrating prescribing pharmacists into interprofessional home care teams in Singapore could provide a scalable approach to managing DRPs. A phased implementation, beginning with higher-risk patient groups, such as those in transitional care program, before wider implementation across the community care settings like community health post, may be most feasible. Pharmacists possess a comparative advantage in DRP management, extending beyond adherence and supply issues to include medication optimisation, deprescribing, dose adjustment based on renal function, and resolving drug interactions. However, integration must contend with manpower, workflow, and resource constraints. Currently, no dedicated funded headcount exists for prescribing pharmacists in home or community care, and the current screening model by prescribing pharmacist is labour-intensive. Implementing a risk-triaging algorithm within electronic health records could automate patient selection and referral, ensuring pharmacist expertise is directed to those most likely to benefit. Successful scaling will require evidence of clinical benefit, alignment with health system priorities, sustainable funding models, and operational enablers to streamline workflows.
This pilot study has several limitations. First, the sample size was small due to the limited number of patients enrolled in the new model of care. To improve reliability and reduce potential confounding, we used a larger control group for comparison. Second, patients in the intervention arm were selected by the collaborative prescribing pharmacist using predefined criteria to target those at higher risk of DRPs. While this approach aligns with clinical priorities, it introduces selection bias when interpreting the number of DRPs identified and resolved. Third, the inconclusive p-value for the safety outcome may reflect limited statistical power, as no a priori sample size calculation was performed; null findings should therefore be interpreted with caution. Fourth, the review panel was not blinded to group allocation when categorising DRPs, which may have introduced bias. Fifth, the observed impact of interprofessional collaboration may have been facilitated by strong, pre-existing professional relationships within the team, an advantage not guaranteed in other care settings where trust and collaboration may need time to develop.
This study was also conducted within a single transitional care programme in one public hospital, limiting generalisability. As a pilot, the evaluation focused solely on DRP identification, without assessing broader economic, clinical, or humanistic outcomes. System-level measures, such as cost-effectiveness, long-term patient outcomes, and resource utilisation were beyond the scope of this study. Future research should involve larger, multi-centre datasets and diverse community settings, such as community health posts, to refine the interprofessional collaboration model. In addition, economic evaluations and patient-reported outcomes should be included to generate more comprehensive evidence. Such data will be essential to assess the added value and long-term sustainability of embedding prescribing pharmacists across various care settings within population health initiatives. This will provide guidance for health system stakeholders about scaling these models across the Regional Health System.
Conclusion
This study highlights the importance of interprofessional collaboration in home care, demonstrating how pharmacists, nurses, and physicians work together to optimise patient care. By integrating pharmacists as collaborative prescribers, the healthcare team was able to identify and resolve significantly more DRPs. Pharmacist collaborative prescribing was safe, with no increase in hospitalisation rates. Scaling this model will require a phased approach, beginning with higher-risk patient groups, coupled with strategies to address manpower, workflow, and funding challenges. With these foundations in place, prescribing pharmacists could make a significant contribution to medication safety and therapy optimisation within population health. Full regression results including covariates are presented in Supplementary Table 1 (Table S1).
Supplemental Material
Supplemental Material - Pilot study of interprofessional collaboration with prescribing pharmacists to address drug-related problems in patients receiving home care services in Singapore
Supplemental Material for Pilot study of interprofessional collaboration with prescribing pharmacists to address drug-related problems in patients receiving home care services in Singapore by Elena Meishan Lee, Dewi Azura Binte Hussain, Cheong Kah Mun, Agnes Lim An Qi, Lee Sueh Ying Patricia, Mon Hnin Tun, Kamsiah Binte Abdul Rahman, Bee Geok Low, Mok Wen Hui Nicole, See Qin Yong in Proceedings of Singapore Healthcare
Footnotes
Acknowledgements
ORCID iDs
Author contributions
Funding
Declaration of conflicting interests
Supplemental Material
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
