MedBridge study

Medication Reviews Bridging Healthcare (MedBridge) is a research project that aims to study the effects of hospital-initiated comprehensive medication reviews, including post-discharge follow-up, on older patients' healthcare utilisation.

The project started in 2016 and the main results were published in April 2021 in the scientific journal JAMA Network Open. The project has been a fruitful collaboration between Region Uppsala, Region Gävleborg, Region Västmanland, Uppsala Clinical Research Center (UCR) and the Departments of Pharmacy and Medical Sciences at Uppsala University.

Information på svenska.

MedBridge forskar- och apotekspersonal.
MedBridge clinical pharmacists and researchers.

Background: Why MedBridge?

Inappropriate prescribing and use of medications are major causes of avoidable healthcare-related harm around the world. Medication review, a structured critical examination of a patient’s medication treatment, by a ward team including a clinical pharmacist has been introduced as an intervention in older hospitalised patients to optimise treatment outcomes and reduce harm. However, evidence on clinically relevant outcomes in secondary care is scarce, and implementation of these interventions in daily practice is challenging. The main aim of this project was therefore to study the implementation, performance and effects of medication reviews by clinical pharmacists in older hospitalised patients.

Main study result: No effect on unplanned hospital visits

A pragmatic cluster-randomised crossover trial (MedBridge) was conducted to study the effects of hospital-based comprehensive medication reviews (CMRs) including post-discharge follow-ups on older patients’ healthcare utilisation, compared with only hospital-based reviews and usual care. The trial was conducted at eight wards with multi-professional teams at four hospitals in Sweden in 2017-2018. Patients aged 65 years or older, admitted to one of the wards, were included in one of three treatment groups: 1, CMR; 2, CMR plus post-discharge follow-up; 3, usual hospital care. In total, 2637 participants were included in the main analysis of the MedBridge trial. The participants were 81 years old on average and the median number of prescribed medications was 9. The primary outcome measure, the incidence of unplanned hospital visits within 12 months, did not differ between the intervention groups and usual care. The results have been published in JAMA Network Open.

Visual abstract.jpg

Figure: Visual abstract from Kempen et al. JAMA Netw Open. 2021;4(4):e216303. doi:10.1001/jamanetworkopen.2021.6303

Process evaluation: medication reviews and collaboration between physicians and pharmacists is appreciated, but there is a lot of room for improvement

A process evaluation was conducted alongside the MedBridge trial to support an understanding of how the interventions were implemented and performed and which factors may have affected the trial's results. Interviews with patients and next-of-kin were conducted to explore their experiences with, and views on, the interventions in the trial. Analysis of these interviews resulted in seven key themes. In general, the experiences and views were positive, but some factors, like problems with receiving and retaining information, could have negatively impacted the effectiveness of the interventions in the MedBridge trial.

Interviews with physicians and pharmacists were conducted to explore the facilitators and barriers for performing the trial’s interventions. Multiple facilitators and barriers were identified and then grouped in six main themes. Examples of themes were: CMRs and follow-ups are needed, but not in all patients; roles and responsibilities are unclear; and personal contact at the ward is essential for physician-pharmacist collaboration.

Finally, the intervention fidelity and process outcomes within the MedBridge trial were assessed. In the CMR and CMR plus post-discharge follow-up groups, the intervention fidelity was 94% to 98% during hospital admission, and 40% to 81% upon and after discharge. On average, one medication discrepancy and two drug-related problems per patient were identified as process outcomes of the CMRs. In 77% of the patients, at least one medication discrepancy or drug-related problem was solved.

A new tool to identify medication related hospital admissions

A practical research tool to identify medication-related hospital admissions. One of the MedBridge trial’s secondary outcomes, was developed and validated. The tool was deemed valid for use to identify medication-related hospital admissions in older patients by final-year undergraduate and postgraduate pharmacy students.

In 2021, AT-HARM10 was updated and applicability and reliability to identify possible drug-related emergency department visits was confirmed. AT-HARM10 has received international attention and has already been used for research purposes in other regions in Sweden, Denmark, the Netherlands and Norway. More information on AT-HARM10 can be found here.

Conclusion and implications for research and practice

Despite a high percentage of patients with medication discrepancies and drug-related problems being solved, hospital-based CMRs with and without post-discharge follow-ups, as conducted in the MedBridge trial, do not decrease the incidence of unplanned hospital visits in older patients. Future research projects and clinical initiatives may benefit from a systematic approach addressing the factors identified through our process evaluation. For trials measuring medication-related hospital admissions, AT-HARM10 can be used as assessment tool. In 2021 and onwards, our research group will investigate the extent to which medication-related hospital visits could have been prevented in order to generate knowledge about how we can improve the health and quality of life of older patients through better use of medications in the future.

Media and publications

All publications, articles and conference proceedings can be found here.

Funding

This project has received research grants from different funds which are listed here.

Contact information

For contact information, visit this page.