Assessing the impact of automated devices for enhancing health and reducing medication errors in older adults with long-term health conditions (ADHERE): a study protocol for a randomised controlled trial in Bedfordshire, Luton and Milton Keynes, UK

Abstract

Introduction

Medication non-adherence in older adults with long-term conditions contributes to significant morbidity, mortality and healthcare costs. While adherence support tools exist, many interventions fail to reach those most at risk. Automated medication dispensers (AMDs) show promise in improving adherence and health outcomes, but their integration into routine community pharmacy practice remains underexplored. This study aims to assess the effectiveness of an AMD intervention with SMS reminders in enhancing medication adherence among older adults and to evaluate how this technology can be integrated into community pharmacy workflows.

Methods and analysis

This randomised controlled trial involves 144 participants recruited from eight community pharmacies who will be randomised to receive either the AMD intervention or usual care. Primary outcomes include medication adherence, measured through pharmacy records and self-report at baseline, 3 and 6 months. Secondary outcomes include Morisky Medication Adherence Scale, health-related quality of life (SF-12), and healthcare resource use. A nested mixed methods process evaluation will explore uptake, acceptability and implementation.

Ethics and dissemination

The study protocol has been approved by the University of Bedfordshire Institute for Health Research Ethics Committee (IHREC1039), the NHS and the local authority Research Governance and Research Ethics Committee (NHS REC reference: 25/EE/0026). The findings will be disseminated via a final report, peer-reviewed journal publications and presentations at relevant conferences.

Trial registration number

ISRCTN18849739.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• This randomised controlled trial evaluates the use of automatic medication dispensers to improve medication adherence in older adults with long-term conditions.

• A mixed methods process evaluation is embedded to assess feasibility, acceptability and contextual factors influencing delivery and engagement.

• Recruitment through community pharmacies enables evaluation of real-world implementation within routine care.

• The study population is limited to older adults with long-term conditions, which may affect generalisability.

• Lack of participant blinding may introduce bias in self-reported adherence and well-being outcomes.

Introduction

Medication non-adherence (ie, not taking medication as directed by healthcare providers)¹ is common, critical and costly.² Around 50%–75% of people aged 65+ do not take medication as prescribed,3,6 often because they simply forget.47,9 Poor adherence produces significant health losses¹⁰ through inadequate symptom management,⁷ for example, patients with Parkinson’s have diminished mobility, increased fluctuations² and more hospital admissions.11,13 Cost implications are vast; conservative estimates of unused medicines in the NHS exceed £100 million annually.^{14 15} Non-adherence is associated with two-fold increases in inpatient costs¹⁶ and accounts for 23% of nursing home and 8%–11% of hospitalisation admissions^{12 17 18} and unnecessary doctor visits, diagnostic tests and treatments.¹⁹

Addressing medication non-adherence requires a particular focus on unintentional causes, which are especially common among older adults with long-term conditions. These include memory-related factors such as forgetfulness or confusion,²⁰ as well as physical challenges like impaired vision, reduced dexterity or difficulty opening packaging.²¹ Such barriers can significantly affect an individual’s ability to take medication as prescribed, even when motivation and understanding are present. Global clinical guidelines recommend person-centred approaches that enhance communication, simplify medication regimens and provide practical, tailored support.^{22 23} Strategies that account for cognitive and physical limitations such as reminder systems, accessible packaging and assistive technologies are essential for improving medication adherence and supporting independence.²⁴ Reminder systems not only help mitigate forgetfulness and physical barriers but can also promote accountability and self-monitoring, which are important in encouraging consistent medication-taking behaviours over time.²⁵

However, in many health systems, including the NHS, support for medication adherence remains fragmented and inconsistently delivered, particularly during transitions of care, such as hospital discharge into the community.^{26 27} These gaps are especially concerning for older adults with long-term conditions who often manage complex medication regimens without consistent follow-up or tailored support. Poor coordination across services and limited access to practical tools can lead to medication errors, reduced adherence and avoidable hospital readmissions. Addressing these challenges aligns with key NHS policy priorities, including the drive towards integrated care, personalised support for self-management and reducing pressure on urgent and emergency services. This study evaluates the use of Automated Medication Devices (AMDs) as a scalable intervention to support adherence, promote independence and improve continuity of care across health and social care settings.

Reminder systems and adherence aids such as pill boxes, blister packs and medication charts are widely used to support older adults in managing complex medication regimens. While these low-tech interventions can improve adherence for some individuals, evidence of their overall effectiveness remains mixed, with challenges around correct use, accessibility and the ability to support more complex regimens.^{28 29} In response, technological innovations such as AMDs are gaining attention. Emerging evidence suggests that AMDs may outperform traditional aids by improving dose accuracy, enhancing adherence, supporting patient autonomy and reducing caregiver burden.²⁵ Systematic reviews highlight the importance of user-centred design and usability, particularly for older adults and those taking multiple medications.³⁰ However, further high-quality research is needed to assess these technologies’ real-world effectiveness, acceptability and sustainability within routine healthcare systems.²⁵

A significant gap in the literature concerns the role of community pharmacies in supporting the use of technologies such as AMDs. Pharmacists are highly accessible and play a key role in supporting medication adherence through reviews, education and compliance aids.²⁸ UK policy, including the NHS Long Term Plan and the Community Pharmacy Contractual Framework, highlights their growing role in personalised care and medicines optimisation.^{22 31} AMDs could streamline adherence support by combining reminders, secure storage and time-locked dispensing. However, there is limited evidence on how these devices can be integrated into pharmacy workflows or how pharmacists perceive their role in delivering and monitoring such interventions. Understanding these factors is essential for developing scalable and sustainable adherence solutions in primary care.

Health policy and regulatory bodies are increasingly prioritising medication adherence, particularly for ageing populations.²³ Local initiatives using automated reminders have shown promising results in improving adherence and reducing healthcare utilisation, though implementation has been inconsistent.²⁵ Many devices rely on self-filling by patients or caregivers, increasing the risk of error. This study aims to evaluate the feasibility, acceptability and potential impact of an AMD used either alone or in combination with SMS reminders, depending on whether participants have family members and/or carers able to receive the messages. This dual approach reflects real-world variability in support networks among older adults and aims to explore how AMDs can best be implemented in community settings.

Randomised controlled trials (RCTs) are widely regarded as the gold standard for assessing healthcare interventions.³² Despite the growing interest in digital health solutions, no RCTs to date have evaluated the use of AMDs either alone or combined with reminders in a community pharmacy-led model of care. This trial addresses a crucial evidence gap and aims to generate findings relevant to healthcare systems worldwide that are seeking accessible, cost-effective strategies to improve medication adherence and promote healthy ageing.

A systematic review on pill and medication dispensers highlighted the importance of user acceptance and a human-centred approach for improving medication adherence and patient outcomes. The review categorised devices based on their development stage and hardware/software architecture, emphasising the need for further research to refine these tools for healthcare delivery.³³ Another review focused on medication adherence in elderly patients with polypharmacy, particularly those living at home. It found limited evidence and concluded that more studies are needed to understand adherence rates and evaluate interventions for this population.³⁴

Research questions

Are AMDs for adults aged over 50 with long-term health conditions more effective than standard care in improving medication adherence, quality of life (QoL) and health status after 6 months? Which groups does it work best for, and why?

Methods and analysis

Trial design

An individual randomised control study, participants will be randomly allocated to either a 6-month phase of treatment with AMD or care as usual. Assessments will take place at (1) Baseline, (2) 3 months throughout treatment (12 weeks) and (3) Post-treatment (24 weeks). A nested process evaluation following Medical Research Council (MRC) guidance will combine trial process measures and qualitative data to explore acceptability, implementation, impact mechanisms and context. This manuscript adheres to the Standard Protocol Items: Recommendations for Interventional Trials guidelines for the reporting of trial protocols (see online supplemental file S1). The future publication of study findings will follow the Consolidated Standards of Reporting Trials guidelines for reporting completed RCTs.³⁵

The intervention

The AMD device used in this study is the Pivotell Advance GSM (see figure 1). This AMD is commercially available and can be easily filled and programmed online to dispense pills up to 24 times a day. At the pre-programmed times, the dispenser rotates, the alarm signal is heard, and the correct dosage comes into view through the opening in the lid. The alarm can be programmed to ring for up to 5 hours. A self-locking shutter over the lid opening prevents access to the medication tray except at the alarm time, which prevents the possibility of double dosing or the user’s forced movement of the medication tray.

Figure 1. Photograph of Pivotell automated medication device used in study.

An all-inclusive SIM card is supplied with the Pivotell GSM dispenser, which allows automated SMS alert messages to be sent to designated family members, carers or the user (where relevant) if a medication dose has or has not been dispensed from the device at the scheduled time. This enables timely checks on the individual’s well-being before a potentially adverse event (AE) occurs.

Study setting

This study will be conducted in pharmacy practices in Bedford Borough, Central Bedfordshire, Luton and Milton Keynes (BLMK). This diverse region, home to almost one million people, is located in the southeast of England and is characterised by significant health disparities, making it an ideal context for this study.³⁶ Luton is the most urban, most deprived and ethnically diverse,³⁷ with four areas in the top 10 per cent most deprived in the country.³⁸ Bedford Borough and Milton Keynes also have significant ethnic minority groups and pockets of deprivation. Central Bedfordshire is the most rural and least diverse of the four areas, although it experiences pockets of deprivation. This region has a rapidly ageing population, with one in seven residents aged 65 or older and a high prevalence of long-term conditions, particularly among underserved and ethnically diverse communities.³⁶ These factors contribute to substantial health inequalities and pose challenges for healthcare services in effectively managing care. There is, therefore, an urgent need for more targeted interventions to improve health outcomes in this population.

Participants

Eligibility to take part will be ascertained using the following criteria.

Inclusion criteria

• Prescribed a stable daily medication regime (no more than four doses per day).

• Have physical and/or memory-related difficulties with medication adherence and/or report low adherence using the Morisky Medication Adherence Scale (MMAS-8) (<8).²⁹

• And/or have unsuccessfully tried other compliance aids (eg, dosette boxes, blister packs).

• Demonstrate decisional capacity to consent using questions informed by the University of California Brief Assessment of Capacity to Consent.^{33 34}

• Have sufficient English language skills.

• Provide informed consent.

Medication suitability:

• The AMD must accommodate the required medication (excluding creams, liquids or unsuitable forms), which can be verified via the NHS stability tool.

Exclusion criteria

Participants will be excluded if they:

• Use cytotoxic medications or medications unsuitable for the AMD (eg, size, storage).

• Have a variable medication regime (eg, Warfarin) or take medication more than four times daily.

• Already use an AMD or similar device.

• They are not responsible for taking their own medications (eg, the carer administers medication).

• Cannot provide informed consent.

Sample size calculation

The sample size focuses on detecting a difference in medication adherence between intervention and control groups (1:1 ratio). With non-adherence rates among adults aged 60+ ranging from 26% to 59%,^{39 40} a pilot study showed significant improvement using the device, achieving a 97.9% adherence rate, a 20% increase.⁴¹ Previous studies suggest 60 participants per arm can detect a clinically significant change.⁴² To ensure 90% power (α=0.05) and account for 20% loss to follow-up, 144 patients (72 per arm) are required.

Recruitment will involve eight pharmacies across BLMK over a 12-month period. Each pharmacy will recruit at least three patients per month for 6 months (totalling 18 patients per pharmacy). Based on inclusion/exclusion criteria, an estimated 50–100 participants per pharmacy per year can be recruited, ensuring the target is achievable.

Randomisation and blinding

Participants who are screened and provide consent will be randomised by the study statistician into either the treatment arm or care-as-usual. To address the risk of unbalanced trial arms in studies with fewer than 200 participants, minimisation, a form of covariate adaptive randomisation, will be used instead of pure randomisation.⁴³ This method reduces imbalance by sequentially allocating participants based on key covariates: gender (Male vs Female), age (50–65 vs 65–75 vs 75+), referral reason (Physical vs Memory-related) and pharmacy location. The process will be managed using the MinimPy2 software.⁴³ The trial statistician will remain blinded to group allocation throughout the analysis phase to reduce the potential for bias.

Figure 2 presents the flow chart of the trial, outlining the key stages from participant screening and consent through to randomisation, intervention delivery, follow-up, and data collection. It illustrates the pathway for both the intervention and control groups, providing a visual overview of the study design and timeline.

Figure 2. Flow diagram of the trial. AMD, automated medication device.

Outcome measures

Primary outcomes

Medication adherence is widely recognised as a complex concept; therefore, both objective and self-report measures will be used at three primary time points: baseline, 3 month and 6 months. Additional time points will be determined by prescription refill frequency each time the device is refilled for the intervention group and a repeat prescription is dispensed for the control group.

Objective adherence data will be captured through routine pharmacist administrative data and Medication Refill Adherence (MRA),⁴⁴ which expresses adherence as the percentage of total days of medication supplied divided by the number of days of study participation. MRA is a recommended and robust measure of adherence using administrative data.⁴⁴

Self-reported adherence will be assessed through structured open-ended questions, a method shown to be reliable, particularly in older populations,³⁹ particularly in older populations.⁴⁰ Patients will be asked questions such as ‘How many medications do you take each day and what is your usual routine?’ and ‘Thinking back over the last 28 days, how many times have you not taken your medication?”. This will be calculated as a percentage (ie, number of tablets taken per day reported by patient/prescribed number of tablets per day on medication label).⁴⁰ For the intervention group, additional adherence data will be collected from the Pivotell Administration Centre, which provides detailed records of medication access. Pharmacy refill logs will further validate missed medication data.

Secondary outcomes

Secondary outcomes will include self-reported adherence, QoL and self-reported resource use data. All outcomes will be assessed at all time points unless otherwise stated.

The MMAS-8 is a validated self-report tool widely used to assess medication adherence across diverse patient populations.⁴⁵ The scale consists of eight questions which measure intentional and unintentional barriers to adherence. It categorises adherence into three levels: high (score of 8), medium (score of 6 to <8) and low (score of <6). The MMAS-8 is recognised for its reliability and predictive validity, particularly in identifying unintentional non-adherence. Due to its simplicity and ease of use, the MMAS-8 has been extensively employed in clinical practice and research to capture adherence behaviours. Studies have demonstrated its effectiveness in predicting patient outcomes,^{46 47} making it a valuable tool in identifying individuals at risk of non-adherence and guiding targeted interventions to improve medication-taking behaviours.

QoL will be assessed using the 12-item Short-Form Health Survey (SF-12).⁴⁸ The SF-12 has been shown to be a reliable and valid tool for assessing mental and physical health status in general populations,⁴⁹ including older adults.⁵⁰ Example questions include: ‘In general, would you say your health is…’ measured by a five-point Likert scale ranging from (1) excellent to (5) poor. ‘During the past 4 weeks, how much has your physical health or emotional problems interfered with social activities?’ measured by a five-point Likert scale ranging from (1) all of the time to (5) none.

The inclusion of QoL as a secondary outcome is intended to explore whether improvements in medication management and adherence, supported by the AMD intervention, are associated with short-term changes in perceived independence, emotional well-being and social participation. Although significant changes in QoL are not expected over the study’s relatively short duration, examining these measures will provide preliminary insight into the broader acceptability and potential psychosocial benefits of the intervention, informing outcome selection for a future definitive trial.

Health resource data will be self-reported. Data will be gathered directly from participants (or caregivers if required) detailing hospital stays, healthcare visits and professional contacts. This approach will ensure a comprehensive understanding of healthcare utilisation and resource demands, enabling an evaluation of the intervention’s broader impact on patient care. Tracking resource use at multiple intervals will provide valuable insights into changes over time, informing future healthcare planning and resource allocation.

Adverse events and safeguarding

All eligible participants will undergo a comprehensive suitability check for both the patient and their prescribed medication, conducted jointly by a researcher and a pharmacist. The AMD has been used safely in UK pharmacies⁴¹ and complies fully with Royal Pharmaceutical Society guidelines.⁵¹ While the device itself is considered low-risk, concerns have been raised in the literature about the potential for increased iatrogenic events when adherence is improved.⁵² To mitigate this, participants will be monitored closely by their pharmacists, and General Practitioner (GPs) will be informed prior to enrolment and given the opportunity to raise any objections.

Adverse event monitoring and reporting

This study will follow the National Institute for Health and Care Research decision tree for AE and serious adverse event (SAE) reporting. Although no serious or unexpected AEs are anticipated, all events will be actively monitored and recorded. The Chief Investigator and trial team will assess the relatedness of events on a case-by-case basis. All AEs and SAEs will be reported to the Sponsor promptly and to the Study Steering Group (SSG) at scheduled intervals. Any SAE deemed both unexpected and related will be reported to the Research Ethics Committee within 15 days.

Criteria for study discontinuation

This is a low-risk trial, and most AEs in this population are expected to be unrelated (eg, hospitalisations, illness or death due to pre-existing conditions). However, if a serious adverse reaction that is at least possibly related to the AMD occurs in one participant or severe adverse reactions occur in two participants, the study will be paused for investigation. Based on findings, the study may be temporarily halted or discontinued, and appropriate authorities (eg, HRA) will be notified. The SSG will provide regular oversight to ensure participant safety.

Process evaluation

A mixed-methods process evaluation will be conducted, in accordance with the MRC guidance on evaluating complex interventions.⁵³ This evaluation will examine the acceptability of and engagement with AMDs, alongside contextual factors influencing implementation, delivery, participant experience and outcomes.

Data collection and analysis

Quantitative and qualitative data will be collected from all participants. In addition, enhanced data collection will be carried out at four purposively selected case study pharmacy sites, chosen to reflect ethnic, socioeconomic and geographical diversity across the BLMK region. Routinely collected data will be used to assess uptake, engagement and dropout rates. Baseline questionnaires completed by participants who decline the use of an AMD will be analysed to identify predictors of non-uptake.

A nested qualitative study will combine trial process measures with interview data to explore variability in quantitative outcomes.⁵³ All interviews will be audio-recorded, transcribed verbatim and analysed using framework analysis.^{54 55} Data synthesis will follow a realist evaluation approach to identify what works, for whom, in what contexts and with what outcomes.

Acceptability and satisfaction

Acceptability will be assessed using a theory-informed questionnaire based on the Theoretical Framework of Acceptability (TFA),⁵⁶ which explores seven constructs (eg, burden, effectiveness, self-efficacy). Follow-up qualitative interviews will be conducted with 5–10 participants (and families/carers) from case study sites to explore evolving views on intervention acceptability, perceived impact and overall experience. Postintervention interviews with 5–10 individuals who declined to use an AMD will explore barriers to engagement. In addition, pharmacists at the case study pharmacies will be interviewed to capture their perspectives on the acceptability, usability and implementation of AMDs in community pharmacy settings.

Case study sites and pharmacist involvement

Pharmacists at the case study sites will maintain recruitment logs, and 8–10 pharmacists will be interviewed to understand their perspectives on the feasibility and integration of AMDs into routine practice. These interviews will provide in-depth insights into how AMDs are adopted, used and perceived within everyday workflows, and identify barriers and facilitators to their implementation. Unstructured observations and focus group workshops with stakeholders, including patients, carers, pharmacists and commissioners, will explore the intervention’s perceived value, usability and sustainability. Unstructured observations will allow the research team to capture real-time interactions between users, pharmacists and the device, offering contextual understanding of practical challenges and adaptations. Focus group workshops will complement these data by enabling collective reflection among stakeholders, highlighting shared experiences, differing priorities and opportunities for improvement. Together, these qualitative methods will triangulate perspectives from multiple levels of the healthcare system, enhancing understanding of how AMDs can be embedded into routine community pharmacy practice and scaled across primary care settings.

Data collection

Data collected in this trial will include questionnaires, qualitative interview data and quantitative data on medication adherence and healthcare resource utilisation. Questionnaires, designed by the University of Bedfordshire, will be completed using Qualtrics. Interviews will be conducted face-to-face, recorded and transcribed verbatim by a professional transcription service. Medication data will be recorded on computers at participating pharmacies. Healthcare resource utilisation will be self-reported by patients during interviews and supplemented with data extracted from NHS records stored securely on university computers. Data relating to the AMDs will be stored on the Pivotell database. All collected data will remain anonymous and accessible only to authorised representatives from the university and participating pharmacies.

Efforts will be made to ensure data accuracy, completeness and reliability. Missing questionnaire data will prompt a researcher to follow-up with participants to retrieve the required information. Participants, including patients, caregivers and pharmacists, will be contacted approximately 2 weeks postintervention via email or phone to complete follow-up questionnaires. Rigorous procedures, including secure data storage and authorised access, will ensure the integrity and confidentiality of all collected data.

Data analysis

The data analysis for this study will focus on evaluating the feasibility, acceptability and potential impact of AMDs on medication adherence, QoL and healthcare resource utilisation. Quantitative and qualitative data will be analysed separately and then integrated during interpretation to explore how and why the intervention works within its implementation context.

All quantitative analyses will be conducted using IBM SPSS V.29⁵⁷ statistical software, and significance will be set at p<0.05. The trial statistician will remain blinded throughout the trial to reduce potential bias. Descriptive statistics will summarise participant characteristics, recruitment and retention rates. Missing data will be handled using multiple imputation techniques where appropriate, and sensitivity analyses will test the robustness of the findings.

Quantitative analysis

• Primary Outcomes (medication adherence): MRA will be calculated as the percentage of days medication was dispensed divided by the total number of study days. Depending on data distribution, paired t-tests or repeated measures analyses will compare adherence between groups over time.

• Secondary outcomes: self-reported adherence (MMAS-8) scores will be categorised as high, medium or low and analysed using χ² or logistic regression to explore group differences and changes over time. SF-12 data will be scored using standard algorithms to derive physical and mental health component scores; comparisons between groups and across time points will use paired t-tests or Analysis of covariance (ANCOVA) to adjust for baseline values. Healthcare utilisation data will be summarised descriptively and compared between groups using t-tests or non-parametric equivalents where assumptions are not met.

Qualitative analysis

Qualitative data will include pharmacist interviews, unstructured observations and stakeholder focus group workshops. These complementary methods will enable triangulation across multiple perspectives and contexts.

• Pharmacist interviews will provide in-depth insights into how AMDs are adopted, used and perceived within everyday pharmacy workflows, identifying barriers and facilitators to integration.

• Unstructured observations will capture real-time interactions between users, pharmacists and devices, offering contextual understanding of practical challenges, adaptations and workflow integration.

• Focus group workshops with patients, carers, pharmacists and commissioners will explore collective perceptions of the intervention’s value, usability and sustainability, highlighting shared experiences and differing priorities.

All qualitative data will be analysed thematically following Braun and Clarke’s six-step framework,⁵⁸ supported by NVivo software. Coding will be both inductive and deductive, guided by the TFA)⁵⁶ to explore constructs such as perceived burden, intervention coherence and self-efficacy. Themes will be refined through iterative team discussions, and triangulation across data sources will enhance analytical rigour. Integration of qualitative and quantitative findings will occur during interpretation, providing a comprehensive understanding of feasibility, acceptability and mechanisms of action to inform future large-scale evaluation.

Process evaluation data analysis

The process evaluation seeks to examine the intervention’s uptake, engagement and acceptability. Quantitative data from surveys will be analysed using descriptive statistics to report on the uptake and engagement rates. Regression analysis will be used to identify predictors of successful engagement, such as socio-demographic factors and baseline adherence.

Qualitative data from interviews and open-ended questions will be analysed using reflexive thematic analysis.⁵⁸ The interviews will be transcribed, coded and categorised to identify key themes regarding participants’ experiences with the intervention, barriers to engagement and factors that contributed to successful medication adherence. This analysis will provide deeper insights into the mechanisms behind the intervention’s effectiveness.

Patient and public involvement

A dedicated Public Advisory Group (PAG) was established early in the design phase to support the development and delivery of the study. The PAG includes older adults with long-term conditions, carers and individuals from local patient advocacy organisations and academic institutions with expertise in ageing and public involvement. This diversity ensures a broad range of lived experiences and perspectives are reflected throughout the research process.

The PAG has made significant contributions to the design of participant-facing materials, including the Participant Information Sheet, consent forms and questionnaires. They have also advised on the structure and language of the trial website to enhance its clarity and accessibility for a wide audience.

To assess the usability and acceptability of the intervention, PAG members participated in hands-on workshops to explore and test the AMD. These sessions allowed them to evaluate the physical interface, trial draft usage instructions and co-develop an accessible user guide that supports a range of literacy and digital skill levels. The group also assessed the potential burden of the intervention and contributed to designing a participant schedule that is realistic and minimally disruptive.

Looking ahead, the PAG will continue to shape the study’s dissemination strategy. They will help identify which findings should be prioritised for communication with participants and the wider public and advise on appropriate and accessible formats such as summary leaflets, community talks and updates via the study website to ensure results are shared in clear and meaningful ways.

Ethics and dissemination

The study protocol was approved by the University of Bedfordshire Institute for Health Research Ethics Committee (IHREC1039) on 26 November 2024 and the NHS and local authority Research Governance and Research Ethics Committee (NHS REC reference: 25/EE/0026) on 20 March 2025. The study will be conducted in accordance with the principles of the Declaration of Helsinki. Informed consent will be obtained from all participants prior to data collection. An example of the participant consent form is provided in the Supplementary Material. The findings will be disseminated through a final report, peer-reviewed journal publications and presentations at relevant conferences. The trial was registered on 31 March 2025 with registration number ISRCTN18849739.

Discussion

Medication non-adherence is a pervasive issue, especially among individuals with long-term conditions, leading to worsened health outcomes, increased healthcare costs and a reduced QoL. This study seeks to address this problem by evaluating the feasibility and potential effectiveness of AMDs in improving adherence. By comparing an intervention group using AMDs with a control group, the study is expected to provide preliminary evidence on the feasibility, acceptability and likely impact of these devices on medication adherence in real-world settings. The use of both objective adherence data from pharmacy records and self-reported measures from participants will allow for a comprehensive assessment of adherence, helping to establish whether AMDs can offer a viable and acceptable solution for improving patient compliance with prescribed medications.

The process evaluation will further enhance the findings by examining the mechanisms behind the feasibility and acceptability of the intervention. By assessing uptake, engagement and contextual influences, the study aims to generate insights into how and why AMDs work (or don’t work) for certain populations. This evaluation will identify barriers and facilitators to implementation, ensuring that the findings from the RCT are not just about the potential value of the intervention, but also about how it can be optimally integrated within community pharmacy practice.

QoL has been included as an exploratory secondary outcome to examine whether improvements in medication management and adherence are associated with short-term changes in participants’ perceived independence, emotional well-being and social participation. Although substantial changes in QoL are not anticipated within the relatively short duration of this feasibility trial, exploring this outcome will provide valuable preliminary data on the broader psychosocial relevance and acceptability of the intervention, helping to inform outcome selection and measurement strategies for a future definitive trial.

Future research should build on the evidence generated by this RCT, focusing on broader, multi-site trials to confirm the findings across diverse populations and healthcare settings. While this study is not designed to test effectiveness, it is expected to provide valuable data on the practicalities of implementing AMDs and on their potential contribution to improving medication adherence. Further research will be required to explore long-term outcomes, including impacts on patient health, healthcare costs and resource utilisation. The anticipated findings will serve as an important step towards understanding how adherence-enhancing technologies can be effectively embedded into community pharmacy services to support individuals with long-term conditions.

Limitations

A key limitation of this study is the retraction of the original MMAS-8 validation article: Morisky, D.E., Ang, A., Krousel-Wood, M., & Ward, H.J. (2008). Predictive Validity of a Medication Adherence Measure in an Outpatient Setting. The Journal of Clinical Hypertension, 10(5), 348–354. https://doi.org/10.1111/j.1751-7176.2008.07572.x. The article was retracted (https://onlinelibrary.wiley.com/doi/10.1111/jch.14718) nearly a decade after publication due to concerns about the reported sensitivity and specificity of the adherence scale. At the time of retraction, all co-authors were retired and no longer had access to the original data needed to formally respond to the statistical allegations. Moreover, the journal did not publish a side-by-side rebuttal or provide an open forum for scholarly exchange, limiting transparency around the process. While the retraction must be acknowledged, this context should be taken into account when interpreting the historical and scientific role of the original validation study.

Current trial status

Recruitment of participants started on 1 April 2025.