Implementation of a comprehensive template to support personalised care for people with multiple long-term conditions: a mixed-methods evaluation in primary care

Background 

Healthcare services are mainly organised around single health conditions and need reconfiguration to meet the needs of people with multiple long-term conditions (multimorbidity). Typically, people are offered annual reviews for each of their long-term conditions separately. In a randomised controlled trial, a comprehensive computerised template based on a personalised care model increased the person-centredness of multimorbidity reviews in primary care, but there were implementation challenges. We sought to understand and address the challenges of implementing a template to support personalised primary care for people with multimorbidity (PP4M).

Objectives To explore the extent of implementation and factors influencing uptake of the PP4M intervention. To understand factors influencing implementation and normalisation of the template.

Design 

Convergent parallel mixed methods within a non-randomised hybrid implementation-effectiveness study. Normalisation Process Theory (NPT) informed design, data collection and analysis.

Setting 

Primary care (general practices) in three English regions.

Participants 

Quantitative: Patients aged 18 years or over and had at least three types of long-term conditions (routine data collection); staff involved in using the template in implementation practices (Normalisation MeAsure Development (NoMAD) questionnaire).

Qualitative: Staff at implementation practices.

Intervention 

A multimorbidity computerised template to support personalised annual reviews. NPT-informed implementation package delivered to implementation practices included: process mapping, software support and training.

Data collection 

Routine medical record data; NoMAD questionnaires and qualitative interviews in implementation practices.

Primary/secondary outcomes 

Measures of reach, fidelity, acceptability and sustainability.

Analysis 

Quantitative data: descriptive statistics, logistic regression and difference-in-difference models. Qualitative data analysis conducted using NPT coding manual.

Results 

In practices that received an NPT-informed implementation package, use of the template increased more, across patients with a range of demographics and health conditions, than in those that did not receive the implementation package (OR 2.86 (95% CI 2.34 to 3.49)). The implementation package successfully triggered NPT processes of coherence and cognitive participation, and, to a lesser extent, collective action and reflexive monitoring. Contextual factors, including a lack of staff generalist skills and disease-specific incentives, impeded engagement and sustained implementation.

Conclusions 

Focusing on the processes of normalisation as mechanisms of implementation facilitated development of an implementation strategy with potential to trigger those mechanisms, but did not sufficiently address contextual factors. Implementation strategies to support personalised care must consider wider system and practice level contextual factors, such as incentives and staff training.

Trail registration number https://doi.org/10.1186/ISRCTN40295449 (2022–08-03, retrospectively registered.)