Assessment of fidelity to interventions aiming to improve appropriate use of polypharmacy in older people

Introduction

Studies demonstrating the effectiveness of an intervention could lack reliability if they do not involve an assessment of intervention fidelity. Fidelity evaluations findings can provide valuable insights for implementing interventions in clinical trials and in finding solutions for more successful interventions.[1] Despite the importance of this methodological feature, there has been a lack of focus on assessing aspects of fidelity to interventions designed to improve appropriate polypharmacy in older adults.[2]

Aim

To assess intervention fidelity in studies aimed at improving appropriateness of polypharmacy in older adults.

Methods

A paper-based assessment was conducted to evaluate intervention fidelity in 10 trials added to a Cochrane review which was being updated.[2] The interventions were assessed by the research team, comprising pharmacists experienced in health research and authors of the aforementioned review.[2] Data were extracted and assessed using the National Institutes of Health Behaviour Change Consortium (NIH-BCC).[1] This framework involves five domains: intervention design, provider training, intervention delivery, intervention receipt and intervention enactment. There are 39 elements across the five domains which were assessed as ‘present,’ ‘absent, but should be present,’ or ‘not applicable’. A scale ranging from zero (strongly disagree) to six (strongly agree) was used to rate fidelity. If an element was judged as ‘present’, it was rated between 1-6 based on the extent of the description of its presence. If an element was judged as ‘absent, but should be present’, this was scored as zero. If the element was graded as ‘not applicable’, it was excluded. The overall fidelity score for each study was calculated as the sum of scores for all elements, ranging between 0-234, with ≤116 equating to poor fidelity, 117-176 equating to fair fidelity, and 177-234 equating to good fidelity. Barriers and facilitators to conducting effective interventions reported in the included trials were compiled and combined into descriptive categories.

Results

Of the 10 included trials, none addressed all NIH-BCC domains. Intervention design contains 16 elements, and nine were reported in the trials such as content of treatment dose of ‘intervention’ and ‘comparison’ conditions. Of the seven elements that comprised provider training, three elements were addressed (e.g. how providers were trained). Six out of nine elements that encompassed delivery of intervention were presented (e.g. method to ensure the dose of intervention is delivered as specified). Of the five elements that comprised receipt of intervention, two elements were acknowledged including multicultural factors considered in development and delivery of intervention. The enactment domain was not considered in trials. Individual total study scores ranged between 19-72, with intervention fidelity for each study considered poor. Barriers (e.g. intervention complexity) and facilitators (e.g. prescribers’ experiences) to implement effective interventions were identified.

Conclusion

This study assessed intervention fidelity of trials aiming at improving appropriate polypharmacy in older adults. This was judged to be poor for each study. The number of included trials was small, and the scoring method was not validated. Future trials should consider addressing intervention fidelity comprehensively to help plan, implement and assess trials.

References

1. Borrelli B. The assessment, monitoring, and enhancement of treatment fidelity in public health clinical trials. Journal of Public Health Dentistry. 2011;71: S52-63.

2. Cole J, Gonçalves-Bradley D, Alqahtani M et al. Interventions to improve the appropriate use of polypharmacy for older people. Cochrane Database of Systematic Reviews. 2023;10(10): CD008165.