To our knowledge, this is the first study to conduct a multi-stakeholder approach to the development and prioritisation of research questions associated with medicines optimisation. Similar processes are used by the James Lind Alliance [25]. The current study adopted a systematic approach that incorporated the opinions of a wide range of stakeholders using participatory methods. The value of these stakeholders, rather than limiting the process to academic researchers and the literature, was illustrated by the additional 62 questions that were identified during Stage 2 as a result of including the extended stakeholder group, as well as the comparison of PPI and non-PPI priorities. Hence, our findings support the conclusions of an earlier review [26] which highlights the lack of utilising qualitative research on patient experiences of polypharmacy, especially where related to multimorbidity.
Several of the high priority questions reflected the extent to which patients’ views and experiences are considered during medication reviews. Few tools exist for eliciting patient priorities and preferences during consultations, including medication reviews [27]. As such, this service might benefit from the introduction of a common framework built on equity, confidence, and perceptions of acceptance [28], for defining and classifying patient-mediated interventions. Most of the questions associated with medication reviews related to ‘structural’ elements, i.e. what constitutes a ‘good’ review, and when is the ‘right’ time to undertake a review. The challenge for future work comes from operationalising these questions in objective, less value-laden terminology. This was also reflected with the prioritisation of questions regarding polypharmacy – for example, the nature of information and education that the patient would require to enable them to engage and have ownership of their medication management.
Polypharmacy is of increasing relevance due to the ageing population and the increasing number of people receiving multiple medicines [1]. The top priorities around polypharmacy identified in the present study reflect the challenge of how best to involve patients in decision making, especially in relation to medication reviews, and how primary care is best engineered to ensure their effectiveness.
In the UK, a range of non-medical health professionals can prescribe medicines for patient [29]. The growth in numbers of non-medical prescribers reflects the need to mitigate increasing demands on the NHS [30]. The ability to prescribe has the potential to enhance the roles of health care professionals and improve patient care by facilitating treatment provision in settings more accessible to patients and possibly in a more timely manner [31]. The top five research questions in the NMP category included the need to raise patient awareness of NMPs and the extent to which NMP training provides practitioners with the confidence to address complex polypharmacy and deprescribing issues [6]. It has been suggested that educational programmes for NMPs would benefit from considering how best to maintain the currency of practitioners’ knowledge [32], and that education aimed at the public may be warranted, to address concerns and limited awareness with regard to the diagnostic skills and status of NMPs [33].
The top-ranked ‘Deprescribing’ priority was “how to empower patients to take a more active role in self-management and self-monitoring of multiple long-term conditions, including deprescribing?”, which was were ranked ‘extremely important’ or ‘important’ by at least three-quarters of participants, indicating a convergence towards the desire for a broad culture shift, whereby patient perspectives are an integral part of the decision-making process. Challenges associated with deprescribing include how it is defined, whether it is safe, and how these aspects are communicated to patients and health care professionals alike [34]. This reflects a desire to adopt a holistic approach to medicines optimisation, whereby patient perspectives and values are central to the consultation process, teams are inherently multidisciplinary, and where the general perspectives as well as complexities of individual cases can be addressed [35-37].
The PPI research infrastructure has existed for over a decade [38] and the importance of PPI in this prioritisation exercise was reinforced by the many differences between their views and those of the non-PPI participants [39]. Given that patient involvement in healthcare decision-making can lead to better affective, cognitive and health outcomes [40, 41], the value of involving patients in the entire research process is implicit in the top-ranked ‘patient concerns’, ‘Is there a shared decision (with patients) about using each medicine?’.
The third WHO Global Patient Safety Challenge, Medication Without Harm, identified medication safety as a priority [42]. A separate consensus exercise identified the need to adopt technology to enhance medication safety, and to develop guidelines and standard operating procedures for high-risk patients, medications and contexts [43]. Within this current prioritisation study, it was seldom stated explicitly medication safety but was implicit in many questions. For example, ‘Patient concerns’ ‘What are the advantages and disadvantages of online pharmacy services in relation to access to medicines and safety, patient experience, out-of-pocket expenses, information provision?’ (rank 19), and ‘Deprescribing’ ‘Where are the gaps in education and safety (about deprescribing) and how these can be addressed is key to ensuring deprescribing is safe and effective?’ (rank 4). This observation may reflect the wording of the original NGT question and that the subsequent apportioning of questions across the four categories decreased the salience of ‘safety’. As such, priorities identified in the current study may differ substantially from global priorities where, for example, medicine safety is a recognised priority [44].
Strengths and Limitations
A major strength of this prioritisation study was the overall proportion of patients and other stakeholders involved in each stage, ensuring that different perspectives and experiences were represented throughout the process. This is likely to have improved the relevance and real-world value of the research outcomes, as well as the validity and reliability of the findings [45]. However, conversely, the high proportion of pharmacist participants may have distorted the results. The main limitation relates to the Stage 3 procedure of aggregating ratings into an overall rank. Ascribing a score to each rating and then summing the scores for a set of ratings by any one individual may produce the same magnitude of outcome from quite different sets of ratings. The Borda count employed in Stage 3 overcomes this to some degree in that it takes all the rating preferences into account by attributing weighted scores to each rating. However, this still assumes that scores are interval in nature and does not take into account the actual ‘attitude’ that the respondent may have towards the question content [46]. Nonetheless, to test whether different scoring approaches significantly altered the nature of rankings, different methods of analysis were explored. For example, given that ratings were skewed toward positive assessments, rankings were constructed based on aggregates of only ‘extremely important’ and ‘important’ ratings. While this did not affect overall ranks substantially, it was decided that a full Borda count that incorporated all ratings was more apposite in that it maintained the full range of perspectives from all participants.