I have just returned from the annual scientific meeting of the Society for Academic Primary Care (SAPC), which was held at Warwick University last week. The central concerns of the APOLLO-MM project - multimorbidity and polypharmacy – featured prominently in this year’s meeting. This may seem unsurprising given that care of patients with multimorbidity is very much the ‘norm’ for the generalist medical practitioner. But it was encouraging to see an increasing emphasis on how to account for the complexity of multimorbidity in research contexts, and there is certainly a growing army of researchers trying to tackle ‘inappropriate’ polypharmacy.
One thing that became clear as I moved from presentation to presentation about polypharmacy is just how ‘baggy’ some of our definitions are. Researchers use a range of cut-off points to define polypharmacy (4+, 5+, 10+, 15+ medicines). They do so whilst acknowledging that definitions based on numbers alone are crude and inadequate, as these measures conceal crucial issues such as: the poor alignment between numbers of drugs prescribed, dispensed and consumed; the contribution of non-prescribed medicines to the medicines count; and the relative complexity of different medicines regimens to the individual patient. The terms ‘appropriate’ and ‘inappropriate’ which are frequent qualifiers of the ‘polypharmacy’ concept are widely used but beg further questions such as ‘appropriate to whom?’ and “appropriate for what purpose or with what in mind?” and cause us to stop and wonder how fixed and certain are these notions of appropriateness.
On this topic, one of the prize-winning abstracts at the SAPC conference was the work of Jenni Burt
, Rupert Payne
and colleagues who are developing a measure of appropriate polypharmacy
in primary care. They have just completed a comprehensive systematic review (which involved screening a breath-taking number of papers!) to identify approaches that already exist to define and measure appropriate polypharmacy. They combined this review with a rigorous expert panel process through which they have refined their list of 160 possible measures to 12 areas. They hope to operationalise these areas within GP IT systems and it will be very interesting to see how this work develops as it moves forward and as its focus shifts from the generation of sound research evidence into the world of clinical practice.
On the subject of bringing technologies into practice, it was great to hear Tony Avery
reporting on the roll-out of the ‘PINCER’ intervention
in the East Midlands. PINCER is a pharmacy-led information technology intervention for medication errors in general practice which has been shown in previous research to reduce the rates of potential errors through a combination of a computerised search to identify ‘at risk’ patients and the support of an experienced pharmacist in helping to correct the problems.
According to the PRACtICe study
, prescribing errors arise in 5% of prescription items in general practice. These are mostly errors of mild to moderate severity, but one in 550 of these errors have the potential to cause severe harm. Given the scale of prescribing it is clearly a hugely important area in which to improve care. I was heartened (and frankly amazed) to discover that the PINCER intervention has already been implemented within 360 general practices
in the UK and has now searched 2.9 million patient records. The research team’s work so far offers early indications of success in reducing prescribing errors, which is very encouraging. It is well recognised that implementing expert technological systems in general practice can be extremely difficult and is often met with resistance
. However this team’s experience suggest that when the information technology is offering something that genuinely aligns with GPs’ professional concern to provide safe and high quality care, GPs are more than willing to innovate and seek to improve their practice (…and this intervention does also involve an expert pharmacist which I am sure increases its appeal!)
presented research on developing and validating a new 10-item scale (the Multimorbidity Treatment Burden Questionnaire
) to assess the burden of treatment in the context of multimorbidity research. Her study involved a review of qualitative literature and working closely with a patient and public involvement group to develop the scale, before testing it with a group of 1554 older patients with 3 or more long term conditions to assess its validity and reliability. One thing which surprised me was that when the researchers looked at the responses of patients of different ages they identified a lower treatment burden in older patients. Although the study was not designed to investigate the causes of this finding, it raises questions about whether and how older people adapt their lives to accommodate their treatment burden, or how they readjust their expectations (and how this then becomes reflected in their responses to a questionnaire, or their accounts of their experience). This is something that we hope we will gain some first-hand experience of as we conduct the APOLLO-MM study.
I was interested to hear Rupert Payne
present work on the patterns of polypharmacy amongst patients who have dementia
and co-existing medical conditions and who are approaching the end of their lives. In this retrospective study of the CPRD database
(Clinical Practice Research Datalink) – a database which covers 70% of the UK population – the research team looked at prescribing data pertaining to 6000 deceased patients. They found that within the last 12 months of these patients’ lives, polypharmacy remained substantial (about 5 drugs), included drugs intended for long-term prevention, and that there was only very modest evidence of reduction in polypharmacy at the extreme end of life. Notwithstanding the difficulties that clinicians may face in predicting the ‘end of life’, this work suggests that there may be scope for optimising medicines in this context. Of course notions like ‘shared decision-making’ around de-prescribing are of limited usefulness in the context that the person with dementia may not have capacity to share in decisions, and in this particular situation doctors may fear that an act of withdrawing medicines might be interpreted as withdrawing ‘care’.
On the subject of the complexity of caring for people with dementia, I was delighted to hear about the work of Nathan Davis
, winner of the Yvonne Carter Award for Outstanding New Researcher 2017. His qualitative work focuses specifically on managing the challenges and uncertainty in end-of-life care for people with dementia. He identifies the cessation of life-sustaining treatment as an area of particular complex decision-making. It is great to see that his published research
, which has incorporated a very strong commitment to co-design with practitioners and family carers, has resulted in the development of a toolkit of heuristics (rules-of thumb) which can aid healthcare professionals in making end-of-life care decisions for people with dementia. This toolkit has already been taken up by several NHS Trusts and is being incorporated into training programmes by the Alzheimer’s Society