Assessing quality in emergency health care

We as a family were recently the receivers of health care, with the birth of our firstborn son who arrived a little early. It is always a learning experience to be the receiver of health care, and it certainly focuses the mind as to what is important in its delivery.

We were fortunate and I am pleased to say that despite the urgent nature of the situation, both mother and baby are well. However for me — usually the consultant in charge — being admitted for emergency surgery was an eye-opening experience. I have nothing but admiration for Miss Wallace who expertly performed the caesarean section and safely brought both mother and baby through the whole experience. The entire undertaking set me thinking about quality assessment in health care and, in particular, the emergency setting.

The PROMs revolution has happened and is the next step in the evidence-based medicine movement focussing outcomes on what patients report. The difficulty of course is not only asking the correct questions, but contextualising the outcome measures to those most important to the patients. One approach has been to ask about satisfaction levels, the so-called ‘friends and family test’: “Would you recommend this treatment to friends and family?” A fantastic question on the face of it, but likely not applicable to the emergency setting.

During the preparation for the Warwick Hip Trauma Evaluation (WHiTE) study,¹ a series of qualitative methods research interviews were undertaken with relatives and carers of hip fracture patients, and the results were surprising. While most orthopaedic surgeons would define the success of hip fracture treatment in terms of hard endpoints like survival or even length of stay, patients themselves were not really concerned with longevity, but more the quality of their recovery. One interesting way to measure quality-adjusted life years (QALY) is an older method where patients are asked how many years of remaining life they would trade to achieve a resolution to their symptoms. The hip fracture patients in the Warwick, now Oxford, study were emphatic that they preferred quality of life with high participation over longer life. Hence, rather unusually, a quality of life measure was chosen in their study as the outcome measure for hip fracture surgery.

I have personal experience of similarly ‘counter-intuitive’ results from qualitative research in terms of addressing outcomes. I was surprised to find that uncertainty and fear for the future were common themes in patients with young hip fractures, but more pronounced in what they didn’t say than what they did. As such, a mental well-being score would be appropriate. That patients who are on an ITU, unconscious and suffering from rib fractures would not mind being randomised into an intervention study with - or indeed without - the agreement of their family, and that survival and return home were the most important outcome measures to them (previous studies had used ventilator days and complications).

All of this tells us that patients do not see success in the way that we, as clinicians, do. We cannot and should not make judgements without appropriate investigation first regarding which outcome measure or suite of measures is suitable for assessment of what is important to patients. This of course is entirely different to an attempted assessment of ‘function’. I wonder whether, as trial design evolves, more and more studies will report objective and subjective outcomes as dual primary outcome measures. We know that when an outcome tool does not address a patient’s concern it is often contaminated with overlay from other, often mental health-related, concerns such as anxiety, depression or financial worries.

Although great strides have been made in the assessment of outcomes and providing evidence for treatments, there is still some considerable way to go. Doing the ‘groundwork’ with well thought-through qualitative research, especially in the emergency setting, is time-consuming, expensive and difficult to achieve. A greater understanding of the purpose of outcome assessment within the general orthopaedic community, whether health economic, patient-related, or joint performance focussed, is a welcome development. Of course we do need to decide as a community which is the correct assessment method. The point is amicably illustrated by the ongoing lambasting of arthroscopic meniscal surgery. On the ‘friends and family test’, the majority of patients would recommend treatment, where several randomised controlled trials now suggest that even at mid-term follow-up there is no difference in either joint-specific or health economic outcomes. Nonetheless, there is a large amount of crossover in these studies from conservative to operative measures, so perhaps the patients are telling us we are asking the wrong questions?