We all like people to do what we want. There are ways to achieve this, usually involving metaphorical sticks or carrots. Government is starting to think about nudging the science around choice (if there is such a thing). However what we are usually talking about in healthcare is ‘compliance’, usually defined as a willingness to follow a course of treatment but in effect meaning whether patients take the medicine prescribed by physicians. In the good old days, doctors prescribed the right pills and model patients took them. Then it emerged that actually as patients we only ever take our medicine correctly about half the time. Even worse, we now know that about only half the clinicians prescribe the right drugs. So an intervention with a capacity to reduce symptoms say by 60% only achieves about 15%.
Previously the only focus was on how to improve patients’ compliance with treatment. This is crucial in some areas of mental health, like psychotic illness, where insight can be sufficiently impaired that compliance with treatment is reduced. However the flip-side is to focus on ensuring clinician, not patient, compliance. This isn’t always easy. Take hand-washing, a known ‘intervention’ to reduce infection and iatrogenic complications within hospital. Despite being a low-cost preventative strategy, accessible to all clinicians, getting successful clinician compliance has been problematic. Notwithstanding high-profile campaigns, clinicians will over-estimate their compliance with hand-washing when covertly observed. This has led to novel strategies to improve hand-washing such as publically shaming under-compliant clinicians, the use of ward computer screensavers depicting bacterial Petri dish growth or best of all getting patients to ask their clinician personally if they have washed their hands.
Healthcare organisations are wrestling with areas of clinician ‘compliance’ in other ways. Take two examples. First, there is a debate is about finding the balance between algorithmic approaches to treatment, usually based on evidence, such as Maps of Medicine or NICE guidelines, with the specific clinical approach required for a particular patient with their specific presentation. This is a challenge in some areas of mental health where evidence about interventions can be scanty or the outcome data are insufficiently rich to evaluate. Most of us can usually make a case for exempting (not complying with) some part of a patient’s treatment from a standard algorithm or pathway, based on our wider experience, although with limited subsequent performance data to determine whether it is a good decision. This needs to change.
Second, clinicians are suffering from policy and procedure compliance overload. A recent BMJ article, whose author is a (ahem) Human Factors Consultant, noted how the clinical management of an elderly patient admitted with a broken hip would include 75 clinical and trust guidelines and policies. Think about how many emails clinicians might get a week helpfully updating them on changes to policy and procedures. Sending the emails doesn’t ensure compliance. If we want clinicians to do the right thing so that patients can have the right thing, simplifying the system (rather than shaming) might help.
Dr John Milton
Consultant Forensic Psychiatrist & Forensic Research Lead
Nottinghamshire Healthcare NHS Trust