There is no doubt that doctors make mistakes. Mostly people forgive them, the charitable view is that it’s because people recognise that their doctors are human and by and large are trying to do a good job. The issue becomes most problematic when the error might cause a delay in the diagnosis of a condition that is best treated sooner rather than later. And especially when the red flag symptoms of that condition are well documented.
Late in 2014 Devesh Oberoi interviewed men who had presented to a specialist late with symptoms that were later diagnosed as cancer. One of the interviews suggested that the delay might in part be due to a late referral:
I spoke to my GP …that time … and. … I was concerned about the symptoms. I told him that I had seen some blood on my toilet paper and he said … umm … yeah that … since it is fresh blood it could be piles (haemorrhoids) or something. Patient with Rectal carcinoma.
Such delays are widely reported in the literature with some experts calling for better research to establish why the diagnosis of cancer is sometimes late in primary care.
Last week our team published secondary data from an experimental study in which we report that the diagnosis of cancer can be missed even when the presentation is straightforward and there are no distracting issues in the consultation (e.g. co-morbidity, psychiatric illness or social problems). One in eight ‘cases’ presented as short video vignettes to doctors in the study failed to elicit a response that included a referral to a specialist or investigations to establish the diagnosis of cancer. What’s also of concern is that where the management decision was to prescribe something, it was hard to see the benefit. In some cases it might even have resulted in harm. Where the decision was to investigate, the indications for some of the investigations were not immediately apparent. Delays may also have occurred in those investigated if the findings were negative or misleading.
None of this is new. Numerous audits have established similar patterns including one we published in 2004 in which three reasons were given for a failure to recognise patterns of cancer:
- A failure to consider the diagnosis of cancer. ‘Blinkered’ approach in assessing patient.
- Inappropriate or incomplete investigation.
- False-negative investigations.
Despite such findings some policy makers think that it is appropriate to pay GPs to focus more on preventive health; to drive payment structures to reflect this public health agenda and distract doctors at the front line of the health service from their core business, namely giving a patient, who consults very briefly, their undivided attention. Doctors need to reflect when they have failed in someway to deliver a satisfactory outcome especially in cases of life threatening illness. That requires a renewal of the commitment to the process of history taking and examination and to updating the skills to make the diagnosis of conditions that are best treated ASAP. When done properly this is time consuming. When doctors are otherwise incentivised to either collect data or tick boxes the result can be less than satisfactory. That it may be already unsatisfactory even before we are driven to adopt practices for which there is very little proven benefit should lead to a rethink.
In relation to pay for performance the King’s Fund reported in 2010:
What evidence does exist suggests that significant improvements have been made in some areas – particularly for diseases such as diabetes, heart failure and chronic obstructive pulmonary disease –but less progress has been made for depression, dementia and arthritis, and these require a more collaborative care model for a higher quality of care to be achieved.
Alternatively it may be that what we can’t afford is to pay GPs to do better at something at the cost of deskilling them in other aspects of their work.
Picture by David Goehring