Trying to predict who will be a happy doctor

Being a doctor can be incredibly stressful and more than a fifth of junior physicians wish they’d never taken medicine in the first place. Moreover, there’s evidence showing that stressed doctors are bad doctors, so it would make sense if applicants to medical school were selected not just on the basis of their academic strengths but also on whether they’re likely to enjoy being a doctor. But in the UK, besides exam results, which are known to be a bad predictor of later stress and burnout, the only information medical school selectors currently have to go on are students’ personal statements and referees’ reports. To find out if these can be used to judge which applicants will become happy, stress-free doctors, Chris McManus and colleagues recruited assessors and showed them applications made in 1990, to see if they could predict which applicants had gone on to become very happy doctors in 2002, and which had become stressed and wanted to change jobs.

Each assessor was shown 20 pairs of personal statements and referees’ reports taken from a pool of 80 applicants, and had to say in each case which belonged to a subsequently happy doctor (based on data collected in 2002) and which belonged to an applicant who had become stressed and wanted to stop being a doctor.

The researchers found that 35 expert medical school selectors performed no better than chance at this task. Nineteen doctors, 22 medical students, and 20 psychology students, who also completed the task, fared no better.

However, the assessors’ judgements were not random – they tended to agree with each other and their predictions also correlated with the applicants’ exam results. It seems that to predict who would become a happy doctor, the assessors were mistakenly using clues to the applicants’ academic prowess that were contained in the personal statements and referees’ reports.

The researchers said “…although many claims are made for the utility of the personal and referees’ information [contained on application forms], we could find no evidence of the long-term predictive validity for an important outcome variable – the judgment of whether or not an applicant will be a happy and satisfied doctor, or instead will be an unhappy, stressed, burned out, dissatisfied doctor who does not enjoy their job and thinks of leaving for another career”.

McManus, I.C., Iqbal, S., Ferguson, E. & Leaviss, J.. (2005). Unhappiness and dissatisfaction in doctors cannot be predicted by selectors from medical school application forms: A prospective, longitudinal study. BMC Medical Education, 5:38.

Post written by Christian Jarrett (@psych_writer) for the BPS Research Digest.

Link to article in The Psychologist by Chris McManus on this topic.