Appointing a new medical consultant colleague is one of the most expensive decisions a hospital can make. Over a period of 25 years the investment in one individual doctor may run in millions of pounds, taking account of salary, pension, discretionary awards, equipment and medicines purchases etc. Typically, such major investments were based on a half hour interview, brief presentation, word-of mouth (\u201cHe\u2019s a good chap we went to the same medical school\u201d) and a couple of references. In the last 5 years, Trusts around the UK have recognised the significant deficiencies of such a process and many have now taken steps to re-design the way they appoint medical consultants. Many have moved to a competency-based or a values-based approach and deploying a range of tools to explore the presence or absence, or potential for the required behaviours and values.<\/p>\n
Is it an exact science? If that means \u2013 \u201ccan you make a 100% accurate and foolproof prediction of the right person for the job\u201d \u2013 the answer is no. Is there a \u201cbest practice\u201d way to do this? Definitely yes \u2013 with several decades of research in selection and assessment to back it up. What is the strongest prediction you are likely to achieve? Research suggests around 70% with the right combination of selection tools. Simply adding more and more tools to the process does not, beyond a certain point, increase the predictive validity of the process. In essence, a well-structured situational interview, psychometric tests (cognitive ability and personality) and \u201cwork sample\u201d tests (exercises that test the candidate in situations close to the reality of the work situation) give the best prediction of job performance.Situational judgement tests have now become a standard element in selection for specialist medical training. However, most Trusts have not gone down the route of cognitive ability tests for consultants on the assumption that medical training has already filtered some of the brightest intellects in the population. Despite the fact that cognitive ability is at least as strong a predictor of performance as personality, the face validity of such measures for the medical population remains fragile. There are three key questions for any selection process \u2013 medical consultants are no different \u2013 1) Can they do the job? 2) How will they do the job? And 3) Why do they want the job? This last question is rarely explored in enough depth, and indeed it is not easy to tap into a candidate\u2019s real motivation for wanting a position. We are learning much more about how to identify motives, values and interests \u2013 supported by an excellent psychometric tool from the Hogan personality test range.<\/p>\n
We have worked with many different Trusts in the UK to support them in the development of a more robust, evidence-based selection process. We have been struck by the conversion (in every case!) from scepticism and resistance to a process involving psychometric tests and behavioural interviews, to the conviction that such methods have produced a far more accurate picture of the candidate and a more confident and robust (defensible) selection decision.<\/p>\n
In the training we provide to interview panel members, we always introduce a break between practice sessions to review the skills. This break in the interview has now been adopted by several Trusts who have found it invaluable to take 10 minutes as a panel to review what they have learned in the first half, re-adjust the questions, and ensure that they are making the most of the interview opportunity.<\/p>\n
We have been providing pen portrait psychometric profiles to the panel with hypotheses about what the profiles might indicate, and suggested interview questions to test these further. These help the selection panel to interpret the psychometrics in relation to the Trust\u2019s competencies or values, with a clear steer on how to probe these traits in greater depth.<\/p>\n
These measures represent a far cry from the \u201cold\u201d way of doing things and there is often loud protest about the restrictions that prevent the use of prior knowledge about a candidate when making an appointment. \u201c We have worked with this trainee for 4 years\u2013 he interviewed poorly but we know he\u2019s a good doctor so we should appoint him anyway\u201d. This is a compelling argument \u2013 but it\u2019s flawed. First, the assumption that a good trainee will automatically become a good consultant, whilst likely, is not foolproof. I have seen several trainees who, once recruited by their senior consultant, have fallen out with them. I have also seen trainees with gold-plated training records who hit the buffers when appointed to a consultant post \u2013 often because expectations of them were unrealistically high. Second, behaviours that may not have been evident during training while trainees are being well-supervised may come to the fore during the often stressful transition to becoming a consultant, and subsequently cause problems. Third, in the interests of a fair process, all candidates must be judged according to the same criteria and using the same process. This simply raises the stakes and means that the selectors must be highly skilled at ensuring that they make best use of the selection process. Several Trusts we work with tell us that using the enhanced process means that they no longer necessarily appoint the best candidate \u201con paper\u201d . They told us that they are more confident about rejecting a candidate with a good track record, but who performed poorly during the selection process.<\/p>\n
No doubt all of this will provoke yet more debate.<\/p>\n
by Dr. Jenny King<\/em><\/p>\n","protected":false},"excerpt":{"rendered":" Appointing a new medical consultant colleague is one of the […]<\/p>\n","protected":false},"author":1,"featured_media":4482,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"footnotes":""},"categories":[2,20],"tags":[],"table_tags":[],"acf":[],"yoast_head":"\n