Psychiatry is a major medical specialty in the UK, accounting for over 10% of all doctors in secondary care in England, and about 11.5% of consultant posts. ¹ Indeed, if the prowess of a medical specialty can be judged from its size, then over the past decade UK psychiatry has gone from strength to strength. From 2000 to 2011, the absolute number of consultant psychiatrists in England rose from 2904 to 4394 (an increase of 51%), with similar increases in Scotland and Wales, in line with the expansion of the consultant workforce in most secondary care specialties. ¹ The majority of the increase (87%) was accounted for by female staff. A need for more psychiatric consultant posts in the period up to 2014 has been predicted based on projections of increasing clinical and medico-legal workload. ² Despite this, the spectre of falling recruitment has haunted UK psychiatry in recent years. ^{Reference Brockington and Mumford3,Reference Brown, Vassilas and Oakley4} A perceived fall in the proportion of UK medical school graduates choosing a postgraduate career in psychiatry, and low competition ratios for first-year core specialist training (CT1) posts in psychiatry, ^5–7 have generated much soul-searching about what is taught in medical schools, and how psychiatry could be made to appear more attractive. ^{Reference Brown and Eagles8–10}

The study by Goldacre and colleagues ^{Reference Goldacre, Fazel, Smith and Lambert11} from the UK Medical Careers Research Group in this issue of the Journal exorcises some of these ghosts, and raises some interesting new visions of the future of psychiatric recruitment in the UK. The researchers ascertained career choices using an established questionnaire methodology that was applied to all medical graduates of all UK medical schools in each of 12 cohorts of graduates between 1974 and 2009. In three cohorts (2002, 2005, 2009), questions were added about careers that had been ‘seriously considered’ but rejected, and the reasons for the rejection. Two important findings challenge established, and deeply ingrained, views about quantitative aspects of recruitment into psychiatry. First, the proportion of graduates in UK medical schools specifying psychiatry as their intended career destination has varied little over the 35 years during which these career surveys have been conducted. That proportion, although low, has remained relatively stable between 4 and 5%; this is comparable with other high-income countries (such as the USA ¹² ), and is capable of maintaining a more than ten-fold difference between the number of psychiatrists per head of population in high-income countries and low- and middle-income countries. ^{Reference Höschl, van Niekerk, L, B and D13,14} Second, although the proportion of male graduates entering psychiatry has been consistent for about 25 years, the proportion of women doctors choosing psychiatry has begun to decline. As the authors suggest, the latter finding might result from psychiatry losing its relative ‘family friendly’ appeal as other (acute medical and surgical) specialties benefit from improved flexible training schemes ¹⁵ and the European Working Time Directive.

Where, then, are we to look for an explanation of the apparent numerical deficiencies in psychiatric recruitment? First, at the population base from which future psychiatrists are recruited: according to the Organisation for Economic Cooperation and Development (OECD), the number of medical graduates in the UK rose from 4432 in 2000, peaking at 6208 in 2007, and stabilising at 5757 in 2010. ¹⁶ This 30% increase was outstripped in the same decade by the >50% increase in medical staff overall (general practice plus hospital specialties). ¹ Thus, at an estimated 5% rate of recruitment, the yield of new doctors from all UK medical schools wishing to enter psychiatry is only about two-thirds of the number required to fill all CT1 vacancies in England. ⁷ Even allowing for the increase in UK medical school intake over the same period (to about 6400 by 2010 but reducing by 2% from 2013), ¹⁷ and assuming all students graduate after a lag period of 5–7 years, applying the same estimated recruitment rate predicts only 90% of posts would be filled. Second, it is probable that failure to fill all psychiatry training posts is not a new phenomenon. Centralised recruitment to a national scheme, rather than by individual trusts or training schemes, has shown up deficiencies that were likely to have been present in the past but were overlooked because popular schemes at major ‘centres of excellence’ were not affected. Third, this shortfall might have been met partially in the past by international (European Union (EU) and non-EU) medical graduates (IMGs). ^{Reference Höschl, van Niekerk, L, B and D13} However, recent changes in UK immigration and visa requirements have limited severely the numbers of non-EU IMGs available for this role, and the number of EU graduates seeking employment in the UK is too small, except in some specialist areas, to make up the deficit. ¹ Whatever the cause of low recruitment rates into psychiatry, uptake from UK medical schools would need to increase to an average of at least 8% if competition ratios for CT1 posts are to be kept above one. The Royal College of Psychiatrists has squared up to this task but the lead time on implementation of the recruitment strategy, ¹⁰ even if maximally successful, is likely to be 5 or 6 years during which clinical psychiatry will suffer continued attrition.