In this study we sought to find out whether students from the Medical Student Research Programme at the University of Bergen continued to engage in research after completing their medical degree course and to identify the factors that caused the participants to proceed with a doctoral degree, to have published articles within the last two years, to engage in academic teaching or supervision or to be employed in an academic position in a university or university college, respectively. We found that 68 % either had completed or were working on a doctoral degree. In comparison, 18.4 % of all Norwegian specialists had a PhD degree in 2016 (5). The median age at completion of the doctoral degree among the participants in our study was 32 years, while the median age at the time of defending their doctoral thesis in medical and health sciences in Norway was 39 years in 2016 (6). Jacobsen et al. have shown that medical students from the research programmes more often take a doctoral degree and complete their degree earlier than medical students who have not attended such programmes (4). When considering that no more than a median number of five years had passed since the participants had graduated from the medical degree course, it seems that many students from the Medical Student Research Programme at the University of Bergen start their academic career early.
We also found that the majority of those who entered PhD candidate positions did so before starting their period as a junior doctor/LIS1 doctor, and that many of them took up PhD candidate positions earmarked for research programme students. Furthermore, we found that students who had not completed the research programme continued to a PhD to the same extent, but they did so at a later time in their career than those who had completed the programme. A doctoral degree is a prerequisite for an academic career. Early recruitment of research programme students can therefore be regarded as an important factor for an early academic career.
The proportion of women who take a doctoral degree in medicine and health sciences has significantly increased since 1980, and in the years 2010–2016 more women than men followed a doctoral programme (6). While the majority of the medical students in Norway are women, our study included equal proportions of men and women. This finding may indicate that a lower proportion of women choose research during their student years, while female students in the research programme more frequently continue to a doctoral degree. Furthermore, we found some indications that men more frequently were involved in academic teaching and supervision. This is a key component of an academic career following a doctoral degree. All students in the research programme, and women in particular, ought to take the initiative and be encouraged to engage in teaching and supervision activities.
In our cohort, regular supervision led to a higher proportion completing the research programme (7), but paradoxically, we found that those who had been provided with such regular supervision had a lower likelihood of proceeding to a doctoral degree. It is crucial that the supervisor is present to provide follow-up at the early stages of a researcher's career, and gradually leaves more latitude for the student to become an independent researcher.
Jacobsen et al. showed that former students in the medical student research programmes more frequently had published at least one article as lead author as compared to medical students who had not attended the research programme (4). In our study, each participant had published a median number of four articles in total, and two as the lead author. The number of articles was higher among those who had completed the doctoral degree, and three in four of these had published articles within the last two years. This may indicate that they continued to engage in research after completing the doctoral degree. However, a median of two years had passed since these had completed the doctoral degree, and these publications may thus include the doctoral thesis. Because of the short follow-up time, we do not know whether the majority of those participants who had completed the doctoral degree continued to engage in research upon completion.
Those who had completed the research programme were more often engaged in academic supervision or teaching than those who had not completed the programme. We assume that this is due to the fact that students who do not complete the programme tend to have a later start to their academic career, and that supervision and teaching thus come at a later stage in their research career. Even with the relatively short follow-up time in our study, those who have completed the research programme will have had a longer research career and thus a higher likelihood of engaging in supervision and teaching activities.
Only 15 % of those who had completed a doctoral degree held a current position in an academic institution. All these participants were working in part-time positions, and a median number of nine years had passed since they had completed their medical degree course. The observation that no more than a median number of three years had passed since the doctoral examination and that many had taken the doctoral degree before becoming junior doctors may explain why so few of the participants with a doctoral degree were working in academic institutions at the time of the survey. Announcements of academic vacancies for doctors often include a requirement for a clinical speciality. A doctor's academic career often grinds to a halt when he/she is undertaking specialist training. In addition, there are relatively few research positions available for PhD graduates in Norway. This may indicate a need for more combined positions in the specialisation training programmes.
One of the weaknesses of our study is its short follow-up time, meaning that many have not yet had the opportunity to establish a research career after completing their doctoral degree. Since no more than 15 students are admitted to the programme at the University of Bergen annually, the study population is relatively small, and this reduces the study's statistical strength. The differences we have detected are nevertheless large and plausible. Despite a high response rate, we cannot exclude the possibility of a selection bias in the sample, for example underestimation of the number of students who did not complete the research programme or go on to take a doctoral degree. Because of the study design, we cannot conclude that the research activity is higher among former students from the Medical Student Research Programme at the University of Bergen than among medical students who have not attended this programme.