Although previous studies have addressed Norwegian medical students’ and young doctors’
hazardous drinking, and possible mediators of this behaviour, we still lack representative and
longitudinal studies on hazardous drinking and life satisfaction among doctors whose careers
are more well-established. Studies have also shown that both individual- and work-based
factors are associated with drinking and life satisfaction among doctors, yet prospective
longitudinal studies are needed to identify possible risk factors and establish causality. Such
information is important both to doctors personally and to the medical associations and
organizational authorities that need to develop interventions to prevent doctors from
developing alcohol problems, and to counteract doctors’ potential for life dissatisfaction. In
addition, long-term longitudinal data from doctors may, to some extent, be generalized to
other populations, especially with respect to individual predictors.
This thesis used data from the Longitudinal Study of Norwegian Medical Students
and Doctors (NORDOC), in which surveys were administered to two nationwide cohorts of
doctors: the Medical Student Cohort and the Young Doctor Cohort. The Medical Student
Cohort (N = 421) commenced in 1993 with new medical students; the Young Doctor Cohort
(N = 631) assessed students graduating from medical school in 1993 and 1994. These cohorts
were spaced six years apart at baseline and both were subsequently surveyed five times, with
the last survey administered in 2014 (NORDOC 6). Most previous NORDOC project papers
have used longitudinal data from the Young Doctor Cohort only, whereas there are fewer
publications from the Medical Student Cohort. The current study is the first to use NORDOC
data from both cohorts, merged at each of four consistent time points in participants’ careers
and analysed longitudinally.
This thesis consists of three research papers. Paper I examined possible risk factors
present during medical school for later hazardous drinking. The prevalence of hazardous
drinking declined from 14% at the end of medical school to 8% by year 10 post-graduation.
About 23% of those who drank hazardously in medical school continued this behaviour
during years 4 and 10 post-graduation. Drinking to cope with tension during medical school
was found to be an important risk factor for hazardous drinking post-graduation, and males
had a higher likelihood of hazardous drinking. Adjusted medical school predictors of
hazardous drinking during post-graduation year 4 were male gender, use of alcohol to cope
with tension, and hazardous drinking; older age, male gender and previous hazardous
drinking were the adjusted predictors of hazardous drinking at year 10 post-graduation.
Paper II identified co-occurring work-related risk factors for hazardous drinking,
while controlling for individual factors and mental distress (i.e., anxiety and depressive
symptoms). Somewhat unexpectedly, there was no independent relationship between work
stress, or other work-related factors, and hazardous drinking. Mental distress and life events
were independently related to hazardous drinking in a repeated-measures model, whereas
drinking to cope (i.e., during medical school) was an independent predictor of hazardous
drinking. Men were at higher risk of hazardous drinking.
Paper III examined trends in life satisfaction over the 15 years following graduation,
and whether work stress had an independent impact on overall life satisfaction among doctors
during this period (after controlling for individual and lifestyle confounders, such as drinking
behaviours). Doctors’ mean life satisfaction ratings did not change across the three
observational follow-up periods. There were no significant gender differences in level of life
satisfaction. The work-related factors that were independently associated with higher life
satisfaction were low work-home stress, low perceived job demands and high colleague
support. Both hazardous drinking and drinking to cope were independently related to life
satisfaction. In addition, those who experienced a decrease in life satisfaction showed higher
levels of neuroticism personality trait and lower colleague support.
Consistent with studies of other populations, we found that individual factors, mental
distress and life stress were the most important risk factors for hazardous drinking among
doctors. With respect to their work-related factors, we found no link with hazardous drinking,
but found an association between these factors and doctors’ life satisfaction. This was true for
work–home stress, perceived job demands and colleague support. Low colleague support was
another important predictor of doctors’ reduced life satisfaction during the follow-up years.
These findings support the notion that doctors’ work-related factors might be improved
through organizational-level interventions, including a more supportive work environment,
towards increasing their life satisfaction. In addition, preventive measures should be taken to
reduce medical students’ and doctors’ drinking to cope with tension and mental distress,
particularly for men who are at higher risk of hazardous drinking.