Continuing medical education and burnout among Danish GPs

Study population

In a cross-sectional questionnaire study, all 458 active GPs in May 2004 in the County of Aarhus were invited to participate. Four GPs were excluded: one was a member of the research group, and three were on leave.

Questionnaire

The 10-page self-administered questionnaire included scales on burnout and questions about CME activities and GP and practice characteristics. Burnout was measured by the Maslach Burnout Inventory – Human Services Survey (MBI-HSS), an international scale which is considered to be reliable and valid.14^–16 This self-rating scale consists of 22 items forming three subscales. Each item is scored on a 7-point ordinal scale, where ‘never’ is scored 0 and ‘every day’ is scored 6. The three subscales are: emotional exhaustion (9 items), which refers to reduced energy and decreased job enthusiasm; depersonalisation (5 items), which refers to cynicism and lack of empathy; and personal accomplishment (8 items), which refers to inefficacy.

Each dimension receives a score which, relative to normative population scores, is categorised as low or high. Burnout was defined by a high score on the emotional exhaustion subscale (>26) and/or on the depersonalisation subscale (>9). High burnout was defined by a high score on the subscales for emotional exhaustion (>26) and depersonalisation (>9), and a low score on the personal accomplishment subscale (<34).14

CME was assessed by asking GPs if they were members of local CME groups or supervision groups (CME with a psychological approach as in the Balint groups), if they used the practice facilitator (outreach visits), and how many days during the past 2-year period they had spent on conferences, courses, and seminars relevant to general practice.

Non-responders were sent a reminder with a new questionnaire after 4 weeks. GPs were remunerated for responding.

Analysis

For each GP, scores were calculated on the three burnout subscales. Whether the scores reached burnout or high burnout was also recorded. Only GPs who had answered all items on a subscale were included in the analysis of subscale scores, and burnout and high burnout was only assessed for GPs who had answered all necessary items.

Days spent during a 2-year period on conferences, courses, and seminars relevant to practice were categorised into four groups: 0–5, 6–9, 10–17, and more than 17 days. This classification was made because every GP in Denmark is entitled to at least 3 full days of course activity a year. GPs who were members of both a CME group and a supervision group were compared with GPs who were members of only one of the groups, and with GPs who were not members of either a CME group or a supervision group.

The association between feeling emotionally exhausted, depersonalised, having low personal accomplishment, suffering from burnout or high burnout, and CME activities was calculated. Associations were calculated as prevalence ratios (PRs) in a generalised linear model with log link for a binomial outcome.

Calculations were made of the crude PR and an adjusted PR including age, sex, practice organisation (solo practice, group practice, or partnership practice), number of enrolled patients per GP (<1385, 1385–1785, or >1785 patients), and weekly working hours (<36, 37–41, 42–46, 47–51, or >51 hours).

Ninety-five per cent confidence intervals (CI) were calculated for proportions and ratios with P-values of 5% or less considered statistically significant. Data were analysed using Stata 9.0.

Summary of main findings

The study results show that GPs who were not members of a CME group or who did not take part in outreach visits were more likely to suffer from burnout and high burnout than GPs who were members of a CME group or took part in outreach visits. It was not possible to show an association between number of days spent on courses and burnout. The prevalence of burnout among Danish GPs was about 25%, and almost 3% suffered from high burnout. One in three GPs were found to be lacking a sense of personal accomplishment, one in five suffered from depersonalisation, and one in 10 GPs was emotionally exhausted.

Strengths and limitations of the study

The strengths of this study were the use of an international validated scale for measuring burnout, a high response rate, and statistical precision. The high response rate indicates that the GPs found the questionnaire relevant. As the study was cross-sectional, it is not possible to make any causal inference; for example, whether GPs would suffer from burnout less frequently if they participated in CME activities or, more reasonably, if the GPs who chose not to participate were those who suffered from burnout.

The validity of the burnout measure has been shown to be valid in cross-sectional studies. However, if the measure is unstable, for example if it depends too much on job stress on a particular day, the measure may be inaccurate. It has not been possible to find data on reproducibility of the MBI-HSS. However, the MBI-HSS has been shown to be highly associated with mental health and job satisfaction, for example,17^–19 which indicates that the measure is stable.

Although the study had a high response rate, a small group of physicians did not respond, which may provide a basis for possible selection bias. It is reasonable to assume that GPs experiencing burnout would have less energy and enthusiasm to complete the questionnaire. Given a higher prevalence of burnout among non-responders, the prevalence of burnout may therefore have been underestimated. The non-responding group would not have been expected to have participated in more CME activities, and a possible selection bias would not change the association found between CME and burnout.

CME activities were measured from GPs' self-reported use of outreach visits, membership of CME and supervision groups, and use of days on courses. The use of days on courses during a 2-year period may invite recall bias which may explain why no association was detected here.

Comparison with existing literature

The prevalence of burnout and high burnout among GPs found in this study is supported by the findings of a Swiss study which showed that one-third of primary care physicians experienced burnout, and 2.8% experienced high burnout.18 Studies of UK and Italian GPs found higher levels of emotional exhaustion and depersonalisation, but lower values on personal accomplishment compared with the present findings.8^,20

Only one study has been identified that examines burnout and CME among physicians.11 Kushnir et al found a negative association between CME activities and burnout among Israeli GPs.11 The authors suggested that CME and professional updating may reduce the level of burnout. The current study is not completely comparable to their study, however, as a different approach to measure CME and burnout was used.

Implications for future research and clinical practice

GPs who were not members of a CME group had a significant higher likelihood of suffering from burnout and high burnout than those who were members of a CME group. Therefore, not being a member of a group could indicate that these GPs may be more likely to suffer from burnout. Although the present study does not unequivocally establish causality, it seems reasonable to be aware of possible implications if a GP stops CME activities, or to encourage GPs to be a member of a CME group. However, it remains unanswered whether it is the group, the CME, or the combination of the two which is most important. Future research could also investigate whether CME groups may be more attractive to GPs with a similar suitable learning style and behaviour, which may indicate that the groups may not be inclusive enough.