A Three-year Cohort Study of the Relationships between Coping, Job Stress and Burnout after a Counselling Intervention for Help-seeking Physicians

Karin E Isaksson Ro; Reidar Tyssen; Asle Hoffart; Harold Sexton; Olaf G Aasland; Tore Gude


BMC Public Health. 2010;10(213) 

In This Article


In this prospective three-year follow-up study of physicians who sought a counselling intervention, we found that the significantly reduced level of emotional exhaustion, job stress and emotion-focused coping seen after one year, compared with baseline, were maintained at three-year follow-up.[26] The clinical significance of these results is indicated by the changes being of moderate effect sizes, describing a reduction in level of emotional exhaustion from significantly higher to a level not significantly different from Norwegian physicians in general, and a reduction in level of job stress from significantly higher to significantly lower than of Norwegian physicians in general.[26] Additionally, the enhanced work capacity, as indicated by a substantial reduction in the proportion of Sana physicians who were on sick leave at follow-up (both at one- and three-years) compared with baseline indicates a clinical significance.

The long-term reduction in emotional exhaustion contrasts with results from the few previous follow-up studies of preventive interventions for physicians that have found a reduction in emotional exhaustion up to a year after the intervention,[25,30,40] but indicate a relapse without additional interventions.[17,32] In the present cohort there were no planned additional interventions. Some of the participants had, however, on their own initiative, chosen to come to a second intervention at the Resource Centre (primarily within the first six months, before the one-year follow-up), some had sought psychotherapy, and some had implemented a practical intervention by reducing weekly work hours, as reported previously.[26] All these post-intervention initiatives may have contributed to the reduction in emotional exhaustion over the years.

In this study there was a reduction in emotion-focused coping strategies occurring before the reduction in emotional exhaustion. Although previous studies have found that physicians under stress report more use of active coping strategies than their colleagues,[11,12] this result indicates that it was a reduction in emotion-focused coping strategies, rather than an increase in active coping, that influenced reduction in emotional exhaustion. The reduction in emotion-focused coping, such as self-blame or wishful thinking, may have been a factor reducing the risk of relapse in emotional exhaustion in experience of renewed stress-exposure. The associations between active and emotion-focused coping strategies and their influence on emotional exhaustion warrant further investigation.

For the group of physicians attending psychotherapy after the counselling intervention a reduction in emotion-focused coping strategies was seen mainly from one- to three-year follow-up, whereas the group not attending psychotherapy showed a reduction from baseline to one-year follow-up. The former group was initially more distressed, reporting higher baseline levels of emotional exhaustion and neuroticism than the rest of the cohort. The results appear to indicate that some physicians can change their coping strategies after a short-term intervention (like the intervention at the Resource Centre, Villa Sana). The most distressed physicians, however, appeared to need additional psychotherapy that may have contributed to the significant reduction in emotion-focused coping found from one- to three-year follow-up in this group. These findings may strengthen previous recommendations of counselling and psychotherapy as primary and secondary preventive interventions for physicians.[41,42] This is in accord with studies of other distressed groups, where therapeutic interventions have been found to have the potential to change unfavourable coping strategies.[13] Since therapy in this study was self-selected, further work is needed to confirm this.

Longitudinal studies have previously shown reciprocal relationships between changes in emotional exhaustion and changes in job stress in normative samples of physicians.[15,16] In this study, however, a unilateral relationship was found, indicating that a reduction in perceived job stress occurred before reduction in emotional exhaustion. This may imply that the relationship between these parameters differs among physicians with high initial levels of emotional exhaustion, as in the present cohort, compared with physicians in normative samples.

As mentioned above, previous studies have shown a relationship between work-home interface stress and emotional exhaustion.[2,4,16] Consistent with this, we found that a reduction in the job stress dimension "social stress" (including both work-home interface stress and time pressure), preceded change in emotional exhaustion from baseline both to one- and three-year follow-up. The results for the two other job stress dimensions, emotional stress and stress due to fear of litigation were less consistent. Focusing on reduction in job stress, especially the social dimension may consequently be important in interventions for distressed physicians.

Neuroticism has previously been described as a relatively stable trait,[28] whereas in this study there was a trend towards reduction from baseline to one year and significant reduction from one- to three-year follow-up. Neuroticism has previously been associated with, and has predicted, emotional exhaustion,[3,8,28] whereas in this study changes in neuroticism (as estimated by the revised Eysenck's Personality Questionnaire) were preceded by reduction in emotional exhaustion. These relationships might differ between normative groups of physicians and a selected group, as in the present cohort. Further studies of the sequential relationships between changes in emotional exhaustion and personality traits in initially distressed doctors are needed.

Strengths and Limitations

A strength in this study is its prospective design with a three-year follow-up period and the relatively high proportion of participants completing one- and three-year follow-up (81%). Due to the study design, without a control group, we cannot determine whether the changes found are related to the intervention or whether they are a spontaneous regression towards the mean. A further possibility is that the changes could be related to factors not assessed in this study, for example social support. Our main objective, however, has been to investigate the three-year course of and the temporal relationships between the factors measured in this study after a counselling intervention

The five-point scoring scale of the Maslach Burnout Inventory (MBI) enabled us to compare the Villa Sana cohort with other Norwegian studies of physicians, but complicates direct comparisons with other international MBI-based studies, where a seven-point frequency scale is used. However, in this instance the former is more important than the latter, and regarding the relative roles of coping and job stress for change in emotional exhaustion we do not believe that the scale differences would have a substantive effect on the conclusions of the present study. Another limitation, reducing transnational generalizability of our results is the difference in working conditions for physicians in different countries. Although reduced work hours yield reduced emotional exhaustion both in Norway and in the U.S.,[23–26] Norwegian physicians work fewer hours per week than e.g. American or British physicians[23–25] which could contribute to different relationship between the variables.

The sample size is important in structural equation modelling techniques.[39,43] In order to limit the number of parameters in the model, observed rather than latent variables were used. This was possible since the internal consistencies of the variables were acceptable. The constructed models fitted the data well, as indicated by the satisfactory model fit indices. In spite of restrictions due to the limited sample size, the main relationships between emotional exhaustion, job stress and emotion-focused coping strategies were generally consistent at one- and three-year follow-up.

This study indicates that reduction in emotion-focused coping and job stress preceded reduction in emotional exhaustion. These findings do not necessarily reflect the only possible lagged relationships among these parameters. As stated above, the intervention focused on the use of coping strategies and reduction of stress, and the temporal associations found might reflect this focus. Also, these changes are found in a group of physicians with initially elevated levels of emotional exhaustion who have decided to seek a counselling intervention and can therefore not be generalized to all physicians. Additionally, the literature indicates that there are other coping strategies like spirituality, not examined in this paper, which also could have importance for changes in emotional exhaustion.

Participants lost to three-year follow-up were more often men and had higher levels of distress (emotional exhaustion and job stress), as well as higher levels of emotion-focused coping strategies at baseline. It is difficult to estimate how inclusion of these participants would have influenced results concerning change of these parameters from baseline to follow-up as well as regarding how measures influence each other during follow-up. However, the proportion lost to follow-up was relatively small (19%).


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: