New insights into causes of burnout

Research by the National Institute of Occupational Health has revealed causes of burnout that have not been investigated before. One of the findings indicates that changes in the levels of burnout appear to predict changes in the levels of sickness absence in a three-year follow-up period.

The recent study Burnout in human service work – causes and consequences (784 Kb PDF) has yielded some interesting insights into the causes of burnout among social services workers (Borritz, 2006). The study is significant, not only because it involves the first mapping of burnout in Denmark, but also because it has uncovered causes of burnout not previously examined.

Study findings

The study defines burnout as ‘exhaustion’ and measures it according to three different scales, reflecting the different domains of life and of job content: personal, work-related and client-related burnout (see ‘Measuring burnout’ below for a definition of these terms). Table 1 outlines the results of a survey conducted in 1999–2000, including survey results concerning the same population in a three-year follow-up period. This longitudinal survey design makes it possible to assess different psychosocial work environment factors as predictors for future burnout levels.

The data show that the following factors all predicted burnout among social services workers, on at least one scale: high emotional demands, need to hide feelings, a heavy workload, a fast pace of work, little scope for development, little meaning to work, unpredictability, poor role clarity and high levels of role conflict.

Most strikingly, the study does not find any evidence of a correlation between the degree of control exercised at work, and susceptibility to burnout, as is often assumed (see, for example, NL0411NU06).

Overall, the relationship between psychosocial work environment factors and burnout seems to suggest that burnout and related sickness absence are preventable, if the psychosocial work environment is improved (as the Foundation report Employment in social care in Europe (2006) also indicates).

Table 1: Predictors for personal, work-related and client-related burnout
Predictors for personal, work-related and client-related burnout
  Personal burnout Work-related burnout Client-related burnout
Client-specific factors      
Emotional demands
Required to hide emotions    
Demands at work      
Workload  
Fast work pace    
Work organisation and job content      
Possibilities for development ÷ ÷  
Meaning of work     ÷ ÷
Interpersonal relations and leadership      
Predictability   ÷  
Role clarity ÷   ÷ ÷
Role conflicts

Notes: The psychosocial work environment factors are adjusted for each other and for age, gender, socioeconomic status, organisation, family status, having children below the age of seven years, smoking and exercise (at the 1999–2000 baseline).

: Statistically significant positive correlations

÷: Statistically significant negative correlations

: Strong statistically significant positive correlations (statistically significant at a lower p value)

÷ ÷: Strong statistically significant negative correlations (statistically significant at a lower p value)

Source: Borritz, 2006

Burnout as a predictor of sickness absence

Burnout is also found to correlate with sickness absence. In relation to work-related burnout, a high level of burnout in the first survey predicted an increase of 21% more sick days and an increase of 9% in spells of sickness absence per year in the follow-up survey. In addition, the study finds that changes in the levels of work-related burnout during the follow-up period are linked to changes in sickness absence: accordingly, an increase in burnout predicts increases in sickness absence and vice versa.

By categorising the respondents into four groups, according to the level of burnout at the time of the two surveys, the impact of burnout on sickness absence can be measured (Table 2). Respondents who had low burnout at the time of both surveys (low–low) experienced an increase of 1.3 sick days. Participants with increasing burnout (low–high) showed an increase of 4.5 sick days, while participants with persistently high levels of burnout (high–high) demonstrated an increase of 3.8 sick days. Finally, those who experienced decreasing levels of burnout (high–low) showed a decrease of 2.8 sick days.

Other potential consequences of burnout, such as job turnover (a potential link mentioned in FI0512NU03), are not measured, as this is not possible at the current stage of research. However, this is to be addressed at a later stage of the research programme.

Table 2: Burnout as a predictor of changes in sickness absence
Burnout as a predictor of changes in sickness absence
Burnout Changes in sickness absence days
Low–low 1.3
Low–high 4.5
High–high 3.8
High–low ÷ 2.8

Notes: Data are adjusted for age, gender, organisation, socioeconomic status, Body Mass Index (BMI), smoking, alcohol consumption, leisure-time physical activity, family status, having children below the age of seven years, and self-reported disease.

Source: Borritz, 2006

Measuring burnout

It is generally agreed that the core symptom of burnout is exhaustion. Based on this understanding, different ways of measuring burnout have been developed. The instrument applied most frequently is the Maslach Burnout Inventory (MBI). It assesses three components of burnout:

  • emotional exhaustion;
  • depersonalisation;
  • reduced personal accomplishments.

The main criticism of the MBI is that, if applied, it results in three independent measures of burnout. Moreover, each dimension may have causes other than burnout. On this basis, the Copenhagen Burnout Inventory (CBI) (126 Kb PDF) has been developed, focusing on exhaustion as being attributable to three different domains: exhaustion in general; exhaustion attributed to work in general; and exhaustion attributed to work with clients. Accordingly, the CBI measures:

  • personal burnout (applies to everyone in and out of the workforce);
  • work-related burnout (applies to everyone in the workforce);
  • client-related burnout (applies to human services workers).

The data show that the three scales for burnout are correlated. This overlap, however, is only partial, thus supporting the notion that burnout is attributable to different domains.

About the study

The study Burnout in human service work – causes and consequences is part of the Project on Burnout, Motivation and Job Satisfaction (PUMA), conducted by the National Institute of Occupational Health. PUMA is a three-wave prospective intervention study on burnout, including six years of follow-up, conducted in the human services sector.

Reference

Borritz, M., Burnout in human service work – causes and consequences, Copenhagen, National Institute of Occupational Health, 2006.

Rune Holm Christiansen and Henrik Stener Pedersen, Oxford Research A/S

Useful? Interesting? Tell us what you think. Hide comments

Eurofound welcomes feedback and updates on this regulation

Коментар