When workplace interventions lead to negative effects: Learning from failures

Aims: To investigate if workplace interventions resulted in changes in the psychosocial work environment. Process evaluation was conducted to study the implementation process and to use this knowledge to understand the results. Methods: Seven intervention units (n = 128) and seven non-randomized reference units (n = 103) of a large hospital in Denmark participated in an intervention project with the goal of improving the psychosocial working conditions. The intervention consisted of discussion days for all staff, employee working groups, leader coaching, and activities to improve communication and cooperation. Measures of the psychosocial work environment were conducted before the start of the intervention and again after 16 months using 13 scales from the Copenhagen Psychosocial Questionnaire, version I (COPSOQ I). Results: In the intervention units there was a statistically significant worsening in six out of 13 work environment scales. The decrease was most pronounced for three scales that measure aspects of interpersonal relations and leadership. In addition, all three scales that measure aspects of work organization and job content decreased. In comparison, the reference group showed statistically significant changes in only two scales. Process evaluation revealed that a large part of the implementation failed and that different implicit theories were at play. Conclusions: Without the insights gained from process data the negative effects of this intervention could not be understood. Sometimes - as it seems happened in this study - more harm can be done by disappointing expectations than by not conducting an intervention.