Background Systematic reviews of case-control studies demonstrated an association between self-reported sexual and physical abuse in childhood/adolescence and FMS in adulthood (1,2). However, the case-control studies reviewed did not analyse if the association of self-reported maltreatments in childhood/adolescence and adulthood FMS was mediated by depression. Depressive mood could induce a negative recall bias of childhood (3).
Objectives We tested if the association of self-reported maltreatments in childhood/adolescence and FMS in adulthood is mediated by depression in a case-control study.
Methods All consecutive patients diagnosed with FMS (either by ACR 1990 or modified ACR 2010 criteria) of one secondary care pain medicine centre and one tertiary care psychosomatic medicine centre from January to June 2011 were included into the study. Randomly selected age- and sex-matched controls from a representative survey of the general German population were used for controls. Childhood/adolescence maltreatments were assessed by the validated German version of the Childhood Trauma Questionnaire CTQ, depression by the two-item depression scale of the validated German version of the Patient Health Questionnaire PHQ 4. The scores of the five CTQ- subscales were compared between FMS-patients and controls using analysis of covariance adjusting for depression. Group was specified as a fixed between-subject factor. Effect sizes for ANCOVA were expressed as partial eta2 which were interpreted as a small effect size ≥0.01, a medium effect size ≥0.06 and as a large effect size when ≥0.13.
Results 153 FMS-patients (87.6% women; mean age 50.3 years, SD 9.6 years) with a mean duration of chronic widespread pain of 10.6 (SD 8.9) years and 153 age- and sex matched participants of the general population were included into the study. The ANCOVA demonstrated a significant group difference with large effect size (partial eta2=0.21, p<0.001) for emotional abuse, with medium effect size (partial eta2=0.12, p<0.001) for physical abuse, with medium effect size (partial eta2=0.07, p<0.001) for sexual abuse, with medium effect size (partial eta2=0.09, p<0.001) for emotional neglect and medium effect size (partial eta2=0.09, p<0.001) for physical neglect. The group differences were partially attributable to depression with a small effect size (partial eta2=0.05, p<0.001) for emotional abuse, with small effect size (partial eta2=0.03, p<0.001) for physical abuse, with medium size (partial eta2=0.07, p<0.001) for emotional neglect and medium effect size (partial eta2=0.06, p<0.001) for physical neglect. There was no significant effect size of depression on group differences in sexual abuse (partial eta2=0.009, p=0.10).
Conclusions FMS-patients reported more severe childhood/adolescence maltreatments than age- and sex-matched persons of the general population. The association between self-reported childhood/adolescence maltreatments and diagnosis of adult FMS was only partially mediated by depression. Childhood/ adolescence maltreatments can be regarded as risk factors of adulthood FMS.
 Paras ML. JAMA 2009: 5;302(5):550-61.
 Häuser W. Arthritis Care Res (Hoboken) 2011;63(6):808-20.
 Kosseva M. Schmerz 2010;24(5):474-84, in German.
Disclosure of Interest None Declared