Testing the validity of telephone interviews to assess chronic pain in children and adolescents: A randomized cross-over trial.

BACKGROUND: Telephone surveys are intended to reduce attrition in longitudinal studies. For paediatric chronic pain patients, the comparability of pain-related information gathered using telephone interviews and postal surveys remain unknown. Furthermore, it remains unknown how social desirability may influence answers.

METHODS: To compare data from telephone interviews and postal surveys, a randomized cross-over design with two measure points 2 weeks apart and four conditions (combinations of telephone interviews (T) and postal surveys (P): P-T, T-P, P-P, T-T) was conducted in a sample of N = 323 paediatric chronic pain patients.

RESULTS: In the inter-group comparison, pain-related information did not differ between telephone interviews and postal surveys except for the information on pain location (back and extremities). Agreement measures of the intra-group comparisons suggest substantial to excellent agreements for all items and did not differ between the groups. The internal consistency of a disability scale was excellent for both assessment modes; the number of missing values did not differ. Participation rate was higher for telephone interviews compared to the postal surveys. Across both time points, attrition was lowest for the groups without a switch in assessment mode compared to the groups with a switch in assessment mode. Except for pain-related school absence, no effect of social desirability occurred.

CONCLUSIONS: Telephone interviews are a useful method to achieve a high response rate. Pain locations should be asked for separately and not in an open question when interviewing children and adolescents on the telephone.

SIGNIFICANCE: Telephone interviews are a good method to achieve a high response rate and obtain valid data in studies with paediatric chronic pain patients.