Effect of discordant physician-patient perceptions on patient adherence in inflammatory bowel disease

Background. Discordant physician-patient perceptions on health-related information have been related to less favourable health outcomes and increased use of health services.;Objectives. To develop a psychometrically-sound measure of physician-patient discordance that could be used by clinicians and researchers working with patients with various chronic diseases. To investigate the relationship between physician-patient discordance and patient adherence to self-care in inflammatory bowel disease.;Study design and population. A prospective cohort study with follow-ups at 2-weeks and 4-months was conducted between February and November 1999 at three gastroenterology clinics affiliated with the McGill University Health Centre. Ten physicians and 200 patients with inflammatory bowel disease participated in the study.;Methods. A 10-item visual analog scale questionnaire was developed which assessed perceptions of the patient's health status and of the clinical visit. Questionnaires were completed independently by physicians and patients following the index clinical visit. Discordance was calculated within physician-patient pairs. Demographic, clinical and psychosocial data were obtained prior to the visit. Patient adherence data were obtained at 2-weeks using a telephone interview and mail-back survey. General adherence was assessed with a visual analog scale; medication adherence was determined with a validated questionnaire. Medication data were obtained by chart review at 4-months. Multivariable generalized estimating equations models and mixed models for unbalanced repeated measures analysis of variance were used to determine associations between discordance and patient adherence.;Results. Satisfactory psychometric properties were obtained for discordance scores. Higher psychological distress was the most important determinant of higher discordance. Higher distress was correlated with active disease, less time since diagnosis, greater number and impact of negative life events. Higher satisfaction with social support reduced psychological distress by buffering the negative impact of perceived stress. Higher discordance on symptoms and treatment increased the risk of general nonadherence in patients with higher social support satisfaction. Medication adherence was associated with active disease, greater disease duration and scheduling another appointment. Higher discordance on well-being decreased the probability of medication adherence in non-distressed patients.;Conclusion. Preliminary evidence has been provided to support the validity of the new measure of physician-patient discordance. Higher discordance was associated with an increased risk of nonadherence in patients with healthy psychosocial characteristics.