DESIGN: Survey.

SETTING: Four primary care clinics.

PATIENTS: Six-hundred twenty-seven adult patients.

MEASUREMENTS: Physician perception of difficulty (Difficult Doctor-Patient Relationship Questionnaire), mental disorders and symptoms (Primary Care Evaluation of Mental Disorders, [PRIME-MD]), functional status (Medical Outcomes Study Short-Form Health Survey [SF-20]), utilization of and satisfaction with medical care by patient self-report.

RESULTS: Physicians rated 96 (15%) of their 627 patients as difficult (site range 11–20%). Difficult patients were much more likely than not-difficult patients to have a mental disorder (67% vs 25%, p<.0001). Six psychiatric disorders had particularly strong associations with difficulty: multisomatoform disorder (odds ratio [OR]=12.3, 95% confidence interval [CI]=5.9–25.8), panic disorder (OR=6.9, 95% CI=2.6–18.1), dysthymia (OR=4.2, 95% CI=2.0–8.7), generalized anxiety (OR=3.4, 95% CI=1.7–7.1), major depressive disorder (OR=3.0, 95% CI=1.8–5.3), and probable alcohol abuse or dependence (OR=2.6, 95% CI=1.01–6.7). Compared with not-difficult patients, difficult patients had more functional impairment, higher health care utilization, and lower satisfaction with care, whereas demographic characteristics and physical illnesses were not associated with difficulty. The presence of mental disorders accounted for a substantial proportion of the excess functional impairment and dissatisfaction in difficult patients.

CONCLUSIONS: Difficult patients are prevalent in primary care settings and have more psychiatric disorders, functional impairment, health care utilization, and dissatisfaction with care. Future studies are needed to determine whether improved diagnosis and management of mental disorders in difficult patients could diminish their excess disability, health care costs, and dissatisfaction with medical care, as well as the physicians’ experience of difficulty.