Mechanisms behind the failure of residents' longitudinal primary care to promote disclosure and discussion of psychosocial issues.

CONTEXT: Longitudinality (care by a single physician over time) and continuity (receipt of most care from a single physician) are believed to enhance patient-physician relationships and facilitate disclosure of emotional distress, but some studies suggest this potential goes unrealized.

OBJECTIVES: To determine whether care in a pediatric residents' continuity clinic promotes, over time, increased discussion, disclosure, and detection of parents' social and emotional distress and to understand physicians' communication behaviors underlying changes with time.

DESIGN: Longitudinal, observational study of parent-physician interaction over the course of 1 year.

PARTICIPANTS: One hundred ninety parents (90% African American) and their infants' primary care physicians (31 [4 Asians and 27 whites] first- and second-year pediatric residents).

MAIN OUTCOME MEASURES: Frequency with which parents and physicians raised topics related to parental mood and family or social functioning; proportion of distressed parents discussing mood or functioning; and physicians' detection of parent distress.

RESULTS: Physician initiation of psychosocial topics fell in the course of longitudinal relationships (odds of initiation in visits > or =6 vs odds of initiation in visits 1-5 = 0.46 [95% confidence limits, 0.31%, 0.67%]); parent initiation did not change over time nor was it increased by greater levels of continuity. Length of relationship was not associated with increased physician detection of parental distress or with increased rates of disclosure by distressed parents. Physicians' positively framed leading questions, and their avoidant responses to prior parental disclosures were significantly associated with decreased odds of problem disclosure. In contrast, visits in which parents or physicians raised psychosocial topics were characterized, on average, by 40% higher levels of physicians' "patient-centeredness" (increases of about 100 utterances per visit [95% confidence limits, 65.7%, 133.9%]).

CONCLUSIONS: Longitudinal relationships between residents and patients may not be sufficient to promote the discussion, disclosure, and detection of psychosocial issues. Training in communication skills may help residents achieve the potential and goals of longitudinal care.