Provider choice and use of mental health care: implications for gatekeeper models

To examine the ways in which the costs of nonresidential mental health care depend on (1) the type of provider who initiates the treatment episode and (2) the level of cost sharing imposed on the patient.

The 1987 National Medical Expenditure Survey, a national probability sample of the U.S. civilian, noninstitutionalized population.

Data were collected during four personal interviews conducted during 1987 and 1988. Key variables include the type of provider contacted at the beginning of treatment (psychiatrist, other physician, nonmedical mental health care specialist) and the cost (total actual payments from all sources) for the treatment episode.

An episodic model of demand for mental health care is estimated using a two-step procedure. Multinomial probit analysis is first used to determine the factors that influence the choice of initial provider type. Right-censored Tobit analysis is used to determine the factors that affect the costs of care, including the type of provider who initiates the care episode.

Results indicate that out-of-pocket price does significantly (p < .05) affect the patient's initial choice of provider type but that, after controlling for the endogeneity of provider choice, price is no longer significant in explaining overall treatment costs. After controlling for selection effects, care episodes initiated by nonspecialist physicians are found to be as expensive as those initiated by psychiatrists and significantly more expensive than episodes initiated by nonphysicians.

The results suggest that nonmedical mental health care specialists may be more effective than physicians in controlling costs when used as case managers in the care of persons with mental illnesses.