6023.0: Thursday, November 16, 2000 - 8:45 AM

Abstract #11496

Estimating the effect of number of treatment sessions on depression outcomes using managed care limits as instruments

Jodi L. Constantine, PhD, Department of Health Care Policy, Harvard Medical School, 180 Longwood Avenue, Boston, MA 02115, 617-432-3447, constantine@hcp.med.harvard.edu and Jeff DeSimone, PhD, Department of Economics, East Carolina University, Brewster A-427, 10th Street, Greenville, NC 27858, 252-328-6742, desimonej@mail.ecu.edu.

Counseling is a common form of treatment for depression, although research has focused on which types of therapy ultimately improve mental health rather than estimating the impact of any counseling therapy session. Our paper estimates the effect of counseling sessions on mental health outcomes for depressed patients using an instrumental variable (IV) estimation method to account for the bias that ordinary least squares (OLS) regression might produce if patients with slower outcome improvements stay in treatment longer.

Data include treatment and outcome information on 244 depressed adults who received outpatient treatment from a community-based social service agency that serves patients under numerous funding contracts. Managed care financing components that attempt to lower total costs through reductions in sessions administered serve as natural identifying variables in this IV framework.

The IV estimate indicates that additional treatment sessions have a positive and statistically significant impact on mental health functioning, which contrasts the OLS finding that mental health outcome does not depend on the number of treatment sessions attended. These findings suggest that the imposition of session limits may reduce mental health outcomes.

We conclude that 1) constraints imposed by managed care in order to reduce costs may negatively affect mental health outcomes, and 2) IV methods can prove useful in situations where OLS is likely to produce a biased estimate and viable instrumental variables are available.

Learning Objectives: 1. Define Instrumental Variables (IV) and describe appropriate use of IV estimation. 2. Discuss differences between OLS and IV estimation results related to managed care and depression treatment. 3. Apply IV estimation to public health research to inform programs, practice, and policy

Keywords: Managed Care, Statistics

Presenting author's disclosure statement:
Organization/institution whose products or services will be discussed: none
I do not have any significant financial interest/arrangement or affiliation with any organization/institution whose products or services are being discussed in this session.

The 128th Annual Meeting of APHA