Bayesian Inference for a Two-Part Hierarchical Model

Profiling is currently an important, and hotly debated, topic in health care and other industries looking for ways to control costs, increase profitability, and increase service quality. Managed care in particular has seen a proliferation in the use of statistical profiling methodology, particularly with regard to monitoring expenditure data. This article focuses on the specific problem of developing statistical methods appropriate for profiling physician contributions to patient pharmacy expenditures incurred in a managed care setting. The two-part hierarchical model with a correlated random-effects structure considered here accounts for both the skewed, zero-inflated nature of pharmacy expenditure data and the fact that patient pharmacy expenditures are correlated within physicians. The random-effects structure has an attractive interpretation in terms of a conceptual model for physician prescribing patterns. Using this model, we propose to rank physicians based on an appropriately constructed provider-level performance measure. This information is subsequently used to develop a novel financial incentive scheme. Inference is conducted in a Bayesian framework using Markov chain Monte Carlo.

[1]  J. Tebbs,et al.  An Introduction to Categorical Data Analysis , 2008 .

[2]  Robyn Tamblyn,et al.  Physician and Practice Characteristics Associated with the Early Utilization of New Prescription Drugs , 2003, Medical care.

[3]  Dongchu Sun,et al.  Fully Bayesian spline smoothing and intrinsic autoregressive priors , 2003 .

[4]  R. Strawderman,et al.  Quantifying the Physician Contribution to Managed Care Pharmacy Expenses: A Random Effects Approach , 2002, Medical care.

[5]  M. Rosenthal,et al.  Transmission of financial incentives to physicians by intermediary organizations in California. , 2002, Health affairs.

[6]  G. Gastaldi.  Bayesian analysis of covariance matrices and dynamic models for longitudinal data , 2002 .

[7]  A. Agresti,et al.  A Correlated Probit Model for Joint Modeling of Clustered Binary and Continuous Responses , 2001 .

[8]  Joseph L Schafer,et al.  A Two-Part Random-Effects Model for Semicontinuous Longitudinal Data , 2001 .

[9]  F. Burge,et al.  Physician characteristics in relation to cardiovascular drugs commonly prescribed for hypertension in Nova Scotia. , 2001, The Canadian journal of clinical pharmacology = Journal canadien de pharmacologie clinique.

[10]  W. Manning,et al.  The Unreliability of Individual Physician “Report Cards” for Assessing the Costs and Quality of Care of a Chronic Disease , 2000 .

[11]  J. Marron,et al.  SiZer for Exploration of Structures in Curves , 1999 .

[12]  G. Larkin,et al.  Ethical issues of managed care. , 1999, Emergency medicine clinics of North America.

[13]  T. Miller,et al.  Disclosing physician financial incentives. , 1999, JAMA.

[14]  K. Vranizan,et al.  Primary care physicians' experience of financial incentives in managed-care systems. , 1998, The New England journal of medicine.

[15]  E. Emanuel,et al.  Ethical guidelines for physician compensation based on capitation. , 1998, The New England journal of medicine.

[16]  M. Cowen,et al.  Casemix adjustment of managed care claims data using the clinical classification for health policy research method. , 1998, Medical care.

[17]  M. Glickman,et al.  Statistical Methods for Profiling Providers of Medical Care: Issues and Applications , 1997 .

[18]  C. Jaén.  Primary-care physicians. , 1997, Tobacco control.

[19]  L. Ryan,et al.  Latent Variable Models for Mixed Discrete and Continuous Outcomes , 1997 .

[20]  T. Louis,et al.  Consistency in performance among primary care practitioners. , 1996, Medical care.

[21]  Harvey Goldstein,et al.  League Tables and Their Limitations: Statistical Issues in Comparisons of Institutional Performance , 1996 .

[22]  G. Verbeke,et al.  A Linear Mixed-Effects Model with Heterogeneity in the Random-Effects Population , 1996 .

[23]  W. Chang,et al.  Applying case mix adjustment in profiling primary care physician performance. , 1996, Journal of health care finance.

[24]  J. Weiner,et al.  Profiling primary care physician resource use: examining the application of case mix adjustment. , 1996, The Journal of ambulatory care management.

[25]  R. Hays,et al.  Managed Care and Capitation in California: How Do Physicians at Financial Risk Control Their Own Utilization? , 1995, Annals of Internal Medicine.

[26]  C A Gatsonis,et al.  Variations in the Utilization of Coronary Angiography for Elderly Patients with an Acute Myocardial Infarction: An Analysis Using Hierarchical Logistic Regression , 1995, Medical care.

[27]  Paul R. Rosenbaum,et al.  Comparing the Contributions of Groups of Predictors: Which Outcomes Vary with Hospital Rather than Patient Characteristics? , 1995 .

[28]  Charles W. Plows,et al.  Ethical Issues in Managed Care: Council on Ethical and Judicial Affairs, American Medical Association , 1995 .

[29]  Bradley P. Carlin,et al.  Markov Chain Monte Carlo conver-gence diagnostics: a comparative review , 1996 .

[30]  J. Berger,et al.  Estimation of a Covariance Matrix Using the Reference Prior , 1994 .

[31]  D. Molloy,et al.  Relation between physician characteristics and prescribing for elderly people in New Brunswick. , 1994, CMAJ : Canadian Medical Association journal = journal de l'Association medicale canadienne.

[32]  J P Kassirer,et al.  The use and abuse of practice profiles. , 1994, The New England journal of medicine.

[33]  C J McDonald,et al.  Estimating Physician Costliness: An Empirical Bayes Approach , 1993, Medical care.

[34]  Adrian F. M. Smith,et al.  Sampling-Based Approaches to Calculating Marginal Densities , 1990 .

[35]  George H. Dunteman,et al.  Principal Components Analysis , 1990 .

[36]  J. Loehlin.  Latent variable models , 1987 .

[37]  C. Morris,et al.  A Comparison of Alternative Models for the Demand for Medical Care , 1983 .