Original ArticleThe Charlson comorbidity index is adapted to predict costs of chronic disease in primary care patients
Introduction
As the burden of chronic disease in the United States continues to grow, providing cost-effective health care services has become increasingly important. Recent studies project that the number of Americans living with chronic illness will increase from 125 million in 2000 to 164 million (or nearly 50% of the population) over the next three decades. Currently, an estimated 78% of total health care resources are already devoted to individuals with chronic diseases [1].
In the past, most studies have sought to measure the costs of specific chronic conditions, including arthritis [2], [3], diabetes [4], [5], [6], skin diseases [7], asthma [8], [9], hypertension [10], depression [11], [12], [13], and other diseases. Although appropriate for evaluating individual conditions, such an approach fails to allow for broad comparisons among chronic diseases or analysis of persons with multiple coexisting medical conditions. In recent years, however, a small number of studies have sought to provide a systematic estimate of costs of multiple chronic conditions in a single patient population, the most notable of which were conducted by the Group Health Cooperative in Seattle [14] and the Kaiser Permanente group in California [15] in the environment of managed care, and by Yu et al. in the VA System [16].
The present study builds upon this work, while providing a unique contribution in analyzing the costs of chronic diseases in an academic general internal medicine practice. Additionally, it develops a model for predicting future costs. The work of predicting costs ultimately shapes models of health care expenditures. Many previous models have been developed, including risk-adjustment models for Medicare enrollees. Such models have often explored the contribution of specific chronic diseases or diagnostic groups to yearly costs [17], [18], [19]. The role of comorbidity (the total burden of chronic diseases) has received surprisingly little attention, despite its proven ability to predict mortality and its importance in determining hospital costs [20], [21]. Our study is the first to incorporate a prospectively obtained comorbidity index in a rigorous statistical model aimed at predicting total cost of care.
The goal of this study was to use prospectively acquired data to determine the predictors of the cost of care of a large cohort of primary care patients. We used computerized electronic medical records to construct a model that identifies the demographic and clinical features (including specific chronic illnesses, comorbidity, and medications) predictive of the total yearly costs.
Section snippets
Population
To qualify for the study, patients had to have been followed for at least 1 year prior to December 1, 1993. Among patients seen at an academic medical practice at New York Hospital, 5,861 met this criterion. Of those, 2,864 patients were seen by 18 attending physicians. These attendings also supervised 153 residents and two nurse practitioners who saw 2,997 patients during continuity and ambulatory block experiences.
Data collection
The data for this paper were captured by CLIMACS©, a practice management system
Average yearly cost per patient
The average cost per year was $2,655. Average hospital costs, which were $1,558 per patient, accounted for 58.7% of the overall cost. Ambulatory costs totaled $935 per year and accounted for 41.3% of overall cost. In total, 667 patients, or 11.4 % of patients, were hospitalized an average of 1.5 times during the year. The total hospitalization cost on average for patients who were hospitalized was $13,693, whereas their ambulatory costs were $1,307. Thus, yearly total cost for patients who were
Discussion
The present study determined the annual health care costs for a large cohort of primary care patients, using prospectively acquired data to determine the predictors of those costs. Our study found an average yearly total cost per patient in 1994 of $2,655, which is consistent with data from contemporary studies. In 1992, a large staff model Health Maintenance Organization in Seattle showed a mean total cost estimate of $2,006 per patient [14]. A study of Health Maintenance Organization
References (36)
- et al.
The health care costs of diabetic nephropathy in the United States and the United Kingdom
J Diabetes Complicat
(2004) - et al.
The economic burden of skin disease in the United States
J Am Acad Dermatol
(2003) - et al.
Resource costs for asthma-related care among pediatric patients in managed care
Ann Allergy Asthma Immunol
(2003) - et al.
A comprehensive study of the direct and indirect costs of adult asthma
J Allergy Clin Immunol
(2003) - et al.
A new method of classifying prognostic comorbidity in longitudinal studies: development and validation
J Chronic Dis
(1987) - et al.
Adapting a clinical comorbidity index for use with ICD-9-CM administrative databases
J Clin Epidemiol
(1992) - et al.
The growing burden of chronic disease in America
Public Health Rep
(2004) - et al.
The economic burden associated with osteoarthritis, rheumatoid arthritis, and hypertension: a comparative study
Ann Rheum Dis
(2004) - et al.
Direct medical costs and their predictors in patients with rheumatoid arthritis: a three-year study of 7,527 patients
Arthritis Rheum
(2003) - et al.
Economic costs of diabetes in the US in 2002
Diabetes Care
(2003)