Patterns of Disease-Modifying Antirheumatic Drug Use, Medical Resource Consumption, and Cost Among Rheumatoid Arthritis Patients

A systematic review of validated methods for identifying patients with rheumatoid arthritis using administrative or claims data

PURPOSE

To review the evidence supporting the validity of billing, procedural, or diagnosis code, or pharmacy claim-based algorithms used to identify patients with rheumatoid arthritis (RA) in administrative and claim databases.

METHODS

We searched the MEDLINE database from 1991 to September 2012 using controlled vocabulary and key terms related to RA and reference lists of included studies were searched. Two investigators independently assessed the full text of studies against pre-determined inclusion criteria and extracted the data. Data collected included participant and algorithm characteristics.

RESULTS

Nine studies reported validation of computer algorithms based on International Classification of Diseases (ICD) codes with or without free-text, medication use, laboratory data and the need for a diagnosis by a rheumatologist. These studies yielded positive predictive values (PPV) ranging from 34 to 97% to identify patients with RA. Higher PPVs were obtained with the use of at least two ICD and/or procedure codes (ICD-9 code 714 and others), the requirement of a prescription of a medication used to treat RA, or requirement of participation of a rheumatologist in patient care. For example, the PPV increased from 66 to 97% when the use of disease-modifying antirheumatic drugs and the presence of a positive rheumatoid factor were required.

CONCLUSIONS

There have been substantial efforts to propose and validate algorithms to identify patients with RA in automated databases. Algorithms that include more than one code and incorporate medications or laboratory data and/or required a diagnosis by a rheumatologist may increase the PPV.

Is it possible to identify early predictors of the future cost of chronic arthritis? The VErA project

This study was conducted to identify early predictors of the total cost of inflammatory arthritis (IA). One hundred and eighty patients affected by undifferentiated arthritis (UA) or rheumatoid arthritis (RA) were included in the French Very Early rheumatoid Arthritis (VErA) cohort between 1998 and 2001. Health economic data for 2003 were collected using a patient self-questionnaire. Results were analysed in terms of direct, indirect and total costs in 2003 euros (2003euro) for the population as a whole and in diagnostic subgroups. A payor perspective (the French National Health Insurance, in this case) was adopted. Multiple linear regression models were used to identify predictors of total cost from among the criteria assessed on recruitment. Results of the study showed that for the study population as a whole, the mean total cost was euro4700 per patient. The costs attributable to the RA and UA sub-groups were euro5928 and euro2424 per patient, respectively. In a univariate analysis, certain parameters were significantly correlated with a higher cost of illness. In the multivariate analysis, some of these parameters were further identified as being predictive of higher cost. Two strong significant, early predictors of total cost were identified: higher pain (P = 0.002) and the presence of rheumatoid factor (P = 0.004). In the RA sub-group, lower grip strength of the dominant hand (P = 0.039) was another predictor of the illness's subsequent economic impact. In conclusion, our data show that simple clinical and laboratory parameters can be used early in the course of IA to predict the condition's impact on healthcare budgets.

Use of tumor necrosis factor inhibitors in rheumatoid arthritis: a national survey of practicing United States rheumatologists

OBJECTIVES

To determine the prescribing practices, laboratory monitoring protocols, and perceived barriers of United States rheumatologists in prescribing tumor necrosis factor (TNF) inhibitors in rheumatoid arthritis (RA).

METHODS

A survey questionnaire was mailed to 1970 rheumatologists who were randomly selected from a national sample of 3008 rheumatologists. A one-page non-response questionnaire was mailed to approximately 200 randomly selected non-responding rheumatologists to assess non-response bias.

RESULTS

Two mailings yielded a response rate of 22.3% (428 completed, usable surveys out of 1922 deliverable surveys). Rheumatologists reported using all three agents in patients with moderate RA (82-87%), severe RA (94-96%), and in newly diagnosed and mild RA patients (10-18%). In patients with severe RA who inadequately responded to methotrexate, 91% of rheumatologists reported using a TNF inhibitor with one other disease modifying anti-rheumatic drug. Over 94% of rheumatologists reported switching patients from one TNF inhibitor to a different TNF inhibitor due to inadequate response or side effects. Most rheumatologists (96%) ordered the purified protein derivative test for tuberculosis, with almost 82% conducting this test at baseline. Costs to patients and insurance coverage were perceived as major barriers to prescribing these agents although the perception was slightly lower with infliximab than with adalimumab or etanercept.

CONCLUSIONS

The use of TNF inhibitors is not restricted to patients with moderate and severe RA. Rheumatologists are fairly similar in their utilization of the three TNF inhibitors although some variation exists in terms of laboratory practices and perceived barriers regarding the use of these agents.

Alternative Decision Analysis Modeling in the Economic Evaluation of Tumor Necrosis Factor Inhibitors for Rheumatoid Arthritis

OBJECTIVES

To provide a review of the studies that use decision models in the economic evaluation of tumor necrosis factor (TNF) inhibitors in rheumatoid arthritis (RA) and to address some important issues surrounding the choice of such modeling techniques in these economic evaluations.

METHODS

A systematic literature search was conducted by 1 author from the literature published from January 1996 to March 2005 through Medline, Embase, and Cochrane library databases.

RESULTS

The review yielded 29 studies that used decision models. Only 10 studies used a decision model in the economic analysis of the TNF inhibitors and were included in the final review. Decision model types included the following in the review articles: decision tree (2), Markov model (7), and discrete event simulation (1). These models vary in complexity and their choice depends on the course of disease, the impact of treatment, and the available data.

CONCLUSIONS

Based on the results derived from alternative modeling techniques, it is safe to say that all methods can provide useful information with regard to economic evaluations of TNF inhibitors. Even though different modeling techniques provide an appropriate representation of available data, their results should be interpreted contingent on the input data, assumptions, sensitivity analyses, and other alternative scenario analyses.

RELEVANCE

The transparency in the models will encourage end users such as policymakers and prescribers to make informed judgments regarding the appropriateness of the methods and the validity of the results.

Rheumatoid arthritis: direct and indirect costs

Rheumatoid arthritis (RA) causes disability, deformities, progressive radiological joint damage often with a need for joint replacement surgery, premature death, and alterations in quality of life. The economic burden created by RA is enormous. Direct costs per patient have been estimated at 1812-11,792 Euro annually and indirect costs at 1260-37,994 Euro annually. These mean values are approximations, as variations occur across countries, healthcare system organizations, and geographic locations. Direct costs account for one-fourth to slightly over a half of the total cost. Costs associated with inpatient care contribute up to 75% of direct costs, as compared to only about 20% for medications, although wide variations occur in costs related to drug monitoring and side-effect management. Physician visits account for about 20% of direct costs. As compared to indirect health costs for individuals from the general population, those for RA patients are increasing at a rapid rate. Indirect costs account for 80% of the excess cost related to RA. Cost estimates may change over time and show huge variations across individuals, with a small minority of patients accounting for most of the costs. Disability as measured by the Health Assessment Questionnaire (HAQ) has a major impact on costs. Early effective treatment may not only postpone and slow disease progression, thereby improving quality of life, but also decrease costs by preserving productivity and reducing the need for surgery, admission to acute-care and extended-care hospitals, and social service utilization. Data are beginning to accumulate on the excess costs associated with biotherapies and other new second-line drugs. They indicate acceptable excess costs relative to the additional medical benefits and to the cost-effectiveness of other healthcare programs. Nevertheless, the threshold that defines an acceptable excess cost is arbitrary and varies with local economic conditions.

Decreases in rates of hospitalizations for manifestations of severe rheumatoid arthritis, 1983-2001

OBJECTIVE

To determine whether the rates of hospitalization for 4 manifestations of severe rheumatoid arthritis (RA), which are used as indicators of long-term health outcomes, have changed from 1983 to 2001.

METHODS

Data on all patients with RA who were hospitalized with rheumatoid vasculitis or to undergo splenectomy for Felty's syndrome, cervical spine fusion for myelopathy, or total knee arthroplasty at hospitals in California were abstracted from a state hospitalization database. Changes in rates of hospitalization from 1983 to 2001 were examined in this serial cross-sectional study.

RESULTS

Rates of hospitalization for rheumatoid vasculitis and splenectomy in Felty's syndrome decreased progressively over time. The risk of hospitalization for rheumatoid vasculitis was one-third lower in 1998-2001 than in 1983-1987. The risk of hospitalization for splenectomy in Felty's syndrome was 71% lower in 1998-2001 than in 1983-1987. There were no significant decreases in the rates of hospitalization for cervical spine surgery or total knee arthroplasty (primary and revision), although in 1998-2001 there was a reversal of the trend of increasing rates of total knee arthroplasty. The risk of hospitalization for primary total knee arthroplasty was significantly lower in 1998-2001 than in 1990-1993 (rate ratio 0.90, 95% confidence interval 0.86-0.95; P < 0.0001).

CONCLUSION

Rates of hospitalization for rheumatoid vasculitis and splenectomy in Felty's syndrome have decreased over the past 19 years, and there has been a recent decrease in the rates of primary total knee arthroplasty in patients with RA. Although several factors may account for these decreases, these findings suggest that since the early 1980s, the long-term health outcomes of patients with RA have improved.

Cost of illness studies in rheumatic diseases

Knowledge about the economic burden of rheumatic diseases has progressed during recent years. In addition to the increasing number of studies published, the Economics Working Group of the Outcome Measures in Rheumatoid Arthritis Clinical Trial (OMERACT) Conference has produced substantial work to improve methodological standards for the economic evaluation of rheumatic diseases. Some of their preliminary results are presented in this review. Recent data have confirmed two main conclusions of previous studies: the total economic burden of rheumatic diseases is often more substantial than other chronic conditions, including cardiovascular diseases and cancer; and the impact of the disability caused by musculoskeletal diseases is significant on both direct (long-term care in osteoporosis for example) and indirect costs (productivity loss in chronic patients). Besides that, cost-effectiveness studies have provided valid information to improve disease management, especially for patients with rheumatoid arthritis or chronic low back pain.