Decline in rheumatoid vasculitis prevalence among US veterans: A retrospective cross-sectional study

Authors


Abstract

Objective

To examine trends in the prevalence of rheumatoid vasculitis in a national US population comprising both hospitalized and ambulatory patients with rheumatoid arthritis (RA).

Methods

In this serial cross-sectional study, we analyzed data on hospitalized and ambulatory patients spanning 22 years (1985–2006) and 10 years (1997–2006), respectively, to determine the prevalence of rheumatoid vasculitis, as defined by the International Classification of Diseases, Ninth Revision. Our search encompassed data collected on a predominantly male study population during 10 million hospitalizations and outpatient visits, and included annual data on >37,000 RA patients. To test for a decrease in rheumatoid vasculitis prevalence, breakpoint analysis was performed using stepwise Chow and Durbin-Watson tests.

Results

There was a clear decline in the prevalence of rheumatoid vasculitis, and this decline remained evident even after accounting for a decreased number of hospitalizations among RA patients. Peak prevalence occurred among hospitalized patients in the 1980s, and prevalence gradually declined throughout the 1990s. Furthermore, simultaneous breakpoints representing a significant drop in rheumatoid vasculitis prevalence between the years 2000 and 2001 were demonstrated for both inpatients (P < 0.000) and outpatients (P < 0.003). The prevalence of vasculitis dropped 53% among inpatients and 31% among outpatients between 2000 and 2001.

Conclusion

Our results demonstrate a significant decline in rheumatoid vasculitis prevalence after 2000 in this nationwide sample of hospitalized and ambulatory patients. The clear, consistent drop in prevalence provides an opportunity for the formulation of causal hypotheses, including consideration of the impact of biologic agents used to treat RA, on rheumatoid vasculitis.

Current practices for managing rheumatoid arthritis (RA) help to reduce articular damage, and the results of some studies indicate that aggressive treatment modifies the entire spectrum of rheumatoid disease by decreasing the frequency of severe disease (1–3). Arguably, severe extraarticular manifestations, such as rheumatoid vasculitis, are the most serious sequelae of RA. Despite an increased comprehension of RA that has resulted, in part, because of new treatments used to combat the disease, important gaps remain in our understanding of trends in the prevalence of extraarticular manifestations, including rheumatoid vasculitis.

Findings of recent studies examining rheumatoid vasculitis reflect a lack of consensus regarding prevalence trends (4–8). Two studies showed a decrease in rheumatoid vasculitis prevalence, while a third showed no change. All 3 studies had methodologic limitations, including 2 that examined relatively few cases in population-based cohorts and 1 that analyzed only hospitalized patients. In the absence of representative national data sets, examining extraarticular complications among US patients with RA is challenging.

The purpose of this study was to examine trends in the prevalence of rheumatoid vasculitis in a national US data set representing both hospitalized and ambulatory RA patients. We hypothesized that rheumatoid vasculitis prevalence has declined in recent years, not due to shifts in the settings of care, but possibly related to aggressive treatment of RA, including the use of biologic agents.

PATIENTS AND METHODS

In this serial cross-sectional study, we analyzed data on patients admitted to hospitals over a period of 22 years (1985–2006) or seen in outpatient settings over a period of 10 years (1997–2006), in order to identify possible trends in the prevalence of rheumatoid vasculitis. After receiving the approvals of the Department of Veterans Affairs and the local Institutional Review Board, we searched the national Veterans Health Administration (VHA) system to assess rheumatoid vasculitis prevalence, including paired inpatient and outpatient data from 1997–2006 (Table 1). As of 2000, 26.5 million veterans qualified for VHA care, and ∼4 million veterans were receiving services (9).

Table 1. VHA populations and RA and rheumatoid vasculitis prevalence data*
YearOutpatientsInpatients
Total/RA casesRheumatoid vasculitis cases, no. (%)Rheumatoid vasculitis cases/ 1,000 RA casesTotal/RA casesRheumatoid vasculitis cases, no. (%)Rheumatoid vasculitis cases/ 1,000 RA cases
  • *

    Numbers reflect unique individuals seen in each study year. For the years 1985–1996, the mean (range) total cases were 570,184 (∼478,000–635,000). The mean (range) RA cases were 5,197 (4,136–5,823). The mean (mean %) (range) rheumatoid vasculitis cases were 245 (4.7) (171 [4.1]–289 [5.0]). The mean (range) rheumatoid vasculitis cases/1,000 RA cases were 47.0 (41.3– 49.6). For this same time period, data on outpatients were not available. VHA = Veterans Health Administration; RA = rheumatoid arthritis.

20065,078,432/43,3471,440 (3.3)33.2366,400/3,36949 (1.5)14.5
20054,998,714/43,6701,370 (3.1)31.4370,414/3,24941 (1.3)12.6
20044,871,136/43,6551,328 (3.0)30.4364,783/3,18951 (1.6)16.0
20034,692,694/44,5601,156 (2.6)25.9356,983/3,21841 (1.3)12.7
20024,434,671/44,1291,148 (2.6)26.0357,427/3,11244 (1.4)14.1
20014,052,934/41,7911,179 (2.8)28.2356,416/3,13948 (1.5)15.3
20003,640,811/38,9611,600 (4.1)41.1351,173/3,03199 (3.3)32.6
19993,390,527/38,7461,630 (4.2)42.1357,507/3,15199 (3.1)31.4
19983,235,212/37,8131,586 (4.2)41.9369,138/3,192129 (4.0)40.4
19972,956,905/37,8781,585 (4.2)41.8404,273/3,397140 (4.1)41.2

Our analysis involved searching through data collected during 10 million hospitalizations and outpatient visits, and included annual data on >37,000 RA patients. Cases were identified using discharge and clinical diagnostic codes from the International Classification of Diseases, Ninth Revision (ICD-9). In selecting rheumatoid vasculitis cases, we utilized a previously validated ICD-9 coding algorithm (6) that captures the rheumatoid vasculitis manifestations described by Scott et al (10). For our study, the algorithm was modified to improve specificity by excluding patients with diabetes mellitus (DM). In order to be classified as having vasculitis, patients could not have DM (ICD-9 code 250), and had to have RA (code 714.0) and 1 of the following additional features: arteritis (code 447.6), mononeuritis multiplex (code 354.1, 354.2, 354.3, 354.8, 354.9, or 355), peripheral neuropathy due to connective tissue disease (code 357.1), gangrene (code 785.4), or chronic ulcers (code 707.1). Cases were selected and sorted to count unique individuals who were encountered in hospital and outpatient settings. An individual may have been counted in both settings within a given year; thus, analysis remained separated by setting.

We used time series analysis and autoregression models to test the hypothesis that rheumatoid vasculitis prevalence has declined. Time series analysis was completed using stepwise Chow tests, and autoregression models were completed using Durbin-Watson tests. Briefly, the Chow test compares a simple linear-trend null hypothesis against a model with a breakpoint and 2 separate regression lines (representing a change in trend or level) to determine if model performance improves when the time period is divided into 2 parts. P values less than or equal to 0.01 were considered significant for trend and breakpoints. The Chow test was performed stepwise for all years, with the smallest P value tested as a breakpoint. Identical analyses were performed to examine prevalence trends among veterans with RA seen in hospital settings and those seen in outpatient settings in each study year.

RESULTS

In our study, >92% of RA patients were men, with a mean age of 64.9 years. The total number of veterans receiving VHA services increased over time, and the total number of RA outpatients increased initially and then leveled off in recent years (Table 1). The number of hospitalizations relative to the total number of veterans receiving treatment declined in the mid-1990s, which reflects an overall US health care trend toward the management of chronic diseases in outpatient settings. Absolute numbers of patients with RA remained consistent.

There was a clear drop in the prevalence of rheumatoid vasculitis among both hospitalized and ambulatory veterans with RA (Figures 1 and 2). Importantly, autoregression models and stepwise Chow testing revealed a breakpoint in the prevalence of rheumatoid vasculitis among both hospitalized patients (P < 0.000) and ambulatory patients (P < 0.003). Examination of both the inpatient and outpatient data sets showed identical breakpoint timing between the years 2000 and 2001 (Figure 1).

Figure 1.

Rheumatoid vasculitis prevalence among hospitalized veterans with rheumatoid arthritis (RA). Circles represent the prevalence of rheumatoid vasculitis measured in each study year as a proportion of all RA patients hospitalized between 1985 and 2006. Note the peak prevalence of 5% with a linear decline until 2000, and the subsequent drop to 1.3–1.6%.

Figure 2.

Rheumatoid vasculitis prevalence among veterans with rheumatoid arthritis (RA). Circles represent the prevalence of rheumatoid vasculitis measured in each study year as a proportion of all inpatients with RA between 1997 and 2006, and triangles represent the prevalence of rheumatoid vasculitis measured in each study year as a proportion of all outpatients with RA for the same years. For both populations, a drop in prevalence occurred between 2000 and 2001.

The magnitude of the drop in rheumatoid vasculitis prevalence between 2000 and 2001 was ∼31% among outpatients and 53% among inpatients. During that same time period, vasculitis prevalence fell from 41 to 28 cases per 1,000 RA cases among outpatients, and from 32 to only 15 cases per 1,000 RA cases among inpatients. An alternate analysis that included inpatients with rheumatoid vasculitis without excluding cases with DM revealed parallel linear trends and a decrease from 33 to 16 cases per 1,000 RA cases from 2000 to 2001. Using our original definition for determining case selection, absolute rates of rheumatoid vasculitis prevalence peaked at 52 cases per 1,000 hospitalized veterans with RA (∼5%) in the 1980s (Figure 1). Recently, prevalence of rheumatoid vasculitis among hospitalized veterans has fallen below 2% (15 cases per 1,000 RA cases), from ∼4% in 1997. Minimal subsequent increases in rheumatoid vasculitis point prevalence among outpatients were noted from 2004 through 2006. These points did not meet significance for trend (P = 0.045), and the number of cases remained between 30 and 33 per 1,000 RA outpatients, compared with >40 cases per 1,000 RA outpatients from 1997 through 2000.

DISCUSSION

Our results demonstrate a significant decline in the prevalence of rheumatoid vasculitis after 2000 in both hospital and outpatient settings. Rates of hospitalization decreased in the mid-1990s, which illustrates the importance of also examining the prevalence of rheumatoid vasculitis among outpatients prior to formulating conclusions. Ultimately, in this simultaneous analysis, the trends among inpatients and outpatients mirrored one another, thus demonstrating that reported declines did not merely reflect changes toward increased outpatient management.

Previous reports have described conflicting conclusions regarding trends in the incidence and prevalence of rheumatoid vasculitis. One population-based study from the UK showed a decrease in rheumatoid vasculitis prevalence (8), but the conclusions of that study were based upon only 51 total cases. A US population-based study showed no change in rheumatoid vasculitis prevalence based upon 42 cases (5). However, in that study, followup of the RA cohort was truncated (the cohort with the most-recent-onset RA had been diagnosed between 1985 and 1994, with study data not collected beyond the year 2000), meaning that extraarticular events that typically occur later in RA might have been missed (4, 7). A study examining data from California hospitals showed that hospitalizations for rheumatoid vasculitis declined between 1980 and 2001 (6). However, analysis of only inpatient data leaves questions regarding whether contemporary rheumatoid vasculitis cases may have escaped detection due to evolving outpatient management.

To our knowledge, this study is the first to examine a national US population for rheumatoid vasculitis prevalence among both inpatients and outpatients. In the current study, by pairing inpatient and outpatient data from the national VHA system, we sought to overcome limitations of previous studies that examined only regional cohorts or included only hospitalized patients. Consistent with the findings of the California study (6), a significant decline in rates of hospitalization for rheumatoid vasculitis was revealed in our study. Importantly, however, we demonstrated a simultaneous drop in the prevalence of rheumatoid vasculitis among ambulatory patients, confirming that the drop was not an artifact of shifts in the setting of RA care. The timing of the decline described in the current report was also consistent with the findings of the UK study, which cited a 3-year decrease in rheumatoid vasculitis prevalence from 1998 through 2000 (7, 8).

The serial cross-sectional design of our study and the demographics of the population of veterans we analyzed impose some limits on a wide application of our results. By design, the study captured only the prevalence of rheumatoid vasculitis, and chart review validation of individual cases was not performed. Cases were selected using a previously validated algorithm that was modified to improve specificity by excluding patients with DM, though the exclusion of these cases may have impacted case definition due to time trends in DM. Nevertheless, prevalence findings in this report are consistent with previously published incidence rates ranging from <1% to 5.4% (11–13). Furthermore, patients included in our study were predominantly men who were older than average RA patients and whose use of tobacco was likely higher than that in most RA cohorts. These demographics may limit the ability to apply our results to the general RA population. However, some previous reports have described increased rates of extraarticular manifestations among men, smokers, and persons who were older at the time of RA onset (14, 15), so perhaps the characteristics of our study sample facilitated detection of changes among the most at-risk population.

A key strength of the current study was the inclusion of data from large numbers of patients who were followed up over many years in both hospital and outpatient settings, which reinforces conclusions drawn with regard to overall trends in rheumatoid vasculitis prevalence. The structure of the data sets facilitated simultaneous comparisons, which demonstrated a consistent drop in vasculitis prevalence among hospitalized and ambulatory RA patients.

Our findings offer an opportunity to propose new testable hypotheses regarding trends in rheumatoid vasculitis prevalence. Given the method by which prevalence is determined (incidence times duration minus case efflux), we believe that the decline we report may indicate one of the following: a decrease in the number of incidences (not directly testable here), a disease cure in some patients, patients either discontinuing use of VHA services or dying, or error. The 1990s through 2000 were years that marked important changes in RA treatment. One may speculate that the use of biologic agents and strategies involving more aggressive combinations of disease-modifying antirheumatic drugs meant to target severe RA may have diminished extraarticular disease, including rheumatoid vasculitis, by treating existing cases and preventing new ones. Additionally, VHA system-wide efforts to decrease tobacco use among patients may have impacted the rates of rheumatoid vasculitis prevalence. Future studies should examine temporal trends in smoking and pharmacotherapy in relation to rheumatoid vasculitis, and ongoing monitoring is merited.

Consistent with our hypothesis, a decrease in rheumatoid vasculitis prevalence was demonstrated among hospitalized and ambulatory veterans with RA, even after accounting for shifts away from hospitalization and toward management of the disease in outpatient settings. Rather than a gradual linear decline, a breakpoint emerged between 2000 and 2001 in both the hospital and outpatient settings, demonstrating timing similar to that described in a previous UK report (8). The consistency in timing provides grounds for querying causality. As RA treatment evolves, it will be important to continue to follow the evolving course of RA, including trends in rheumatoid vasculitis prevalence, among samples with a sex ratio that more closely reflects that found in the general population.

AUTHOR CONTRIBUTIONS

All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Bartels had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study conception and design. Bartels, Bell, Bridges.

Acquisition of data. Bartels.

Analysis and interpretation of data. Bartels, Bell, Rosenthal, Shinki, Bridges.

Ancillary