SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. REFERENCES

Objective

To quantify the incremental direct medical expenditure associated with rheumatoid arthritis (RA) in the US population from a payer's perspective.

Methods

A probability-weighted sample of adult respondents from the Medical Expenditure Panel Survey (2008) was used to identify a cohort of patients with RA and compared to a control cohort without RA. Annual expenditure outcomes, including total expenditure and subgroups related to pharmacy, office-based visits, emergency department visits, hospital inpatient stays, and residual expenditures were estimated. Differences between the RA and control cohort were adjusted for sociodemographic factors, employment status, insurance coverage, health behavior, and health status using a generalized linear model with log link and gamma distribution. Statistical inferences on difference in expenditures between RA and non-RA controls were based on nonparametric cluster bootstrapping using percentiles.

Results

The adjusted average annual total expenditure of the RA cohort in 2008 US dollars (USD) was $13,012 (95% confidence interval [95% CI] $1,737–$47,081), while that of the control cohort was $4,950 (95% CI $567–$17,425). The incremental total expenditure of the RA patients as compared to non-RA controls was $2,085 (95% CI $250–$7,822). RA patients also had a significantly higher pharmacy expenditure of $5,825 (95% CI $446–$30,998) that was on average $1,380 (95% CI $94–$7,492) higher as compared to the controls. The summated total incremental expenditure of all RA patients in the US was $22.3 billion (2008 USD).

Conclusion

RA exerts considerable incremental economic burden on US health care, which is primarily driven by the incremental pharmacy expenditure.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. REFERENCES

Rheumatoid arthritis (RA) presents an enormous economic burden on society in terms of the direct medical costs, the indirect costs that include lost wages and a caregiver's time, and the intangible costs of pain, fatigue, lowered self-esteem, or other psychological problems (1–7). Using data from the 2003 Medical Expenditure Panel Survey (MEPS), the Centers for Disease Control and Prevention estimated that arthritis and other rheumatic conditions collectively cost the US economy nearly $128 billion ($80.8 billion in direct and $47.0 billion in indirect costs, excluding pain and suffering) (8). This spending equaled approximately 1.2% of the US gross domestic product (GDP) in 2003 (8). Furthermore, in this cohort, the average per-person total direct expenditure in 2003 was $1,752, of which ambulatory care ($914) was the highest subcategory of expenditure, followed by emergency department and inpatient services ($352), prescriptions ($338), and other costs ($146) (8). Information about incremental economic burden of chronic diseases can inform health policies aimed at reducing disparities and unnecessary medical expenditure and quantify the excess economic burden of the disease borne by society. It can also help define and improve treatment and disease management guidelines.

Not much is explicitly known about the economic burden distinctively associated with RA, specific to the US population. In chronic conditions such as RA, primary cost drivers tend to be in demand for new health care technology, pharmacologic treatment, and surgical procedures. Prior to the introduction of biologic response- modifying agents, medications constituted the second largest component of RA-related costs, accounting for approximately 8–24% of total medical expenditure (9, 10). Among the medications, disease-modifying antirheumatic drugs (DMARDs) were responsible for approximately two-thirds of the total drug cost and nonsteroidal antiinflammatory drugs for most of the remainder (9). Recently, the introduction of the expensive biologic DMARDs in RA has more than doubled prescription spending, and the future expectation is that this expenditure will continue to increase (11–13). However, since current therapeutic options are not curative, one-third of RA patients may require surgery, although this rate may have decreased considerably in recent years for patients below age 60 years (1, 14). Escalation in pharmacy expenditures has increased the interest in the comparative analyses of biologic and traditional DMARDs (15–17).

Given these remarkable changes in the magnitude and distribution of spending in RA, it is important to understand the impact of direct medical expenditures attributable to RA (17). There are very few studies reporting total and subgroups of expenditures specific to RA based on data from recent years, and none of these studies have used samples that are nationally representative of the US population (18–24). There is no literature that can quantify the incremental expenditure of RA as compared to the US general population to inform the incremental economic burden presented by RA. Results from non-US studies have limited use to enlighten the economic impact of RA in the US due to significant differences in structure and financing of health care markets (20, 21).

The objective of this study was to evaluate the incremental total annual expenditure associated with RA based on a nationally representative sample from the MEPS. Secondary objectives were to assess the expenditures associated with pharmacy, office-based visits, emergency department visits, hospital inpatient stays, and residual expenditure, including hospital outpatient visits, ambulatory hospital visits, home health care, dental care, vision care, and other medical supplies and devices.

Significance & Innovations

  • This study quantifies the incremental expenditure of rheumatoid arthritis (RA) as compared to the US general population to inform the incremental economic burden presented by RA, using postbiologic era contemporary data from samples that are nationally representative of the US population.

  • Key findings from our study indicate that RA exerts a significant incremental economic burden of $22.3 billion (in 2008 US dollars) annually on US health care. Furthermore, in less than a decade, the primary driver of this incremental expenditure in RA has shifted from hospital expenditure to pharmacy expenditure.

PATIENTS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. REFERENCES

Study population.

The data used for the study were from the 2008 MEPS Household Component (MEPS-HC), a nationally representative survey of the US civilian noninstitutionalized population. The MEPS-HC uses the National Health Interview Survey conducted by the National Center for Health Statistics as its sampling frame. Data are collected at the person and household level in a series of 5 interviews over the course of 2.5 years, using computer-assisted personal interviews. At each interview, the survey collects information about each household member, and the survey builds on this information from interview to interview (25).

The MEPS-HC collects detailed data on demographic characteristics, health conditions, health status, use of medical care services, charges and payments, access to care, satisfaction with care, health insurance coverage, income, and employment. This data set is primarily designed to provide nationally representative data on the types of health care Americans use, frequency of use, amount paid for the services, and, specifically, who pays for what portion of these payments (25). Inferences based on the weighted MEPS data are representative of the entire US population.

Certain racial and/or ethnic populations, low-income families, and minorities are oversampled in the MEPS to improve precision of estimates when examining these groups for particular policy interest. Thus, all estimates from MEPS need to be adjusted with the weights provided on the public use files to adjust for sampling design and survey nonresponse (26). The current analysis accounted for the complex survey design's primary sampling units, strata, and person level sampling weights using Stata 11 survey data features (27).

Patient inclusion.

The Institutional Review Board of Kaiser Permanente Southern California approved the study. The study included persons ages 18–100 years with RA and other inflammatory polyarthropathies (referred to as RA from hereon) identified based on the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) code (714) or clinical classification code (202) observed in the MEPS medical conditions file (28). The MEPS medical conditions file provides information on all household-reported medical conditions for that year (29). The control group consisted of all eligible respondents who did not have RA.

Exclusions.

We excluded patients who had RA in combination with osteoarthritis (n = 18) or an unknown type of arthritis (n = 6). Furthermore, we also excluded those patients who had a diagnosis for RA in the medical conditions file, but who had self-reported not having been diagnosed with RA (n = 8). No exclusions were applied to the control cohort. Based on the inclusion/exclusion criteria, 585 adults with RA and 18,892 adults without RA were identified. The weighted sample of this study was representative of the approximately 198 million US civilian noninstitutionalized population.

Outcomes.

The primary outcome for the study was the direct total annual expenditure. It was defined as the expenditure incurred for all of the health services, including office-based visits, hospital outpatient visits, emergency room visits, inpatient hospital stays, dental visits, home health care, vision care, prescriptions, other medical supplies, and equipment expenditure. Additionally, important subgroups of expenditures, namely pharmacy, office-based visits, emergency department visits, inpatient hospital stays for at least 1 night, and residual expenditures (which included hospital outpatient visits, same day discharge from hospital visits, paid home health care, other medical supplies and devices, vision aids, and dental visits) between the RA and non-RA controls were also compared. All expenditures reported were in 2008 US dollars (USD).

Confounders adjusted.

The differences between RA patients and controls were adjusted for sociodemographics, health status, comorbidities, and health behavior attributes that could potentially confound the estimate of expenditure exclusive to RA. The sociodemographic covariates included age (18–24 years, 25–34 years, 35–44 years, 45–54 years, 55–64 years, and ≥65 years, with 18–24 years as the reference category), sex, race (white versus nonwhite), ethnicity (Hispanic versus non-Hispanic), marital status (married versus not married), family size (log transformed), education (high school graduate, some college, college graduate, some graduate school, and other degree, with less than high school as the reference category), income level (poor, near poor, low income, middle income, and high income, with low income as the reference category), employment status, and geographic (northeast versus not northeast) and metropolitan area location. Covariates for health behavior and health status included enrollment or the purchase of insurance coverage (private, public, or uninsured, with privately insured as the reference category), body mass index, current smoking status, and physical activity. Physical activity was defined based on a binary indicator if an individual spent half an hour or more in moderate to vigorous physical activity at least 3 times a week. A binary indicator defined obesity if body mass index was >30 kg/m2 ([weight in pounds × 703]/height in inches2). Additionally, based on the ICD-9-CM diagnosis codes reported in the medical conditions files, chronic condition indicators based on Charlson individual comorbidities for myocardial infarction, congestive heart failure, peripheral vascular disease, cerebrovascular disease, chronic pulmonary disease, dementia, diabetes mellitus with and without complications, paraplegia and hemiplegia, renal disease, cancer, acquired immunodeficiency syndrome/human immunodeficiency virus, peptic ulcer disease, and liver diseases were included to control for the influence of these comorbid conditions on expenditures (30).

Statistical analysis.

Expenditure data generally has non-negative values, high zero mass, nonconstant variance, distribution with longer right tail, and has higher peaks as compared to normal distribution. Zero mass refers to the high proportion of insurance enrollees who have zero or no expenditure during the year. Among those who have any expenditures, the distribution is right skewed. These 2 peculiarities in the distribution of expenditure outcomes are a result of the small proportion of patients who are responsible for a high proportion of total aggregate health care costs (31). These high-cost patients influence some of the parameters of interest to health policy and their data cannot be ignored or trimmed (31–35).

The use of ordinary least squares regression in such analyses may be biased and/or provide imprecise estimates of means and marginal/incremental effects (31, 35). Therefore, we employed a generalized linear regression model (GLM) with log link and gamma family distribution to estimate the expenditure function. Incremental annual expenditure was the difference in the annual medical expenditure between those who had RA and controls after adjusting for sociodemographics, health status, comorbidities, and the other covariates. The incremental expenditure was estimated by the method of recycled predictions to avoid the problem of reintroduction of covariate imbalance (36).

In the case of expenditure outcomes, hypothesis testing based on normality can also produce biased results. To avoid this issue, all statistical inferences were based on nonparametric clustered bootstrapped (1,000 repetitions) confidence intervals (CIs) using percentiles approach, preserving the MEPS survey weights. All statistical analyses accounted for the primary sampling units, strata, and sampling weights in the complex survey design.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. REFERENCES

The mean age of the RA patients was 61.0 years (95% CI 59.2–62.5 years), while that of the control group was 47.4 years (95% CI 46.9–48.0). Based on Table 1, the majority of the RA patients were female (61%), white (81%), married (55%), unemployed (62%), and residing in a metropolitan area (77%). As compared to the RA patients, the control group was more likely to be younger, employed, privately insured, physically active, and less likely to be obese. While 26.6% of the RA patients had only 1 comorbid condition and 24.0% had 2 or more comorbid conditions, in the control group, approximately 71.5% had no comorbid conditions, 22.1% had exactly 1 comorbidity, and 6.5% had 2 or more comorbid conditions.

Table 1. Descriptive statistics on the sociodemographic, health status, and expenditure variables*
VariablesRA (n = 5.8 million)Non-RA (n = 190 million)
  • *

    Values are the weighted percentage unless indicated otherwise. Data are from the 2008 Medical Expenditure Panel Survey's (MEPS) Household Component. RA = rheumatoid arthritis; GED = general equivalency diploma; HIV = human immunodeficiency virus; AIDS = acquired immunodeficiency syndrome; IQR = interquartile range.

  • The income categories are computed by dividing Current Population Survey (conducted by the Bureau of the Census) family income by the applicable poverty line (based on family size and composition) and reported as a percentage. The 5 income categories in MEPS are: negative or poor (less than 100%), near poor (100% to less than 125%), low income (125% to less than 200%), middle income (200% to less than 400%), and high income (greater than or equal to 400%).

  • Unemployed were respondents who were unemployed throughout the entire year.

  • §

    Physical activity was defined based on a binary indicator if an individual spent half an hour or more in moderate to vigorous physical activity at least 3 times a week.

  • A binary indicator defined obesity if body mass index was greater than 30 ([weight in pounds × 703]/ height in inches2).

  • #

    Includes all office-based visits to physicians, nonphysicians, chiropractors, nurse or nurse practitioner visits, optometrists, physician assistant visits, and physical or occupational therapist visits.

  • **

    Sum of all facility expense and separately billing provider's expenses associated with emergency department visits.

  • ††

    Includes facility and separately billed provider expenses for those with at least 1 night's stay in the hospital.

  • ‡‡

    Includes expenses on home health care (agency-sponsored and paid independent providers); dental visits (general dental visits and orthodontist visits); other medical supplies and equipment; vision aids; zero night's hospital stays (includes facility and separately billed provider expenses); and hospital-based outpatient visits to physicians and nonphysicians (facility and separately billed provider expenses).

Female61.153.8
Age, years  
 18–241.710.5
 25–344.017.4
 35–447.517.9
 45–5421.119.3
 55–6425.216.6
 ≥6540.518.3
Race/ethnicity  
 White81.382.6
 Hispanic11.612.1
Married54.755.9
Family size  
 127.220.6
 243.534.2
 3 or 423.434.1
 >45.911.1
Education  
 No degree23.914.6
 GED/high school diploma50.849.6
 Bachelor's degree12.718.0
 Master's/doctorate4.89.6
 Other degree7.88.2
Family income  
 Poor15.610.7
 Near poor8.44.0
 Low income15.312.9
 Middle income30.830.8
 High income29.941.6
Unemployed61.532.9
Reside in metropolitan statistical area76.884.1
Region  
 Northeast18.518.6
 Midwest22.322.6
 South38.135.9
 West21.123.0
Insurance coverage  
 Private53.970.7
 Public38.316.9
 Uninsured7.812.5
Health behavior  
 Physically active§39.257.1
 Obese40.431.4
 Smoker25.219.5
Chronic conditions  
 Myocardial infarction8.22.9
 Congestive heart failure3.51.0
 Peripheral vascular disease3.90.6
 Cerebrovascular disease5.92.3
 Dementia1.20.7
 Chronic pulmonary disease23.712.7
 Peptic ulcer disease1.70.5
 Liver diseases1.70.7
 Diabetes mellitus22.510.5
 Paraplegia/hemiplegia0.00.2
 Renal disease0.80.4
 Cancer10.74.9
 HIV/AIDS0.10.1
Annual expenditure outcomes, median (IQR) dollars  
 Total4,677 (1,628–12,165)1,229 (268–4,199)
 Pharmacy1,418 (357–3,750)126 (0–933)
 Office visits#720 (222–2,212)252 (20–919)
 Emergency department**0 (0–0)0 (0–0)
 Inpatient††0 (0–0)0 (0–0)
 Residual‡‡170 (0–1,155)84 (0–494)

Based on unadjusted analyses (Table 1), average expenditure of RA patients was nearly 1.5- to 3-fold higher than that of the non-RA control cohort in the various subgroups of expenditures. On average, as a proportion of total expenditure, pharmacy expenditure of RA patients was approximately 39.9% (29.1% in non-RA), office-based visits expenditure was 30.4% (35.3% in non-RA), emergency department expenditure was 2.3% (4.5% in non-RA), hospital inpatient expenditure was 9.4% (5.8% in non-RA), and residual expenditure was 18.0% (25.3% in non-RA patients).

After adjusting for sociodemographics, employment, insurance, health behavior, and 13 chronic conditions in the GLM, the incremental total expenditure was significantly higher in the RA patients as compared to the control group. Additionally, the incremental expenditure in all the subgroups of expenditure was also significantly higher in the RA patients as compared to the control group. The average total annual expenditure in the RA patients was $13,012 (95% CI $1,737–$47,081), while that in the control group was $4,950 (95% CI $567–$17,425) (Table 2). The additional/incremental annual total expenditure of the RA patients as compared to non-RA controls was $2,085 (95% CI $250–$7,822). RA patients also had significantly higher average pharmacy expenditure of $5,825 (95% CI $446–$30,998) that was on average approximately $1,380 (95% CI $94–$7,492) higher as compared to the controls. While the incremental office-based visits expenditure was higher by $587 (95% CI $83–$1,939) in RA patients, the incremental differences between inpatient and emergency department visits were $258 (95% CI $6–$1,208) and $13 (95% CI $5–$35), respectively. Based on the survey bootstrapped CIs, the incremental differences were statistically significant in all expenditure categories at an alpha level of 0.05.

Table 2. Adjusted average and incremental annual expenditures (in 2008 US dollars) in rheumatoid arthritis (RA) and non-RA control group*
Annual expenditure outcomesRA (n = 5.8 million)Non-RA (n = 190 million)Incremental difference
  • *

    Values are the mean (95% confidence interval). Confidence intervals are based on nonparametric percentiles from 1,000 cluster bootstrap estimates preserving the Medical Expenditure Panel Survey (MEPS) survey weights. Data are from the 2008 MEPS Household Component. All incremental differences are statistically significant at P ≤ 0.05. Adjusted for age, sex, race, ethnicity, marital status, family size, education, income level, employment status, geographic and metropolitan area, insurance coverage, body mass index, smoking status, physical activity, and 13 chronic conditions.

  • Includes all office-based visits (physicians, nonphysicians, chiropractors, nurse or nurse practitioner visits, optometrists, physician assistant visits, and physical or occupational therapist visits).

  • Sum of all facility expense and separately billing provider's expenses associated with emergency department visits.

  • §

    Includes facility and separately billed provider expenses for those with at least 1 night's stay in the hospital.

  • Includes expenses on home health care (agency-sponsored and paid independent providers); dental visits (general dental visits and orthodontist visits); other medical supplies and equipment; vision aids; zero night's hospital stays (includes facility and separately billed provider expenses); and hospital-based outpatient visits (physicians and nonphysicians [facility and separately billed provider]).

Total13,012 (1,737–47,081)4,950 (567–17,425)2,085 (250–7,822)
Pharmacy5,825 (446–30,998)1,264 (83–6,358)1,380 (94–7,492)
Office visits2,838 (464–9,510)1,221 (156–3,607)587 (83–1,939)
Emergency department279 (84–1,004)183 (70–492)13 (5–35)
Inpatient§5,021 (141–30,519)1,765 (47–8,461)258 (6–1,208)
Residual2,098 (226–7,276)1,036 (100–3,337)261 (27–934)

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. REFERENCES

The objective of this study was to evaluate the incremental direct annual medical expenditure associated with RA based on a nationally representative sample of the US population from the contemporary postbiologic DMARD period. A key finding from our study indicates that in the postbiologic era, the primary driver of the incremental total expenditure associated with RA has shifted from hospital stays to pharmacy expenditure. These findings are noteworthy since about a decade ago, the most important driver of direct costs in the US and internationally was hospitalization, especially in moderate and severe RA, while costs of medication represented only a minute proportion of these direct costs (37). This shift in the primary cost driver is very likely related to early aggressive DMARD treatment, as well as the introduction of expensive biologic DMARDs and their ever-increasing use in recent years. These findings further reinforce the need for evidence-based judicious prescribing of expensive DMARDs as proposed in the recent American College of Rheumatology and European League Against Rheumatism recommendations (15, 17, 38). Our results also quantify the enormous incremental economic burden posed by RA and can inform providers, decision makers, and payers how current dollars are being spent in the treatment of RA patients.

Consistent with the literature, the majority of the RA sample was comprised of women of an older age. Large majorities of these patients were unemployed, which may be due to the high prevalence of disability and functional limitations common in RA patients (39). Given the functional limitations posed by the disease, the RA patients also had a lower likelihood of being physically active, which in turn may have resulted in a higher proportion with obesity. Modifiable health risk factors such as smoking and obesity are responsible for a majority of the health care utilization and expenditure related to chronic diseases. It has been reported that the average US health care expenditure for people who were obese was $5,148 compared to $3,636 for those who were overweight and $3,315 for people who were normal weight (40).

While a majority of RA patients in our study had at least 1 comorbid condition, many of them had multiple severe comorbid conditions, which is similar to results reported by Sullivan et al (39). In contrast, only a small number of the control cohort had multiple comorbidities. The most prevalent chronic conditions in RA patients were chronic pulmonary disease, diabetes mellitus, and cancer, which seem consistent since all these conditions are associated with underlying inflammation. Evidence suggests that treating patients with multiple comorbid chronic conditions costs up to 7 times as much as treating patients with a single chronic condition (41).

Our results on total expenditure (Table 1) are comparable to those reported recently by Wolfe and Michaud (2009), where they estimated the out-of-pocket expenses and their burden on patients with RA from the National Data Bank for Rheumatic Diseases (NDB) (18). Depending on the level of out-of-pocket expense burden, the average total medical expenditure reported in that study ranged between $9,470 and $11,194 (2007 USD). However, the results on subgroups of expenditure reported in their study, especially for pharmacy and inpatient expenditure, were very different. Inpatient costs ranged between $900 and $1,599, while prescription spending was between $7,522 and $8,790 depending on the out-of-pocket expenses level. In our study, the average unadjusted inpatient expenditure was $2,791, while average prescription spending was $3,099 (both amounts deflated to 2007 USD). There are differences between the duration with disease, age, sex, ethnic, and educational characteristics of the US population as compared to the NDB population, and they may not be comparable (22).

The primary driver for the incremental expenditure in the RA patients was the additional pharmacy expenditure, which accounted for approximately 66% of the incremental total expenditure of $2,085 (95% CI $250–$7,822). Using data from MEPS over multiple years, Cisternas et al (42) and Yelin et al (8) have shown that the average prescription expenditure for patients with arthritis and other rheumatic conditions (including ICD-9 codes 274, 354, 390, 391, 443, 446, 710–716, 719–721, and 725–729) had nearly doubled since 1997; it is likely that RA in particular would be treated much more aggressively as compared to other forms of arthritis (8, 11, 12, 42). Moreover, comparing to trends reported by Cisternas et al (42) in contrast to arthritis and other rheumatic conditions, the average prescription expenditure of RA patients in our study was even higher than the average inpatient expenditure ($3,214 versus $2,894). These results highlight the need for additional data on comparative and cost effectiveness of DMARDs in RA that can reduce suboptimal or inefficient spending and reduce disparities. Currently, the majority of RA patients are treated with nonbiologic DMARDs, although the rate of use of biologic DMARDs has been increasing (39, 43–45). Since the cost of biologic therapy can range between $15,000–$25,000 USD per year depending on dose and other factors, it is important to identify subgroups of patients who will most likely benefit from these expensive treatments.

Based on our model, summing the total expenditure over the entirety of RA patients identified in the study, the total direct medical expenditure of RA on the US civilian population was $73.4 billion USD in 2008. To put into context, this expenditure was equivalent to 0.5% of the GDP of the US in 2008 and 6.5% of the approximately $1.14 trillion USD spent on health care in the US in 2008 by the civilian noninstitutionalized population. Additionally, the incremental or additional direct total expenditure of RA to the US was $22.3 billion USD annually or 0.16% of the GDP in 2008. These results indicate the enormous impact of RA on US society, where approximately 0.5–1.5% of the US population that has RA is consuming the equivalent of 0.5% of the GDP in just direct medical expenditure. The true societal burden of RA is much larger since we have not incorporated the impact of RA on out-of-pocket expenses, productivity, disability, limitations, health-related quality of life, and mortality.

Comparing the observed expenditures in RA patients to the overall arthritis and other rheumatic conditions reported by Yelin et al (8), it is evident that the per person average and incremental expenditure in RA is much higher, and as its subgroup, RA exerts a significant proportion of the net economic burden of arthritis and other rheumatic conditions on US health care. Yelin et al estimated that the aggregate incremental expenditures attributable to arthritis and other rheumatic conditions increased from $64.8 billion USD in 1997 to $80.8 billion USD in 2003 (8, 46). By inflation adjusting these estimates (8, 46) to 2008 USD using the medical component of the consumer price index, we get approximately $99 billion USD as the aggregate incremental expenditures attributable to arthritis and other rheumatic conditions. Ignoring the gains due to advances in health technology and the availability of new therapeutic treatments between 2003 and 2008, the incremental expenditure attributable to RA ($22.3 billion USD) seems to contribute approximately 22.5% of the entire incremental medical expenditure burden ($99 billion USD) attributable to arthritis and other rheumatic conditions. Given that the economic impact exerted by arthritis and other rheumatic conditions is similar to that of a moderate recession, the individual contribution of RA in this context is enormous (46).

These results should be interpreted with caution as they may be plagued similar to a prevalence-based cost of illness analysis. From an economic perspective, marginal costs (or cost effectiveness) of DMARD treatment are much more informative as compared to the incremental expenditure associated with RA. For example, even if a cure for RA were discovered, it does not necessarily imply that we would save the $22.3 billion USD additionally spent on RA patients. Essentially, the costs associated with the cure and remission would remain. However, our estimates of incremental expenditures associated with RA do inform rheumatologists and other providers, policy or decision makers, and payers how these dollars are currently spent in the treatment of RA patients.

There are potential limitations related to the retrospective observational study design. Since MEPS is based on self-report, prevalence estimates of RA and other chronic conditions may be misreported and this could vary systematically by cohort status, race, and ethnicity. In RA, disease progression and disease severity are important factors that can lead to heterogeneity of estimated costs, although our study could not control for these effects. We also study patients at a single point in time, whereas a longitudinal study could reduce the effects of unobserved confounding and provide insights to the long-term burden posed by this debilitating disease. We have only estimated the expenditure models using GLM with log link and gamma distribution; hence, changing the econometric models using different link functions could change the results. For example, although misspecification of the gamma distribution would only lead to efficiency losses, misspecification of the log link used in the study could produce biased results.

Finally, our study only evaluated the direct medical expenditure associated with RA, which only provides a lower bound on the true societal burden imposed by this disease. Due to its chronic nature, coupled with long-term disability occurring during a patient's most productive period, the indirect costs associated with RA are significant (47). Indirect costs such as productivity losses and time spent by family members and caregivers to support the patient can be 3 to 4 times higher than direct medical expenditure, and is a greater burden on society (47). Additionally, the intangible costs of pain, fatigue, lowered self-esteem, or other psychological problems imposed by RA are also significant economic burdens on society. Future studies should additionally incorporate productivity losses and health-related quality of life decrements experienced by RA patients in longitudinal nationally representative cohorts.

Despite these limitations, our results show that RA exerts substantial incremental economic burden on US health care. Furthermore, in less than a decade, the primary cost driver for RA has shifted from hospital expenditure to pharmacy expenditure. To reduce the impact of the incremental pharmacy expenditure in RA, future health policies should be designed to promote judicious use of the expensive biologic agents in the patients most likely to benefit from these agents and limit treatment only during the phases of the disease when the biologic agents can be most effective and are necessary.

AUTHOR CONTRIBUTIONS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. REFERENCES

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 submitted for publication. Dr. Kawatkar 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. Kawatkar, Venkatasubramaniam.

Acquisition of data. Kawatkar.

Analysis and interpretation of data. Kawatkar, Jacobsen, Levy, Medhekar, Herrinton.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. REFERENCES
  • 1
    Kvien TK. Epidemiology and burden of illness of rheumatoid arthritis. Pharmacoeconomics 2004; 22: 1–12.
  • 2
    Maetzel A, Li LC, Pencharz J, Tomlinson G, Bombardier C. The economic burden associated with osteoarthritis, rheumatoid arthritis, and hypertension: a comparative study. Ann Rheum Dis 2004; 63: 395401.
  • 3
    Jantti J, Aho K, Kaarela K, Kautiainen H. Work disability in an inception cohort of patients with seropositive rheumatoid arthritis: a 20-year study. Rheumatology (Oxford) 1999; 38: 113841.
  • 4
    Coyte P, Asche C, Croxford R, Chan B. The economic cost of arthritis and rheumatism in Canada. Toronto: Institute for Clinical Evaluative Sciences; 1998.
  • 5
    Backman CL. Employment and work disability in rheumatoid arthritis. Curr Opin Rheumatol 2004; 16: 14852.
  • 6
    Sokka T, Kautiainen H, Mottonen T, Hannonen P. Work disability in rheumatoid arthritis 10 years after the diagnosis. J Rheumatol 1999; 26: 16815.
  • 7
    Allaire S, Wolfe F, Niu J, LaValley M, Michaud K. Work disability and its economic effect on 55–64-year-old adults with rheumatoid arthritis. Arthritis Rheum 2005; 53: 6038.
  • 8
    Yelin E, Murphy L, Cisternas MG, Foreman AJ, Pasta DJ, Helmick CG. Medical care expenditures and earnings losses among persons with arthritis and other rheumatic conditions in 2003, and comparisons with 1997. Arthritis Rheum 2007; 56: 1397407.
  • 9
    Yelin E, Wanke LA. An assessment of the annual and long-term direct costs of rheumatoid arthritis: the impact of poor function and functional decline. Arthritis Rheum 1999; 42: 120918.
  • 10
    Cooper NJ. Economic burden of rheumatoid arthritis: a systematic review. Rheumatology (Oxford) 2000; 39: 2833.
  • 11
    Pucino F Jr, Harbus PT, Goldbach-Mansky R. Use of biologics in rheumatoid arthritis: where are we going? Am J Health Syst Pharm 2006; 63: S1941.
  • 12
    Kasmann NM, Power JD, Mamdani MM, Badley EM. Use of medication. In: Badley EM, Glazier RH, editors. Arthritis and related conditions in Ontario: ICES research atlas. 2nd ed. Toronto: Institute for Clinical Evaluative Sciences; 2004. p. 87104.
  • 13
    Yazici Y, Shi N, John A. Utilization of biologic agents in rheumatoid arthritis in the United States: analysis of prescribing patterns in 16,752 newly diagnosed patients and patients new to biologic therapy. Bull NY Hosp Jt Dis 2008; 66: 7785.
  • 14
    Louie GH, Ward MM. Changes in the rates of joint surgery among patients with rheumatoid arthritis in California, 1983–2007. Ann Rheum Dis 2010; 69: 86871.
  • 15
    Smolen JS, Landewe R, Breedveld FC, Dougados M, Emery P, Gaujoux-Viala C, et al. EULAR recommendations for the management of rheumatoid arthritis with synthetic and biological disease-modifying antirheumatic drugs. Ann Rheum Dis 2010; 69: 96475.
  • 16
    Van der Velde G, Pham B, Machado M, Ieraci L, Witteman W, Bombardier C, et al. Cost-effectiveness of biologic response modifiers compared to disease-modifying antirheumatic drugs for rheumatoid arthritis: a systematic review. Arthritis Care Res (Hoboken) 2011; 63: 6578.
  • 17
    Schoels M, Wong J, Scott DL, Zink A, Richards P, Landewe R, et al. Economic aspects of treatment options in rheumatoid arthritis: a systematic literature review informing the EULAR recommendations for the management of rheumatoid arthritis. Ann Rheum Dis 2010; 69: 9951003.
  • 18
    Wolfe F, Michaud K. Out-of-pocket expenses and their burden in patients with rheumatoid arthritis. Arthritis Rheum 2009; 61: 156370.
  • 19
    Wolfe F, Michaud K. The loss of health status in rheumatoid arthritis and the effect of biologic therapy: a longitudinal observational study. Arthritis Res Ther 2010; 12: R35.
  • 20
    Fautrel B, Clarke AE, Guillemin F, Adam V, St-Pierre Y, Panaritis T, et al. Costs of rheumatoid arthritis: new estimates from the human capital method and comparison to the willingness-to-pay method. Med Decis Making 2007; 27: 13850.
  • 21
    Jacobsson LT, Lindroth Y, Marsal L, Juran E, Bergstrom U, Kobelt G. Rheumatoid arthritis: what does it cost and what factors are driving those costs? Results of a survey in a community-derived population in Malmo, Sweden. Scand J Rheumatol 2007; 36: 17983.
  • 22
    Scott DL, Kingsley G. Clinical effectiveness of biologics in clinical practice. Arthritis Res Ther 2010; 12: 115.
  • 23
    Gabriel SE, Crowson CS, Campion ME, O'Fallon WM. Direct medical costs unique to people with arthritis. J Rheumatol 1997; 24: 71925.
  • 24
    Gabriel SE, Crowson CS, Campion ME, O'Fallon WM. Indirect and nonmedical costs among people with rheumatoid arthritis and osteoarthritis compared with nonarthritic controls. J Rheumatol 1997; 24: 438.
  • 25
    Agency for Healthcare and Research Quality. MEPS HC-121: 2008 full year consolidated data file. 2010. URL: http://www.meps.ahrq.gov/mepsweb/data_stats/download_data/pufs/h121/ h121doc.pdf.
  • 26
    Ezzati-Rice TM, Rohde F, Greenblatt J. Sample design of the medical expenditure panel survey household component, 1998–2007. Methodology Report No. 22. 2008. URL: http://www.meps.ahrq.gov/mepsweb/data_files/publications/mr22/mr22.pdf.
  • 27
    Stata statistical software: release 11. College Station (TX): Stata Corporation; 2009.
  • 28
    Agency for Healthcare and Research Quality. Healthcare cost and utilization project (HCUP). 2011. URL: http://www.hcup-us.ahrq.gov/toolssoftware/ccs/ccs.jsp.
  • 29
    Agency for Healthcare and Research Quality. MEPS HC-120: 2008 medical conditions. 2010. URL: http://www.meps.ahrq.gov/mepsweb/data_stats/download_data/pufs/h120/h120doc.pdf.
  • 30
    Li B, Evans D, Faris P, Dean S, Quan H. Risk adjustment performance of Charlson and Elixhauser comorbidities in ICD-9 and ICD-10 administrative databases. BMC Health Serv Res 2008; 8: 12.
  • 31
    Manning W. Dealing with skewed data on costs and expenditures. Cheltenham (UK): Edward Elgar Publishing; 2006.
  • 32
    Manning WG. The logged dependent variable, heteroscedasticity, and the retransformation problem. J Health Econ 1998; 17: 28395.
  • 33
    Manning WG, Basu A, Mullahy J. Generalized modeling approaches to risk adjustment of skewed outcomes data. J Health Econ 2005; 24: 46588.
  • 34
    Manning WG, Mullahy J. Estimating log models: to transform or not to transform? J Health Econ 2001; 20: 46194.
  • 35
    Partha D, Manning WG, Norton E. Modeling health care costs and counts. ASHE. 2010. Cornell University.
  • 36
    Greene W. Econometric analysis. 5th ed. Upper Saddle River (NJ): Prentice Hall; 2002.
  • 37
    Pugner KM, Scott DI, Holmes JW, Hieke K. The costs of rheumatoid arthritis: an international long-term view. Semin Arthritis Rheum 2000; 29: 30520.
  • 38
    Saag KG, Teng GG, Patkar NM, Anuntiyo J, Finney C, Curtis JR, et al. American College of Rheumatology 2008 recommendations for the use of nonbiologic and biologic disease-modifying antirheumatic drugs in rheumatoid arthritis. Arthritis Rheum 2008; 59: 76284.
  • 39
    Sullivan PW, Ghushchyan V, Huang XY, Globe DR. Influence of rheumatoid arthritis on employment, function, and productivity in a nationally representative sample in the United States. J Rheumatol 2010; 37: 5449.
  • 40
    Stagnitti M. Trends in health care expenditures by body mass index (BMI) category for adults in the US civilian noninstitutionalized population, 2001 and 2006. Rockville (MD): Agency for Healthcare Research and Quality; 2009.
  • 41
    Stanton M. The high concentration of US health care expenditures. Rockville (MD): Agency for Healthcare Research and Quality; 2006.
  • 42
    Cisternas MG, Murphy LB, Yelin EH, Foreman AJ, Pasta DJ, Helmick CG. Trends in medical care expenditures of US adults with arthritis and other rheumatic conditions 1997 to 2005. J Rheumatol 2009; 36: 25318.
  • 43
    Pressman Lovinger S. Use of biologics for rheumatoid arthritis tempered by concerns over safety, cost. JAMA 2003; 289: 322930.
  • 44
    Scott DL, Kingsley GH. Tumor necrosis factor inhibitors for rheumatoid arthritis. N Engl J Med 2006; 355: 70412.
  • 45
    Yazici Y, Shi N, John A. Utilization of biologic agents in rheumatoid arthritis in the United States: analysis of prescribing patterns in 16,752 newly diagnosed patients and patients new to biologic therapy. Bull NYU Hosp Jt Dis 2008; 66: 7785.
  • 46
    Yelin E, Cisternas MG, Pasta DJ, Trupin L, Murphy L, Helmick CG. Medical care expenditures and earnings losses of persons with arthritis and other rheumatic conditions in the United States in 1997: total and incremental estimates. Arthritis Rheum 2004; 50: 231726.
  • 47
    Wu E, Chen L, Birnbaum H, Yang E, Cifaldi M. Cost of care for patients with rheumatoid arthritis receiving TNF-antagonist therapy using claims data. Curr Med Res Opin 2007; 23: 174959.