Comments
Abstract
Rationale: There have been no recent comprehensive studies of the epidemiology of sarcoidosis in the United States. Changes in health care use have made available access to data on large numbers of patients with sarcoidosis.
Objectives: To use a U.S. national health care database to gather data on patients with sarcoidosis identified over a 3-year period who were 18 years of age and older, and to determine health care costs for these patients.
Methods: The Optum health care database was queried for a 3-year period (2010–2013). This database includes approximately 15% of U.S. residents. The incidence rate of sarcoidosis was calculated for new cases identified in each year. Calculation of prevalence was based on any patient with sarcoidosis seen during the year. Incidence and prevalence rates are reported per 100,000 patients.
Measurements and Main Results: A total of 29,372 adult patients with sarcoidosis were identified. Of these, 14,700 (55%) were over 55 years of age at the time of diagnosis. The incidence and prevalence rates were higher for African Americans (17.8 and 141.4 per 100,000, respectively) than for white individuals (8.1 and 49.8), Hispanics (4.3 and 21.7), or Asians (3.2 and 18.9). Women were two times more likely to have sarcoidosis, with the highest prevalence for sarcoidosis noted in African American women (178.5). Overall, the yearly health care cost reported for patients with sarcoidosis was low, with a median of $18,663 per year. However, the yearly cost for the top 5% was $93,201.
Conclusions: For patients 18 years of age and older enrolled in a U.S. national administrative database, sarcoidosis was more common among African Americans, but it was reported for all four of the major ethnic groups studied. While health care costs were relatively small for most patients, the cost of care for some patients was considerable.
Sarcoidosis is a worldwide disease of unknown etiology. The disease is characterized by the finding of granulomas in the lung and other organs in the absence of any associated foreign material or pathogenic microorganisms (1). In prior studies, the total number of cases in the United States has been estimated to be between 150,000 and 200,000 people (2, 3). In the United States, the disease was found to be more common among African Americans than among white individuals (3, 4) and more common in women than in men (3, 5, 6).
Early studies characterized sarcoidosis as a disease of young people, with most patients diagnosed before age 40 years (7). The onset of sarcoidosis was felt to be relatively uncommon in those over the age of 50 years (8). In more recent epidemiologic studies, researchers observed two peaks of incidence, one at ages 25–29 years and another at ages 65–69 years (9, 10). In recent epidemiologic studies, investigators found a shift in the peak incidence to ages 40–59 years in white individuals (11). In studies focused on American women, the incidence and prevalence were higher among African Americans than among white individuals, with increased rates in the Northeast (12, 13). However, since the mid-1990s, there has not been a systematic study of the epidemiology of sarcoidosis among all U.S. residents (3).
The reported fractional rate of U.S. patients with sarcoidosis who require therapy has ranged from 10 to 80% (14–18). Referral centers often report a higher rate of treatment. The most commonly used treatment remains prednisone (16, 17). However, alternatives to prednisone are available (19). These include the antimetabolites methotrexate and azathioprine, as well as biologic agents such as the anti–tumor necrosis factor (TNF) monoclonal antibody infliximab (20).
In 2013, Mitsubishi-Tanabe Pharma Development America contracted with OptumInsight to update and extend this information. OptumInsight is a business unit of Optum, which is the health services platform for UnitedHealth Group. OptumInsight provides software and information products, advisory consulting services, and business process outsourcing to participants in the health care industry.
In preparation for compiling this report, OptumInsight staff members selected and assembled information from Optum’s closed U.S. managed health care database, which contains detailed information regarding a well-defined portion of the U.S. population. This database is derived from a deidentified patient database sourced from a large, nationwide commercial health insurance and Medicare parts C and D database. The data are continuously updated. The database includes information on patients’ complete history, demographics, drug use, and health care use over fixed time periods. Therefore, it allows a realistic estimation of the phenotypic expression and clinical course, as well as specific information regarding the incidence and prevalence of various conditions, across the entire U.S. population for individuals who had health care insurance between 2009 and 2013 (21–23).
An analytical file of patients with sarcoidosis was created using Optum’s anonymized patient-level database, which represents approximately 32 million patients annually. We selected patients who filed medical claims between January 2009 and December 2013 with the 135.xx code of the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM). Patients included in the cohort were 18 years of age or older and enrolled in a health plan within a large national health insurer affiliated with Optum.
To exclude most patients for whom sarcoidosis was “ruled out,” we required subjects to have filed two or more medical claims at least 14 days apart using ICD-9-CM code 135.xx. Patient information collected included age, race, sex, and insurance type (commercial insurance plan vs. Medicare managed care plan). Treatment of sarcoidosis was captured from the database on the basis of information provided by health care providers treating patients diagnosed with sarcoidosis. Health care providers were encouraged but not required to indicate the condition for which any treatment was prescribed. The Optum database captures health care provider treatment decisions regarding sarcoidosis; however, the database does not indicate the specific motivation for treatment decisions made by health care providers.
All patients seen between 2010 and 2013 were characterized for each 1-year period. Each unique case was identified during these 1-year periods as “new” if the patient had at least 12 months of preeligibility and no sarcoidosis diagnosed in the 3 years prior to that year. Patients were considered “continuing” if they had filed one or more claims for sarcoidosis in the 3 years prior to eligibility. Patients who had less than 12 months of preeligibility and no claims for sarcoidosis in the previous 3 years were considered “undefined.” Calculation of the incidence of sarcoidosis was based on the number of new cases seen in the year. Calculation of the prevalence of sarcoidosis was based on the total number of cases seen in the individual year (new, continuing, and undefined).
The proportion of patients captured in the system was then extrapolated to the U.S. population, correcting for the age, race, and sex of the patients in the present study versus the corresponding U.S. census data. The sarcoidosis population, a subset of the Optum database, was segmented by patient type (new, continuing, or undefined). Once the patients were selected, Optum developed prevalence estimates for sarcoidosis by taking the patients with at least two medical claims for sarcoidosis during a calendar year and dividing that number by the total number of patients in the Optum database for that year.
The available data were segmented by year, age, sex, race, line of business, and state. The data were projected to the U.S. population aged 18 years and older. The incidence rate was multiplied by the total U.S. population for each year to project the number of newly diagnosed patients. Similarly, the prevalence rate was multiplied by the total U.S. population for each year to project the total number of patients with sarcoidosis. For estimates of the U.S. population, we did not correct for socioeconomic differences between the Optum database and the U.S. population.
The cost of care was determined using patients with both medical and pharmacy benefits. Cost-of-care categories included condition- and non–condition-related costs. We further segmented these two cost categories to better understand whether the cost of care was driven by treated or untreated patients with sarcoidosis. Across the treated and untreated groups, the individual cost categories were recorded. The cost was considered either related or not related to sarcoidosis on the basis of billing claims and was not further verified. We also gathered information about the cost per individual drug used to treat sarcoidosis.
Comparisons were made by using chi-square tests where indicated. A P value less than 0.05 was considered significant.
To create the database for analysis, patients with sarcoidosis were selected using the time frame of January 2009 through December 2013. The data from 2013 were assumed to be incomplete, since there may have been a lag in insurance claims. In summary, between 2009 and 2013, a total of 29,372 unique patients with sarcoidosis were identified who were older than 18 years of age and were coded as having sarcoidosis at least twice, 14 days or more apart.
The first diagnosis date was used as the index date. As shown in Table 1, the absolute numbers of total patients per year suggest a slight increase of about 1% annually. The number of undefined patients with sarcoidosis remained stable. The overall incidence of sarcoidosis in the United States per 100,000 insured patients was calculated per year, and we found it to be fairly stable during the investigated period (8.4, 8.3, 8.8, and 7.6 for 2010, 2011, 2012, and 2013, respectively). Consistent with these data were the analysis of the corresponding overall prevalence in the same period and the resulting incidence/prevalence ratio (Table 1).
| Patients with Sarcoidosis Aged ≥18 yr | 2010 | 2011 | 2012 | 2013* |
|---|---|---|---|---|
| New cases† | 1,620 (12.9%)‡ | 1,662 (13.8%)‡ | 1,858 (14.6%)‡ | 1,691 (12.8%)‡ |
| Continuing cases† | 8,297 | 8,697 | 9,150 | 9,751 |
| Undefined cases† | 2,152 | 2,397 | 2,172 | 2,101 |
| Total patients per year | 12,069 | 12,756 | 13,180 | 13,543 |
| Cases per 100,000 | ||||
| Incidence (I) | 8.4 | 8.3 | 8.8 | 7.6 |
| Prevalence (P) | 59.94 | 60.11 | 60.18 | 59.04 |
| I/P ratio | 0.14 | 0.14 | 0.15 | 0.13 |
The age distribution of the sarcoidosis population is shown in Table 2. We also show the age distribution of the U.S. Optum members. Among the patients with sarcoidosis, 17,316 (59%) were 55 years of age or older, while only 14,200,000 (42.5%) of the U.S. Optum members were 55 years of age or older (chi-square = 1,688.769; P < 0.0001). The distribution of patients between 55 and 64 years of age was highest in both absolute percentage of patients with sarcoidosis and percentage of the Optum population in the same age group (31.2% vs. 15.0%). In contrast, the distribution of patients between 18 and 34 years of age was lowest in both absolute percentage of patients with sarcoidosis and percentage of the Optum population in the same age group (3.5% vs. 25.1%). Table 2 also shows the age distribution for new cases seen during the study period (6,831 unique patients). The proportion of new cases was similar to the prevalent cases. Interestingly, 55% of patients were 55 years of age or older at the time of diagnosis.
| Age, yr | All Cases | New Cases | Optum Members | |||
|---|---|---|---|---|---|---|
| 18–34 | 1,014 | 3.5% | 342 | 5% | 8,400,000 | 25.1% |
| 35–44 | 3,506 | 11.9% | 956 | 14% | 5,200,000 | 15.6% |
| 45–54 | 7,536 | 25.7% | 1,776 | 26% | 5,600,000 | 16.8% |
| 55–64 | 9,162 | 31.2% | 1,913 | 28% | 5,000,000 | 15.0% |
| 65 or older | 8,154 | 27.8% | 1,844 | 27% | 9,200,000 | 27.5% |
| Total | 29,372 | 29,372 | 6,831 | 33,400,000 | ||
As expected, the incidence of sarcoidosis among African Americans was higher than the incidence among white individuals (17.8 vs. 8.1, respectively). The incidence of sarcoidosis in Hispanics was less than half that for white individuals but was higher than that for Asians (4.3 vs. 8.1 vs. 3.2, respectively) (Figure 1A). The prevalence rates followed the same pattern (Figure 1B). The differences in incidence between African Americans, white individuals, Hispanics, and Asians were all significant (Z = 7.9191, 9.0913, and 6.3213, respectively; P < 0.0001 for all comparisons). The incidence among white individuals was significantly higher than those for Hispanics (Z = 4.9573; P < 0.0001) and Asians (Z = 3.4350; P < 0.001). There was no significant difference between Hispanics and Asians (Z = 0.8576; P > 0.05).
Figure 1. The (A) incidence and (B) prevalence of sarcoidosis versus self-identified race. To be included in this study, patients must have been 18 years of age or older and engaged in the affiliated health plan, seen at least twice, been appropriately diagnosed, and been correctly coded. Rates were determined on the basis of all eligible patients in the Optum database.
[More] [Minimize]The differences in prevalence between African Americans and white individuals, Hispanics, and Asians were all significant (Z = 19.22.12, 20.7965, and 13.5208, respectively; P < 0.0001 for all comparisons). The prevalence for white individuals was significantly higher than those for Hispanics (Z = 10.7379; P < 0.0001) and Asians (Z = 6.219; P < 0.0001). There was no significant difference in prevalence between Hispanics and Asians (Z = 1.2215; P > 0.05).
To ensure that these observations were not impacted by health care coverage, we examined the interaction between health insurance and new versus continuing and undefined cases. Figures 2A and 2B show the distribution of new, continuing, and undefined cases versus commercial insurance and Medicare, respectively. There was no major difference between the white individuals and African Americans regarding commercial versus Medicare insurance. The number of cases may be too small to detect differences for the other racial or ethnic groups.
Figure 2. The distributions of new, continuing, and undefined cases versus (A) commercial insurance and (B) Medicare, respectively. There was no major difference between these groups for either commercial or Medicare insurance. Patients must have been 18 years of age or older and engaged in the affiliated health plan, seen at least twice, been appropriately diagnosed, and been correctly coded. Rates were determined on the basis of all eligible patients in the Optum database.
[More] [Minimize]The authors of previous reports have found that women are at higher risk for the sarcoidosis. We compared sex differences across racial and ethnic groups within the time frame investigated (Figure 3). We observed a higher prevalence for women than for men in all ethnic groups, and African American women had a higher prevalence than white women.
Figure 3. Prevalence of sarcoidosis versus race and sex. For all ethnic groups, women (red) had a higher prevalence than men (blue). African American women had the highest prevalence of disease. Patients must have been 18 years of age or older and engaged in the affiliated health plan, seen at least twice, been appropriately diagnosed, and been correctly coded. Rates were determined on the basis of all eligible patients in the Optum database.
[More] [Minimize]To investigate the impact of geography on sarcoidosis in the United States, we analyzed the prevalence among insured residents by region of the country. The Optum members in 2012 included 40.2 million people in four U.S. census regions: the Northeast (n = 11.1 million [27% of the total]), the South (14.6 million [36%]), the Midwest (8.1 million [20%]), and the West (6.5 million [16%]). In all four geographic regions, the estimated insured population prevalence remained relatively stable over time for all 4 years investigated. However, a striking difference in prevalence was observed between the West and the other U.S. geographic regions. This difference could not be explained by a different pattern between continuing, new, and undefined patients in the population analyzed. This observation suggests that the lower prevalence of sarcoidosis in the geographic West is based on reasons other than simple changes in population kinetics (Figure 4). The West has a significantly lower incidence and prevalence of disease than the Northeast, Midwest, or South (incidence chi-square = 27.912, P = 0.0010; prevalence chi-square = 17.915, P = 0392).
Figure 4. The (A) incidence and (B) prevalence of sarcoidosis per insured U.S. residence in 2012 versus geographic area. Patients for whom the region was not specified are not shown. Patients must have been 18 years of age or older and engaged in the affiliated health plan, seen at least twice, been appropriately diagnosed, and been correctly coded. Rates were determined on the basis of all eligible patients in the Optum database.
[More] [Minimize]Among those whose insurance status was unchanged during the course of the study, there were a total of 14,482 patients with sarcoidosis with commercial insurance versus 5,107 patients with Medicare insurance. Treatment for 4,381 newly diagnosed patients with sarcoidosis was determined by examining medications prescribed in the 12 months after the initial visit coded for sarcoidosis (index visit). Only 1,056 (24.1%) of these patients were treated within the first 12 months of diagnosis.
Table 3 summarizes the drugs most commonly prescribed specifically for sarcoidosis during the first 12 months after diagnosis. We include in this table only those drugs prescribed at a rate of 0.5% or more among the patient population. Treatments included glucocorticoids (prednisone), antimetabolites (methotrexate, azathioprine, mycophenolate mofetil, and leflunomide), biologics (infliximab, adalimumab, and rituximab), and others (hydroxychloroquine, colchicine, and pentoxifylline). The mean number of days until treatment was initiated with each drug among the new cases is also listed in Table 3. On average, all drugs were initiated within 4 months of the index diagnosis.
| New* | Continuing | Mean Days from First Sarcoidosis Diagnostic Code to First Treatment for New Cases | |
|---|---|---|---|
| Total number of cases† | 4,381 | 9,099 | |
| Number treated | 1,056 (24.1%) | 2,082 (22.8%) | |
| Glucocorticoids | |||
| Prednisone | 49.1% | 56.0% | 99.1 |
| Antimetabolites | 21.8% | 32.0% | |
| Methotrexate | 12.5% | 16.8% | 79.6 |
| Azathioprine | 5.4% | 8.5% | 83.9 |
| Mycophenolate mofetil | 3.0% | 4.9% | 81.7 |
| Leflunomide | 0.9% | 1.8% | 54.1 |
| Biologics | 2.6% | 5.9% | |
| Infliximab | 0.9% | 3.0% | 64.9 |
| Adalimumab | 0.8% | 1.9% | 46.8 |
| Rituximab | 0.9% | 1.0% | 130.0 |
| Others | 25.2% | 33.9% | |
| Hydroxychloroquine | 19.7% | 25.2% | 70.5 |
| Colchicine | 5.1% | 7.7% | 87.1 |
| Pentoxifylline | 0.4% | 1.0% | 106.8 |
Of the 9,099 continuing patients with 12 months of treatment after the index visit, 2,082 (22.8%) received one or more treatments. Table 3 summarizes the treatments used for these patients. Compared with the new cases, continuing patients were more likely to have more drugs prescribed. Combination therapy included an increased used of biologics.
Table 4 summarizes the overall cost of care for sarcoidosis-related and non–sarcoidosis-related health care per year. The cost was divided as follows: pharmacy, inpatient, office, laboratory, outpatient, emergency department, home health services, durable medical equipment, and other costs. For patients with sarcoidosis, the median cost of care was $18,663. However, the mean cost was over $32,000, since the cost of care for 25% of patients was over $36,000.
| Cost Category | Minimum | 25% | Median | 75% | Maximum | Mean |
|---|---|---|---|---|---|---|
| Condition related | ||||||
| Pharmacy | $0 | $7 | $30 | $97 | $16,857 | $255 |
| Inpatient (acute care) | $742 | $9,528 | $14,659 | $27,364 | $182,629 | $22,193 |
| Office | $0 | $168 | $342 | $621 | $54,811 | $557 |
| Laboratory | $0 | $32 | $76 | $167 | $4,842 | $141 |
| Outpatient | $0 | $190 | $604 | $1,461 | $166,123 | $1,588 |
| Emergency | $12 | $108 | $420 | $840 | $12,726 | $796 |
| Home health | $0 | $266 | $1,169 | $2,640 | $119,150 | $2,444 |
| DME | $16 | $312 | $1,238 | $2,503 | $6,259 | $1,425 |
| Other | $0 | $45 | $125 | $376 | $458,341 | $2,633 |
| Total cost | $770 | $10,656 | $18,663 | $36,069 | $1,021,738 | $32,032 |
| Non–condition related | ||||||
| Pharmacy | $0 | $356 | $1,294 | $3,478 | $35,993 | $2,840 |
| Inpatient (acute care) | $0 | $7,513 | $14,812 | $32,311 | $235,708 | $27,573 |
| Office | $3 | $784 | $1,644 | $3,198 | $204,562 | $2,853 |
| Laboratory | $0 | $105 | $235 | $511 | $69,046 | $484 |
| Outpatient | $0 | $708 | $2,706 | $7,453 | $753,700 | $7,677 |
| Emergency | $0 | $433 | $978 | $2,602 | $302,048 | $2,594 |
| Home health | $0 | $258 | $949 | $2,765 | $621,687 | $3,150 |
| DME | $0 | $238 | $760 | $2,261 | $33,788 | $1,435 |
| Other | $0 | $151 | $489 | $2,521 | $1,065,033 | $6,745 |
| Total cost | $3 | $10,546 | $23,867 | $57,100 | $3,321,565 | $55,351 |
We also calculated the cost of care for the top 5% of patients seen over the 4-year period of the study. This represents approximately 2,600 patients. Table 5 gives the breakdown of average yearly costs directly related to sarcoidosis care and costs for non–sarcoidosis-related health issues. The total cost for this 5% of patients was over $240 million. The cost per patient per year in that group was $93,201.
| Cost Category | Sarcoidosis Related | Not Sarcoidosis Related |
|---|---|---|
| Inpatient hospital | $51,584,563.02 | $43,855,949.01 |
| Inpatient | $36,505,186.91 | $35,744,535.75 |
| Office | $19,882,945.67 | $25,731,947.92 |
| Pharmacy | $5,624,261.50 | $3,619,992.91 |
| Home | $2,272,588.62 | $2,653,214.34 |
| Skilled nursing facility | $2,030,747.73 | $2,643,810.20 |
| Emergency room | $1,326,567.21 | $1,747,919.66 |
| Other unlisted facility | $977,725.72 | $756,796.38 |
| Inpatient psychiatric facility | $14,224.50 | $3,341.38 |
| Outpatient | ||
| Ambulance, land | $550,514.30 | $530,869.95 |
| DME | $422,706.31 | $435,505.54 |
| End-stage renal disease treatment facility | $352,459.09 | $692,459.79 |
| Independent laboratory | $331,182.93 | $332,842.87 |
| Residential substance abuse treatment facility | $196,535.83 | $70,475.25 |
| Ambulatory surgical center | $184,369.98 | $583,485.03 |
| Hospice | $112,952.44 | $138,759.88 |
| Nursing facility | $51,315.64 | $49,417.74 |
| Ambulance, air or water | $13,589.98 | $263,537.50 |
| Comprehensive inpatient rehabilitation facility | $12,089.52 | $10,556.00 |
| Independent clinic | $2,856.50 | $2,612.32 |
| Assisted living facility | $2,344.36 | $2,004.43 |
| Community mental health center | $607.26 | |
| Birthing center | $326.42 | $113.10 |
| Custodial care facility | $156.02 | $135.72 |
| Military treatment center | $103.24 | |
| Mass immunization center | $54.52 | |
| State or local public health clinic | $38.28 | |
| Mobile unit | $40.60 | |
| Federally qualified health center | $151.38 | |
| Pharmacy | $1,570.06 | |
| Total cost | $122,452,975.22 | $119,872,115.47 |
| Cost per patient | $47,097.30 | $46,104.66 |
The predominance of sarcoidosis in women and African Americans has remained a significant feature of the disease (3). However, our data suggest that sarcoidosis occurs in an older population than originally thought. In fact, more than half of the new cases were diagnosed in patients who were 55 years of age or older. We did not calculate the age at diagnosis by race or by sex. This shift in the peak age of sarcoidosis was noted in a longitudinal study of sarcoidosis spanning 60 years of observation (11, 24).
In the present study, we identified over 29,000 adult patients with sarcoidosis. This included over 2,000 new patients every year. We found an overall incidence and prevalence of sarcoidosis of 8 and 60 cases per 100,000, respectively. These findings are similar to data reported by others (3, 11–13). However, our study included a uniquely large database. On the basis of U.S. census data, we estimated that there are over 25,000 cases of sarcoidosis diagnosed in the United States each year. In a given year, approximately 185,000 patients with sarcoidosis seek medical care. It appears that some patients stop seeking care if they are in clinical remission. Care is stopped in only a small number of patients due to death, since the mortality of sarcoidosis in sarcoidosis clinics is about 5% (25). The authors of a recent analysis of death certificate data found an age-adjusted mortality of sarcoidosis of 0.28–0.43 per 100,000 (26, 27).
The incidence and prevalence of disease in United States is strongly influenced by sex and self-identified race. The increased rate for women versus men has been reported for African Americans (3); however, we believe that this is the first report of incidence and prevalence for Hispanics in America. African Americans had much higher incidence and prevalence rates than white individuals. The prevalence for African Americans was 141.6 per 100,000 versus 49.8 per 100,000 for white individuals (Figure 3B). The highest prevalence was for African American women, at 178 per 100,000. This has been noted in prior studies performed in the United States (3, 13, 28). Hispanic and Asian patients were less likely to have sarcoidosis than white individuals, which has been suggested in the past (4). However, in this study, we were able to identify sarcoidosis cases in sufficient numbers in both of these groups to be able to compare them with the larger populations of African Americans and white individuals. The overall incidence among Hispanics and Asians was approximately half that of the white population (Figure 3).
In the present study, we found a difference in the number of sarcoidosis cases in different regions of the country. These differences may be a reflection of the lower percentage of African Americans in the West. In one study of African American women, the incidence and prevalence of disease were similar for all regions of the United States (12). In another study, female nurses had the highest rates in the Northeast and the lowest rates in the West (13).
Previous studies have suggested that health insurance and socioeconomic status may influence the diagnosis and outcome of sarcoidosis (29). In a study done in one southeastern state, researchers found higher rates of sarcoidosis in those counties close to the Atlantic Ocean than inland (30). We did not perform a multivariate analysis of these various factors to explain the difference in rates of sarcoidosis for different areas of the United States. Future studies taking these factors into account seem warranted.
About one-fourth of patients with sarcoidosis in our study received treatment for the disease. In contrast, in most previously published U.S. studies, researchers have reported that about half of patients were on therapy (12, 16, 17). However, most studies have been done with data from sarcoidosis referral centers, which may have introduced a bias toward including patients with more severe disease (16, 17), or from patients completing questionnaires, perhaps introducing bias toward sicker patients (12). In one study of treatment of sarcoidosis in a non–referral center, researchers found a percentage of patients being treated for disease similar to that in our present study (14). Since glucocorticoids do not appear to change the natural course of asymptomatic disease (31), the lower percentage of treated patients in this study may be a reflection of the identification in our study of asymptomatic patients with sarcoidosis.
Interestingly, initiation of therapy for sarcoidosis averaged 3 months from the time the diagnosis of sarcoidosis was first made. Since the diagnosis of sarcoidosis may have been only part of a differential diagnosis, this delay may reflect a lag in the time to confirmation of the sarcoidosis diagnosis. Patients who eventually receive a diagnosis of sarcoidosis often have been visiting a health care provider for months until a specific diagnosis is made (32).
The most commonly used treatment for sarcoidosis in this study was prednisone. This is consistent with evidence-based recommendations for the use of glucocorticoids as first-line therapy for symptomatic disease (31). Of the antimetabolites, methotrexate was the most commonly used agent. This agent has been reported as the preferred steroid-sparing agent among pulmonary physicians (20). Azathioprine and mycophenolate mofetil were occasionally used, though there have been fewer studies on the effectiveness of these drugs for sarcoidosis (19, 33–35). Of the biologic agents, the largest studies that have been published support the use of infliximab (19, 36). However, there have been some reports of the effectiveness of adalimumab (37) and rituximab (38).
This study provides unique information regarding the yearly cost of care of patients with sarcoidosis. Treatment was prescribed for less than one-fourth of patients. The sarcoidosis-associated costs were not only for medications but also for evaluation and monitoring. Additional costs for non–condition-related issues may still have been related to the cost of sarcoidosis care. This would include complications of prednisone use, such as diabetes and osteoporosis. In addition, immunosuppressive therapy and underlying disease may have led to infections and increased health care costs (39). For the 5% of highest-cost patients, treatment options can become expensive. In this study, the average for this group exceeded $90,000 per year. This includes the use of anti-TNF agents, which are not only expensive but also may require costly supportive care (40). A significant proportion of patients with chronic dyspnea have sarcoidosis-associated pulmonary hypertension (41). Care for these patients may include costly treatments for pulmonary hypertension (42, 43). Lung transplant has been a successful option for some patients with refractory pulmonary disease (44).
In the present study, we examined patients who were part of a specific health care plan. Although there was no significant difference between those with commercial insurance and those with Medicare, we do not know whether our findings are applicable to uninsured patients or those covered by Medicaid. Another limitation of this study is that it is based on an administrative claims database. This means we were not able to capture the reason for one treatment or another.
The rate of prednisone use may have been underestimated in this study, since its use for sarcoidosis was tied to the diagnosis of sarcoidosis. Patients who had previously encountered the drug and had intolerance for the medication would not be noted. Also, a patient may have begun prednisone for an alternative diagnosis, such as asthma, and that diagnosis may not have been changed in the system subsequently. Also, the database provides only information on whether a prescription was filled. Longitudinal medication adherence information was not available.
The identification of sarcoidosis was based on diagnostic claims. We required that there be at least two claims for sarcoidosis for a patient to be included in the study. On one hand, we may have missed patients who were seen only once for their sarcoidosis. On the other hand, we did not confirm the diagnosis on the basis of biopsy or other criteria (1). This may represent overdiagnosis of the disease.
Patients in the Optum database were enrolled in one of two different insurance plans: commercial insurance or Medicare. While there appeared to be no major differences in rates of disease for white individuals and African Americans for the two different plans, there may have been some small differences, especially for the smaller numbers of cases in the other ethnic groups. Since the population in the Optum database is older than the general population of the United States, we may have oversampled this group in determining the incidence and prevalence of sarcoidosis. Also, we did not correct for differences in socioeconomic status of those in the Optum database versus the U.S. population to calculate the rate of disease. Since patients may enter and leave the Optum database over time, the denominator could not be precisely defined.
Using a robust, nationwide administrative database, we found that the prevalence of sarcoidosis in the United States is approximately 60 per 100,000 adults. The highest incidence and prevalence rates are among women and African Americans. Contrary to prior studies, the average age of newly diagnosed patients with sarcoidosis was older than 55 years. While many patients do not require treatment for sarcoidosis, for those who are treated, the yearly cost of care can be considerable.
| 1 . | Hunninghake GW, Costabel U, Ando M, Baughman R, Cordier JF, du Bois R, Eklund A, Kitaichi M, Lynch J, Rizzato G, et al. ATS/ERS/WASOG statement on sarcoidosis. American Thoracic Society/European Respiratory Society/World Association of Sarcoidosis and Other Granulomatous Disorders. Sarcoidosis Vasc Diffuse Lung Dis 1999;16:149–173. |
| 2 . | Denning DW, Pleuvry A, Cole DC. Global burden of chronic pulmonary aspergillosis complicating sarcoidosis. Eur Respir J 2013;41:621–626. |
| 3 . | Rybicki BA, Major M, Popovich J Jr, Maliarik MJ, Iannuzzi MC. Racial differences in sarcoidosis incidence: a 5-year study in a health maintenance organization. Am J Epidemiol 1997;145:234–241. |
| 4 . | Izumi T. Symposium: Population differences in clinical features and prognosis of sarcoidosis throughout the world. Sarcoidosis 1992;9:S105–S118. |
| 5 . | Handa T, Nagai S, Miki S, Fushimi Y, Ohta K, Mishima M, Izumi T. Incidence of pulmonary hypertension and its clinical relevance in patients with sarcoidosis. Chest 2006;129:1246–1252. |
| 6 . | Baughman RP, Teirstein AS, Judson MA, Rossman MD, Yeager H Jr, Bresnitz EA, DePalo L, Hunninghake G, Iannuzzi MC, Johns CJ, et al.; Case Control Etiologic Study of Sarcoidosis (ACCESS) research group. Clinical characteristics of patients in a case control study of sarcoidosis. Am J Respir Crit Care Med 2001;164:1885–1889. |
| 7 . | Sartwell PE, Edwards LB. Epidemiology of sarcoidosis in the U.S. Navy. Am J Epidemiol 1974;99:250–257. |
| 8 . | Lenner R, Schilero GJ, Padilla ML, Teirstein AS. Sarcoidosis presenting in patients older than 50 years. Sarcoidosis Vasc Diffuse Lung Dis 2002;19:143–147. |
| 9 . | Hillerdal G, Nöu E, Osterman K, Schmekel B. Sarcoidosis: epidemiology and prognosis: a 15-year European study. Am Rev Respir Dis 1984;130:29–32. |
| 10 . | Byg KE, Milman N, Hansen S. Sarcoidosis in Denmark 1980-1994: a registry-based incidence study comprising 5536 patients. Sarcoidosis Vasc Diffuse Lung Dis 2003;20:46–52. |
| 11 . | Ungprasert P, Carmona EM, Utz JP, Ryu JH, Crowson CS, Matteson EL. Epidemiology of sarcoidosis 1946-2013: a population-based study. Mayo Clin Proc 2016;91:183–188. |
| 12 . | Cozier YC, Berman JS, Palmer JR, Boggs DA, Serlin DM, Rosenberg L. Sarcoidosis in black women in the United States: data from the Black Women’s Health Study. Chest 2011;139:144–150. |
| 13 . | Dumas O, Abramovitz L, Wiley AS, Cozier YC, Camargo CA Jr. Epidemiology of sarcoidosis in a prospective cohort study of U.S. women. Ann Am Thorac Soc 2016;13:67–71. |
| 14 . | Reich JM, Johnson RE. Course and prognosis of sarcoidosis in a nonreferral setting: analysis of 86 patients observed for 10 years. Am J Med 1985;78:61–67. |
| 15 . | Hunninghake GW, Gilbert S, Pueringer R, Dayton C, Floerchinger C, Helmers R, Merchant R, Wilson J, Galvin J, Schwartz D. Outcome of the treatment for sarcoidosis. Am J Respir Crit Care Med 1994;149:893–898. |
| 16 . | Baughman RP, Judson MA, Teirstein A, Yeager H, Rossman M, Knatterud GL, Thompson B. Presenting characteristics as predictors of duration of treatment in sarcoidosis. QJM 2006;99:307–315. |
| 17 . | Gottlieb JE, Israel HL, Steiner RM, Triolo J, Patrick H. Outcome in sarcoidosis: the relationship of relapse to corticosteroid therapy. Chest 1997;111:623–631. |
| 18 . | Judson MA, Boan AD, Lackland DT. The clinical course of sarcoidosis: presentation, diagnosis, and treatment in a large white and black cohort in the United States. Sarcoidosis Vasc Diffuse Lung Dis 2012;29:119–127. |
| 19 . | Baughman RP, Nunes H, Sweiss NJ, Lower EE. Established and experimental medical therapy of pulmonary sarcoidosis. Eur Respir J 2013;41:1424–1438. |
| 20 . | Schutt AC, Bullington WM, Judson MA. Pharmacotherapy for pulmonary sarcoidosis: a Delphi consensus study. Respir Med 2010;104:717–723. |
| 21 . | Byfield SA, McPheeters JT, Burton TM, Nagar SP, Hackshaw MD. Persistence and compliance among U.S. patients receiving pazopanib or sunitinib as first-line therapy for advanced renal cell carcinoma: a retrospective claims analysis. J Manag Care Spec Pharm 2015;21:515–522. |
| 22 . | Voss EA, Ma Q, Ryan PB. The impact of standardizing the definition of visits on the consistency of multi-database observational health research. BMC Med Res Methodol 2015;15:13. |
| 23 . | Rodriguez F, Olufade T, Heithoff K, Friedman HS, Navaratnam P, Foody JM. Frequency of high-risk patients not receiving high-potency statin (from a large managed care database). Am J Cardiol 2015;115:190–195. |
| 24 . | Sawahata M, Sugiyama Y, Nakamura Y, Nakayama M, Mato N, Yamasawa H, Bando M. Age-related and historical changes in the clinical characteristics of sarcoidosis in Japan. Respir Med 2015;109:272–278. |
| 25 . | Baughman RP, Winget DB, Bowen EH, Lower EE. Predicting respiratory failure in sarcoidosis patients. Sarcoidosis Vasc Diffuse Lung Dis 1997;14:154–158. |
| 26 . | Swigris JJ, Olson AL, Huie TJ, Fernandez-Perez ER, Solomon J, Sprunger D, Brown KK. Sarcoidosis-related mortality in the United States from 1988 to 2007. Am J Respir Crit Care Med 2011;183:1524–1530. |
| 27 . | Mirsaeidi M, Machado RF, Schraufnagel D, Sweiss NJ, Baughman RP. Racial difference in sarcoidosis mortality in the United States. Chest 2015;147:438–449. |
| 28 . | Keller AZ. Hospital, age, racial, occupational, geographical, clinical and survivorship characteristics in the epidemiology of sarcoidosis. Am J Epidemiol 1971;94:222–230. |
| 29 . | Rabin DL, Thompson B, Brown KM, Judson MA, Huang X, Lackland DT, Knatterud GL, Yeager H Jr, Rose C, Steimel J. Sarcoidosis: social predictors of severity at presentation. Eur Respir J 2004;24:601–608. |
| 30 . | Buxbaum L, Lackland D, Judson M, Hoel D, Mohr L. Geographic patterns of pulmonary disease in South Carolina. Ann Epidemiol 2000;10:460–461. |
| 31 . | Baughman RP, Nunes H. Therapy for sarcoidosis: evidence-based recommendations. Expert Rev Clin Immunol 2012;8:95–103. |
| 32 . | Judson MA, Thompson BW, Rabin DL, Steimel J, Knattereud GL, Lackland DT, Rose C, Rand CS, Baughman RP, Teirstein AS; ACCESS Research Group. The diagnostic pathway to sarcoidosis. Chest 2003;123:406–412. |
| 33 . | Hamzeh N, Voelker A, Forssén A, Gottschall EB, Rose C, Mroz P, Maier LA. Efficacy of mycophenolate mofetil in sarcoidosis. Respir Med 2014;108:1663–1669. |
| 34 . | Vorselaars AD, Wuyts WA, Vorselaars VM, Zanen P, Deneer VH, Veltkamp M, Thomeer M, van Moorsel CH, Grutters JC. Methotrexate vs azathioprine in second-line therapy of sarcoidosis. Chest 2013;144:805–812. |
| 35 . | Müller-Quernheim J, Kienast K, Held M, Pfeifer S, Costabel U. Treatment of chronic sarcoidosis with an azathioprine/prednisolone regimen. Eur Respir J 1999;14:1117–1122. |
| 36 . | Baughman RP, Drent M, Kavuru M, Judson MA, Costabel U, du Bois R, Albera C, Brutsche M, Davis G, Donohue JF, et al.; Sarcoidosis Investigators. Infliximab therapy in patients with chronic sarcoidosis and pulmonary involvement. Am J Respir Crit Care Med 2006;174:795–802. |
| 37 . | Sweiss NJ, Noth I, Mirsaeidi M, Zhang W, Naureckas ET, Hogarth DK, Strek M, Caligiuri P, Machado RF, Niewold TB, et al. Efficacy results of a 52-week trial of adalimumab in the treatment of refractory sarcoidosis. Sarcoidosis Vasc Diffuse Lung Dis 2014;31:46–54. |
| 38 . | Sweiss NJ, Lower EE, Mirsaeidi M, Dudek S, Garcia JG, Perkins D, Finn PW, Baughman RP. Rituximab in the treatment of refractory pulmonary sarcoidosis. Eur Respir J 2014;43:1525–1528. |
| 39 . | Baughman RP, Lower EE. Frequency of acute worsening events in fibrotic pulmonary sarcoidosis patients. Respir Med 2013;107:2009–2013. |
| 40 . | Drent M, Cremers JP, Jansen TL, Baughman RP. Practical eminence and experience-based recommendations for use of TNF-α inhibitors in sarcoidosis. Sarcoidosis Vasc Diffuse Lung Dis 2014;31:91–107. |
| 41 . | Baughman RP, Engel PJ, Taylor L, Lower EE. Survival in sarcoidosis-associated pulmonary hypertension: the importance of hemodynamic evaluation. Chest 2010;138:1078–1085. |
| 42 . | Baughman RP, Culver DA, Cordova FC, Padilla M, Gibson KF, Lower EE, Engel PJ. Bosentan for sarcoidosis-associated pulmonary hypertension: a double-blind placebo controlled randomized trial. Chest 2014;145:810–817. |
| 43 . | Barnett CF, Bonura EJ, Nathan SD, Ahmad S, Shlobin OA, Osei K, Zaiman AL, Hassoun PM, Moller DR, Barnett SD, et al. Treatment of sarcoidosis-associated pulmonary hypertension: a two-center experience. Chest 2009;135:1455–1461. |
| 44 . | Shorr AF, Davies DB, Nathan SD. Predicting mortality in patients with sarcoidosis awaiting lung transplantation. Chest 2003;124:922–928. |
Supported by Mitsubishi-Tanabe Pharma Development America.
Author Contributions: R.P.B.: wrote the first draft of the manuscript; all other authors provided input on and revised the first and other drafts of the manuscript; S.F.: was responsible for data compilation and created the tables and figures; R.P.B., M.D., and G.W.: revised the tables and figures based on the desired analysis; and S.F.: was contracted by Mitsubishi-Tanabe Pharma Development America and the University of Cincinnati to provide data for analysis. All authors contributed to data interpretation and the preparation and writing of the final manuscript. All authors reviewed and contributed to the manuscript.
Author disclosures are available with the text of this article at www.atsjournals.org.
