Incidence and prevalence of adult systemic lupus erythematosus in a large US managed-care population

Abstract

Objective

The objective of this paper is to determine the incidence and prevalence of adult systemic lupus erythematosus (SLE) in a large US managed-care population.

Methods

Subject inclusion in the incidence cohort required a medical claim with an SLE diagnosis and a service date from 2003 to 2008 that satisfied the following criteria: 1) ≥18 years on service date; 2) continuously enrolled for 24 months before and 12 months after service date; 3) in the 12 months after service date, ≥ one inpatient claim or ≥ two office or ER visits with an SLE diagnosis; 4) no SLE diagnosis 24 months prior to service date; and 5) no SLE medications 12 months prior to service date. Prevalence cohort subjects were identified using a similar algorithm and were not required to satisfy criteria 4) and 5).

Results

A total of 1,557 subjects were included in the incidence cohort, and 15,396 were included in the prevalence cohort. The overall age- and gender-adjusted SLE incidence rate (2003–2008) was 7.22 cases per 100,000 person-years. The annual prevalence of SLE (per 100,000 individuals) varied from 81.07 in 2003 to 102.94 in 2008.

Conclusion

The SLE incidence in this large managed-care plan with geographic diversity was slightly higher than previous estimates, and the prevalence was within the range of previous estimates.

Keywords

Epidemiology incidence prevalence systemic lupus erythematosus

Introduction

Systemic lupus erythematosus (SLE) is a chronic autoimmune disease with symptoms including skin rashes, photosensitivity, oral ulcers, arthritis and fever.^1,2 Subjects with SLE may have involvement of the cardiovascular system, the hematologic system, the nervous system, or renal system.^2,3 The latter, lupus nephritis, is observed in about 30% of subjects with SLE and is a major cause of mortality.^4,5 Subjects typically experience periods of remission and periods of flares, and the 10-year survival probability for subjects with SLE is currently about 90%.⁶ Mortality among subjects with SLE has decreased considerably in recent decades⁷ in part because of better management with pharmacotherapy or early diagnosis and intervention. Therapies for SLE currently include immunosuppressive drugs, antimalarials and corticosteroids. Even with therapy, quality of life assessments show decreased physical, social and emotional functioning over time among SLE subjects.⁸ Females are more commonly affected by SLE than males, and African-Americans and Asians appear to be more susceptible to developing SLE than Caucasians.^9–11 Although individuals of all ages are at risk for SLE, disease onset (incidence) appears to be most common among individuals aged 15–44.¹²

Previous estimates of SLE incidence in the US have ranged from 1.5 to 5.6 per 100,000 (1950–2004), and previous estimates of SLE prevalence in the US have ranged from 41.8 to 241 cases per 100,000 (1989–2004).^13–22 Most of these studies of SLE epidemiology in the US have been limited to specific geographic regions, and little information exists on the prevalence or incidence of SLE in the US over the last decade. In the current study, we examined the prevalence of adult SLE and incidence of adult SLE, lupus nephritis and central nervous system (CNS) lupus using a large managed-care plan claims database with diverse geographic distribution to better understand the epidemiology of SLE in the US.

Patients and methods

Study design and subject identification

This study used a proprietary managed-care organization (MCO) claims database affiliated with OptumInsight. Eligible subjects were enrollees consisting of approximately 35 million commercially insured members with medical and pharmacy benefits during the study period from 2003 to 2008 (i.e. approximately 14 million total lives were covered in 2008). The study data contain demographic information and enrollee medical claims information submitted by individual providers for ambulatory services (using the Health Care Financing Administration (HCFA)-1500 format), submitted by institutions for facility services (using the Uniform Bill (UB)-82 or UB-92 formats), and for pharmacy services (typically submitted electronically by the pharmacy at the time prescriptions were filled). No identifiable protected health information was extracted or accessed during the course of the study. Pursuant to the Health Insurance Portability and Accountability Act, the use of de-identified data does not require Institutional Review Board approval or waiver of authorization.

We defined SLE cases according to the following algorithm. For incident cases, the earliest date of service of a medical claim with a diagnosis of SLE from 01 January 2003 through 31 December 2008 was defined as the incidence date for subjects with incident SLE, provided the following inclusion criteria for SLE incidence were met: 1) the subject was at least 18 on the date of service; 2) the subject was continuously enrolled with medical and pharmacy benefits for 24 months prior to and 12 months following the date of service; 3) in the 12 months following the date of service the subject had either 3a) evidence of at least one inpatient claim with an SLE diagnosis (ICD-9-CM diagnosis code 710.0x) or 3b) two or more office or emergency room (ER) visits (combinations allowed) at least 30 days apart, but not more than 365 days apart with a diagnosis code for SLE (where the first visit occurred during the same calendar year as the date of service); 4) the subject had no claims with a diagnosis for SLE in the 24 months prior to the date of service; and 5) the subject had no claims for antimalarials, systemic corticosteroids, methotrexate, azathioprine, or mycophenolate mofetil for 12 months prior to the date of service. For prevalence of SLE, for each calendar year between 2003 and 2008, we used the same inclusion criteria as the incidence cohort, with the exception of requirements numbers 4) and 5) above and only required 12 months of continuous enrollment prior to the service date. A subject was required to have a date of service in each year he/she was counted as prevalent. Subject information in the database was reviewed back to the year 1999. Subjects with evidence of SLE from 2000 to 2002 (per the above criteria) who were a part of the study were excluded from the incidence cohort (but not from the prevalence cohort).

Subject characteristics

Age, gender and geographic location were captured from enrollment data based on the SLE claim associated with the earliest of all incidence and prevalence dates. The index date was defined as the incidence date for the incidence population. For the prevalence population, the index date for each person was chosen at random from among all the person’s prevalence dates between 2003 and 2008. Medications that could be used to treat SLE were identified from pharmacy claims (using National Drug Code (NDC) code bundles) and medical claims (using appropriate Healthcare Common Procedure Coding System (HCPCS) codes submitted by individual providers that use the HCFA-1500 or Centers for Medicare and Medicaid Services (CMS)-1500 formats or for facility services submitted by institutions that use the UB-82, UB-92, UB-04, or CMS-1450 formats) for the one-year period following the index date. Presence of specific physician specialties (rheumatologist, dermatologist, nephrologist, or neurologist) was captured from medical claims on the index date and during the year following the index date. Subjects identified as SLE incident/prevalent were also eligible for assignment to a lupus nephritis group or a lupus CNS group based on the presence of diagnosis or procedure codes for nephritis conditions (acute glomerulonephritis, nephrotic syndrome, chronic glomerulonephritis, nephritis unspecified, acute renal failure, chronic kidney disease, renal failure, kidney biopsy, hemodialysis or peritoneal dialysis, kidney transplant) or CNS conditions (seizure, psychosis, organic brain syndrome, cranial nerve disorder, demyelination syndromes, encephalitis, myelitis, encephalomyelitis, myelopathy, aseptic meningitis, cerebral vasculitis, mononeuritis multiplex, plexus disorders) at any time between 2003 and 2008. Subjects could be assigned to both a CNS lupus group and a lupus nephritis group if conditions for each manifestation were met.

Incidence and prevalence calculations

Annual crude incidence rates and prevalence were calculated for each calendar year 2003–2008 as the ratio of a numerator to a denominator, as described below.

For annual incidence rates, the year-specific numerator included subjects who were incident during that specific calendar year (per the criteria above). By definition, each subject was included in only one of the annual incidence numerators, as no more than one incidence date was assigned to each unique subject. The annual incidence denominators included all commercial health plan enrollees who 1) were 18 years of age or older as of 01 January of the specific year; 2) were continuously enrolled that year and two prior years (total of three years); and 3) did not have an incidence date or a prevalence date in any year prior to the specific year (subjects found incident or prevalent for SLE from 2000 to 2002 were also excluded).

An overall incidence rate for the period from 2003 to 2008 was also calculated whose numerator was defined as the number of unique subjects with an incidence date for SLE between 2003 and 2008. The 2003–2008 overall incidence denominator was defined as the total number of ‘at-risk’ person-years from 2003 to 2008. Each enrollee could contribute up to six ‘at-risk’ person-years, one for each calendar year between 2003 and 2008. To contribute a single ‘at-risk’ person-year to the denominator, an individual must have been continuously enrolled for the entire calendar year, continuously enrolled for a total of at least three years (including the calendar year required above), at least 18 years of age as of 01January of that calendar year, and not identified as SLE incident or SLE prevalent prior to the calendar year. The 2003–2008 overall incidence rate was calculated as the ratio of the 2003–2008 incidence numerator divided by the 2003–2008 incidence denominator, and reported per 100,000 person-years. The crude overall and year-specific incidence rates were age- and gender-adjusted to the 2000 US census population to allow comparison of rates in different populations.^23,24

For annual prevalence rate calculations, the year-specific numerator included all subjects who were identified as prevalent within that year. Each subject could be identified as prevalent for up to six years (each year from 2003 to 2008). Unlike with the incidence rates, prevalent subjects were not excluded from subsequent numerator calculations once identified as prevalent. The annual prevalence denominators were determined in the same way as the annual incidence denominators, except for requirement number 3 of the incident denominator (described above), which did not apply here.

Statistical and sensitivity analyses

Yearly trends in incidence rates were modeled in SAS 9.2 using a generalized linear model (GLM) with covariates, described below. The negative binomial distribution was used to model the incidence counts as a function of covariates, using a log link function and using as an offset the natural logarithm of the corresponding denominator. Equality of crude incidence rates across age groups, gender and region was each tested using likelihood ratio tests (LRT). Confidence intervals for the crude rates and the age- and gender-adjusted rates were obtained using the normal approximation to the binomial distribution. The uncertainty in each yearly age- and gender- adjusted rate was calculated using basic properties of variance, as each yearly age- and gender-adjusted rate was a simple linear combination of the crude age- and gender-adjusted rates for that same year. Sensitivity analyses for incidence and prevalence followed the calculations described above, but with the numerators limited to subgroups of subjects with specific clinical characteristics.

Results

Incidence

A total of 1557 subjects were identified for inclusion in the incidence cohort; of these, 15.99% were identified with lupus nephritis and 13.23% with CNS lupus (Table 1). For the 2003–2008 period, the estimate of the overall crude incidence rate of SLE was 6.63 cases per 100,000 person-years (Table 2). The overall crude incidence was higher among individuals who were 65 years of age or older, when compared to individuals 45–64 or 18–44 years of age (10.44 vs. 9.37 or 5.10 cases per 100,000 person-years; LRT Chi square = 14.71, degrees of freedom (df) = 2, p < 0.001). Also, overall crude SLE incidence was higher among females than males (11.10 vs. 1.89 cases per 100,000 person-years; LRT Chi square = 323.84, df = 1, p < 0.001). The (overall) age- and gender-adjusted SLE incidence rate from 2003 to 2008 was estimated at 7.22 cases per 100,000 person-years (95% confidence interval (CI) = 6.78–7.65) (Table 2). The overall age- and gender-adjusted incidence rates of lupus nephritis and CNS lupus over this period were 1.35 per 100,000 person-years and 1.03 per 100,000 person-years, respectively. The estimates of the year-specific age- and gender-adjusted incidence of SLE ranged from 6.64 to 7.82 cases per 100,000 individuals between 2003 and 2008 (Table 2).

Table 1

Characteristics of systemic lupus erythematosus (SLE) subjects (2003–2008)

	Incidence cohort subjects		Prevalence cohort subjects
	n	%	n	%
SLE (overall)^a	1557	100.00	15,396	100.00
Lupus nephritis	249	15.99	2508	16.29
CNS lupus	206	13.23	1994	12.95
Year
2003	216	13.87	3947	25.64
2004	252	16.19	4297	27.91
2005	255	16.38	4491	29.17
2006	262	16.83	5277	34.28
2007	278	17.85	5444	35.36
2008	294	18.88	6027	39.15
Age
18–44	634	40.72	6865	44.59
45–64	794	51.00	7584	49.26
65+	129	8.29	947	6.15
Gender
Female	1342	86.19	13,868	90.08
Male	215	13.81	1528	9.92
Region
Northeast	199	12.78	1461	9.49
Midwest	446	28.64	3893	25.29
South	664	42.65	7777	50.51
West	248	15.93	2265	14.71

Subjects could have more than one manifestation of SLE; CNS: central nervous system.

Table 2

Systemic lupus erythematosus (SLE) incidence (2003–2008)

	Incidence (per 100,000)^a
	Crude rate	Age- and gender-adjusted rate^d	95% CI (adjusted rate)
SLE (overall)^b	6.63	7.22	6.78–7.65
Lupus nephritis	1.06	1.35	1.14–1.55
CNS lupus	0.88	1.03	0.86–1.21
Year
2003	6.72	7.31	6.15–8.48
2004	7.12	7.82	6.66–8.97
2005	6.54	7.19	6.11–8.28
2006	6.36	6.76	5.76–7.75
2007	6.36	6.64	5.70–7.58
2008	6.79	7.72	6.64–8.80
Age (LRT Chi-square = 14.71, df = 2, p < 0.001)^c
18–44	5.10	4.96	4.57–5.35
45–64	9.37	9.34	8.69–9.99
65+	10.44	10.95	9.05–12.84
Gender (LRT Chi square = 323.84, df = 1, p < 0.001)^c
Female	11.10	11.89	11.11–12.66
Male	1.89	2.21	1.87–2.56

Rates are per 100,000 person-years for the period from 2003 to 2008, except year-specific rates, which are per 100,000 individuals; ^bSubjects could have more than one manifestation of SLE; groups were not mutually exclusive; ^cGeneralized linear model using negative binomial family and log link. Not adjusted for other covariates. Year-specific analysis on adjusted rates; demographic analyses on crude rates; ^dAge and gender adjustment was done for all variables, except age distribution was gender adjusted and gender distribution was age adjusted. Rates reflect 2000 US population. CNS: central nervous system; CI: confidence interval; LRT: likelihood ratio tests; df: degrees of freedom.

Sensitivity analyses were performed to investigate whether the SLE incidence estimates were robust evidence of clinical and treatment characteristics typical for SLE (Table 3). Individuals with SLE and a specialist visit on the index date had an overall age- and gender-adjusted SLE incidence rate of 2.66 cases per 100,000 person-years (95% CI = 2.41–2.91), and individuals with SLE and a specialist visit within a year following the index date had an overall adjusted SLE incidence of 5.21 cases per 100,000 person-years (95% CI = 4.85–5.57). The overall adjusted SLE incidence of subjects who had claims for mycophenolate mofetil, immunosuppressants, antimalarials and/or systemic corticosteroids within a year after the index date was 3.66 cases per 100,000 person-years (95% CI = 3.36–3.95). Finally, the overall age- and gender-adjusted SLE incidence rate among individuals who had at least one of the above characteristics was 5.80 cases per 100,000 person years (95% CI = 5.42–6.18).

Table 3

Sensitivity analysis of incidence rates (2003–2008)

	2003–2008 overall incidence (per 100,000 person-years)
	n	Crude rate	Age- and gender-adjusted rate	95% CI (adjusted rate)
Base case analysis	1557	6.63	7.22	6.78–7.65
Lupus nephritis	249	1.06	1.35	1.14–1.55
CNS lupus	206	0.88	1.03	0.86–1.21
Presence of specialist visit on index date^a
SLE	607	2.59	2.66	2.41–2.91
Lupus nephritis	63	0.27	0.33	0.23–0.43
CNS lupus	46	0.20	0.26	0.17–0.35
Presence of specialist visit in the year following index date^a
SLE	1157	4.93	5.21	4.85–5.57
Lupus nephritis	197	0.84	1.03	0.85–1.20
CNS lupus	161	0.69	0.78	0.64–0.93
Presence of antimalarials, immunosuppressants, mycophenolate mofetil, and/or systemic corticosteroids one year following index date^b
SLE	825	3.51	3.66	3.36–3.95
Lupus nephritis	141	0.60	0.70	0.56–0.84
CNS lupus	95	0.40	0.47	0.35–0.59
Presence of any of the above^c
SLE	1284	5.47	5.80	5.42–6.18
Lupus nephritis	216	0.92	1.13	0.95–1.32
CNS lupus	171	0.73	0.84	0.69–1.00

Specialist included rheumatologist, dermatologist, nephrologist, or neurologist; ^bAntimalarials included chloroquine HCL, chloroquine phosphate, hydroxychloroquine sulfate, or quinacrine HCL; immunosuppressants included methotrexate or azathioprine; systemic corticosteroids included betamethasone, corticotropin, dexamethasone, methylprednisolone, prednisolone, prednisone, or triamcinolone; ^cPresence of specialist visit on the index date; specialist visit within a year following the index date; and/or used antimalarials, immunosuppressants and/or systemic corticosteroids in the year following the index date; CI: confidence interval; CNS: central nervous system; SLE: systemic lupus erythematosus: LRT: likelihood ratio tests; df: degrees of freedom.

Prevalence

A total of 15,396 subjects were identified for inclusion in the SLE prevalence cohort (Table 1). Of these, 2508 were identified with lupus nephritis, and 1994 were identified with CNS lupus (Table 1). Annual SLE prevalence was calculated for each year from 2003 to 2008, and ranged between 81.07 cases per 100,000 individuals (95% CI = 78.54–83.60) in 2003 to 102.94 cases per 100,000 individuals (95% CI = 100.35–105.54) in 2008 (Table 4). Sensitivity analyses were performed to determine how annual SLE prevalence estimates varied when counting only SLE subjects who had a specialist visit on the prevalence date and/or a specialist visit within a year following the prevalence date. In general, annual SLE prevalence estimates from the sensitivity analysis were slightly lower than the base case prevalence estimates (Table 4).

Table 4

Yearly systemic lupus erythematosus (SLE) prevalence (2003–2008)

Year	Annual SLE prevalence (per 100,000 individuals)	95% CI
Base case analysis
2003	81.07	78.54–83.60
2004	81.08	78.66–83.50
2005	80.87	78.51–83.24
2006	89.48	87.07–91.90
2007	91.07	88.65–93.49
2008	102.94	100.35–105.54
Sensitivity analysis^a
2003	67.90	65.59–70.22
2004	68.19	65.97–70.42
2005	68.36	66.18–70.53
2006	77.41	75.16–79.65
2007	78.20	75.96–80.45
2008	89.35	86.93–91.77

SLE subjects with specialist visit on the prevalence date and/or specialist visit within a year following the prevalence date. Specialist included rheumatologist, dermatologist, neurologist, or nephrologist. Presence of specialist visit captured at every non-missing prevalence date between 2003 and 2008.

Discussion

In this study, we evaluated the incidence and prevalence of adult SLE in a large national managed-care organization in the US using historical data. Over the time period from 2003 to 2008, we estimated the overall age- and gender-adjusted SLE incidence rate to be 7.22 cases per 100,000 person-years. The overall adjusted SLE incidence rates of lupus nephritis and CNS lupus over the period from 2003 to 2008 were 1.35 and 1.03 per 100,000 person-years, respectively. Inpatient as well as outpatient claims with diagnosis codes for SLE were used to achieve more robust disease detection. As this approach could decrease specificity, an additional analysis that required specific clinical and treatment characteristics for case inclusion was performed. Because the result from this sensitivity analysis (5.80 per 100,000 person-years) was similar to the value calculated for overall SLE incidence (7.22 cases per 100,000 person-years), we conclude the analytic approach used to identify SLE subjects had good specificity. Yearly estimates of SLE prevalence in this study ranged from about 81 to 103 per 100,000 (sensitivity analysis range: 68–89 per 100,000).

A number of studies have evaluated the incidence and/or prevalence of SLE in the United States. Compared to previously published studies of SLE incidence in the US, the incidence rates determined in this study were slightly higher. Prior studies have reported the incidence of SLE in the US at fewer than six per 100,000,^13–16 whereas the year-specific incidence rates reported in the present study ranged from about six to eight per 100,000. Also, the SLE prevalence estimates generated in this study were within the range of previous reports of SLE prevalence in the US, but the estimates were much tighter than those published across the different studies (41.8 to 241).^15–22 Some estimates from previous studies are based on data that are not recent (e.g. from 10 years ago or more); moreover, some previous studies were limited to a specific geographic region. The present study used geographically diverse data from across the US and reported incidence and prevalence estimates from as recently as 2008. Differences in study design, methodology, population characteristics, study period and data sources could account for the variation in incidence and prevalence estimates among studies.

Limitations related to the use of administrative claims data should be considered when interpreting the results of this study. Importantly, the algorithm used to identify SLE patients in this study was not validated, and we could not ensure that a patient had SLE according to American College of Rheumatology (ACR) classification criteria.² Also, the heterogeneous clinical presentation of SLE may complicate diagnosis. To explore the robustness of SLE subject selection in this study, sensitivity analyses were performed based on the inclusion of additional clinical characteristics. Future studies in which subjects’ medical charts are reviewed could help confirm findings from this study. Our data did not allow for the breakdown of incidence and prevalence as a function of ethnic group(s), and future studies should investigate the effects of this variable on SLE epidemiology. Another limitation common to using administrative claims data is that we were not able to obtain entire medical histories for subjects in the study. Importantly, SLE subjects may have long periods of quiescent disease, with an average length of remission of 2.3 years (range of 1.0–5.7 years).²⁵ In the present study, we required that all subjects in the SLE incidence cohort have a two-year pre-index period with no evidence of SLE, and we required that incidence cohort subjects have no claims for SLE medications in the year prior to the index date. However, it is possible that some subjects included in the incidence cohort had evidence of SLE earlier than the plan data allowed us to determine, and this limitation may have caused the incidence rate values to be inflated. Due to this limitation, subjects identified as incident in this study may be better described as subjects with either newly diagnosed or newly active disease. Other study limitations may have affected prevalence estimates. When prevalence was calculated, subjects who were prevalent in one year were not necessarily counted as prevalent in subsequent years; subjects were required to meet study criteria for presence of SLE each year they were counted as prevalent. This requirement may have led to lower calculated values for prevalence rates in some years.

In conclusion, this study provided recent incidence and prevalence data of adult SLE in a managed-care plan from a large, geographical diverse population and adds to the SLE epidemiology literature in the US. The incidence observed in this plan was similar to, though slightly higher than, estimates from previous literature. Additionally, the prevalence observed in this plan was within the range of previous estimates.

Footnotes

Acknowledgement

The authors would like to thank Jesse Potash, PhD, at OptumInsight for assistance with preparation of this manuscript.

Funding

This study was supported by MedImmune LLC.

Conflict of interest statement

DEF has served as medical advisory board member to MedImmune LLC. WG and AWF are full-time employees of MedImmune LLC. AEC has served as a consultant to MedImmune LLC, Bristol Myers Squibb and GlaxoSmithKline/Human Genome Sciences and has received research funding from GlaxoSmithKline/Human Genome Sciences. TB and SRI have served as consultants to MedImmune LLC as employees of OptumInsight.

References

Manson

Isenberg

. The pathogenesis of systemic lupus erythematosus. Neth J Med 2003; 61: 343–346.

Tan

Cohen

Fries

. The 1982 revised criteria for the classification of systemic lupus erythematosus. Arthritis Rheum 1982; 25: 1271–1277.

The American College of Rheumatology. Nomenclature and case definitions for neuropsychiatric lupus syndromes. Arthritis Rheum 1999; 42: 599–608.

Manson

Rahman

. Systemic lupus erythematosus. Orphanet J Rare Dis 2006; 1: 6–6.

Moss

Ioannou

Sultan

Haq

Isenberg

. Outcome of a cohort of 300 patients with systemic lupus erythematosus attending a dedicated clinic for over two decades. Ann Rheum Dis 2002; 61: 409–413.

Cervera

Khamashta

Font

. Morbidity and mortality in systemic lupus erythematosus during a 10-year period: A comparison of early and late manifestations in a cohort of 1,000 patients. Medicine (Baltimore) 2003; 82: 299–308.

Urowitz

Gladman

Abu-Shakra

Farewell

. Mortality studies in systemic lupus erythematosus. Results from a single center. III. Improved survival over 24 years. J Rheumatol 1997; 24: 1061–1065.

Thumboo

Strand

. Health-related quality of life in patients with systemic lupus erythematosus: An update. Ann Acad Med Singapore 2007; 36: 115–122.

Pons-Estel

Alarcón

Scofield

Reinlib

Cooper

. Understanding the epidemiology and progression of systemic lupus erythematosus. Semin Arthritis Rheum 2010; 39: 257–268.

10.

McCarty

Manzi

Medsger

Jr Ramsey-Goldman

LaPorte

Kwoh

. Incidence of systemic lupus erythematosus. Race and gender differences. Arthritis Rheum 1995; 38: 1260–1270.

11.

Samanta

Feehally

Roy

Nichol

Sheldon

Walls

. High prevalence of systemic disease and mortality in Asian subjects with systemic lupus erythematosus. Ann Rheum Dis 1991; 50: 490–492.

12.

Danchenko

Satia

Anthony

. Epidemiology of systemic lupus erythematosus: A comparison of worldwide disease burden. Lupus 2006; 15: 308–318.

13.

Hochberg

. The incidence of systemic lupus erythematosus in Baltimore, Maryland, 1970–1977. Arthritis Rheum 1985; 28: 80–86.

14.

Naleway

Davis

Greenlee

Wilson

McCarty

. Epidemiology of systemic lupus erythematosus in rural Wisconsin. Lupus 2005; 14: 862–866.

15.

Uramoto

Michet

CJJr

Thumboo

Sunku

O'Fallon

Gabriel

. Trends in the incidence and mortality of systemic lupus erythematosus, 1950–1992. Arthritis Rheum 1999; 42: 46–50.

16.

Lim SS, Bayakly R, Gordon C, et al. The Georgia Lupus Registry: The incidence and prevalence of systemic lupus erythematosus. In: Abstracts of the American College of Rheumatology/Association of Rheumatology health professionals annual scientific meeting, Chicago, IL, USA, 4–9 Nov 2011. Arthritis Rheum 2011; 63 (Suppl 10): 2444.

17.

Chakravarty

Bush

Manzi

Clarke

Ward

. Prevalence of adult systemic lupus erythematosus in California and Pennsylvania in 2000: Estimates obtained using hospitalization data. Arthritis Rheum 2007; 56: 2092–2094.

18.

Maskarinec

Katz

. Prevalence of systemic lupus erythematosus in Hawaii: Is there a difference between ethnic groups? Hawaii Med J 1995; 54: 406–409.

19.

Balluz

Philen

Ortega

. Investigation of systemic lupus erythematosus in Nogales, Arizona. Am J Epidemiol 2001; 154: 1029–1036.

20.

Massachusetts Department of Public Health, Center for Environmental Health, Environmental Epidemiology Program. Final report on the incidence and prevalence of systemic lupus erythematosus (SLE) in Boston and environmental factors, http://www.mass.gov/eohhs/docs/dph/environmental/tracking/lupus-report-web.pdf (2007, accessed 12 April 2011).

21.

Ward

. Prevalence of physician-diagnosed systemic lupus erythematosus in the United States: Results from the third national health and nutrition examination survey. J Womens Health (Larchmt) 2004; 13: 713–718.

22.

Feldman CH, Hiraki LT, Solomon DH, et al. Prevalence and demographics of systemic lupus erythematosus among U.S. adults with Medicaid coverage, 2002–2004. In: Abstracts of the American College of Rheumatology/Association of Rheumatology health professionals annual scientific meeting, Chicago, IL, USA, 4–9 Nov 2011. Arthritis Rheum 2011; 63 (Suppl 10): 2441.

23.

United States Census Bureau. United States Census 2000, http://www.census.gov/main/www/cen2000.html (2000, accessed 20 April 2012).

24.

Insinga

Itzler

Pellissier

Saddier

Nikas

. The incidence of herpes zoster in a United States administrative database. J Gen Intern Med 2005; 20: 748–753.

25.

Barr

Zonana-Nacach

Magder

Petri

. Patterns of disease activity in systemic lupus erythematosus. Arthritis Rheum 1999; 42: 2682–2688.