Prevalence estimates for rheumatoid arthritis (RA) vary across regions and populations, likely reflecting differences in population prevalence of RA susceptibility genes combined with variable exposure to RA-associated environmental risk factors. Globally, some indigenous people have among the highest reported prevalence of RA, with estimates ranging from 2 to 6 times greater than those reported in nonindigenous populations; other indigenous populations do not have higher rates. Central Canadian population-based estimates including those from the province of Manitoba found indigenous people were affected by RA 2 to 3 times more than the general population. However, these studies used variable methods and data sources to identify indigenous individuals with RA.1–5
The indigenous population residing in the Canadian province of Manitoba is comprised mainly of Cree, Ojibway, and Oji-Cree persons, hereafter referred to as First Nations (FN).6 7 8 9 4,5
METHODS
Data Sources
Manitoba is a central Canadian province with a population of approximately 1.2 million. Census data indicate 18% of the Manitoba population self-reports as FN. This study linked 3 data sources including health claims data, the Federal Indian Registry file, and clinical data from the UMRC. Universal health care coverage is provided to 98% of the Manitoba population. Records for nearly all health services accessed, including service date, medical provider accessed, and diagnosis, are recorded in the Manitoba Population Research Data Repository (MPRDR) housed at the Manitoba Centre for Health Policy (MCHP).10 International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM ) or International Classification of Diseases, Ninth Revision, Canada (ICD-10-CA ) codes. The ICD-9-CM codes are used for ambulatory care visits. Health service data are available from 1984 onward. Data on prescription drug use regardless of payer are available from 1995, including date dispensed and Drug Identification Number; the Drug Identification Numbers are linked to the World Health Organization's Anatomic Therapeutic Chemical Classification System. Repository data are deidentified. Each resident has a unique scrambled identifier linked to all their service claims. For this study, we accessed health records from April 1, 1995, to March 31, 2010. In Manitoba, health privacy legislation requires that categories that contain fewer than 6 individuals are not reported, and these data are “censored.”
First Nations individuals were identified in the administrative data using the Federal Indian Registry File, which contains records for all registered FN for the purposes of legal entitlement. Permission to use the file was obtained from Indian and Northern Affairs Canada, the University of Manitoba Ethics Review Board, Manitoba Heath Information Research Governance Committee, and the Assembly of Manitoba Chiefs Secretariat Inc. (now the First Nations Health and Social Secretariat of Manitoba), under the auspices of the Health Disparity Research Program (PI2006-00209, principal investigator B.E.). The file was linked to the MPRDR. Linkage was expanded based on familial kinship identifiers in the MPRDR to include individuals eligible but not yet registered as FN and those of FN heritage and likely eligible to be registered as FN.11 12,13
Since 1990, the UMRC has maintained a clinical database of all patients seen in the clinic, which includes rheumatologist-recorded diagnoses and visit dates. Records of patients seen between 2000 and 2010 (RA = 2281, non-RA n = 7044; 73% female, self-reported FN 13%) were linked to the MPRDR using scrambled identifiers. The linked data set was used to validate administrative definitions for RA. Self-reported ethnicity was not used in the validation studies or for population-based estimates of prevalence and incidence.
Validation of the Administrative Definition for RA
Administrative definitions for RA were validated in the MPRDR using the UMRC diagnosis as the criterion standard. We adapted validated administrative definitions currently used to identify other chronic immune-mediated diseases in the regional administrative data.14,15 ICD-9-CM /ICD-10-CA codes 714/M05, M06. For those who were residents for less than 2 years, 3 or more claims sufficed (definition 1). As a comparison, we applied other administrative definitions used in North American data sets during the study period16–18 https://links.lww.com/RHU/A143 ICD codes and definitions). Validity of the administrative definitions for RA was determined for use in the entire Manitoba population and then estimated separately for FN and non-FN populations. The κ, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) with 95% confidence intervals (95% CIs) and Youden J statistic (sensitivity + specificity − 1) * 10019
Prevalence and Incidence
Definition 1 was applied to identify RA cases. Using end-year population figures as the denominators, crude prevalence rates for RA per 100 population (i.e., %) for each year from 2000 to 2010 were determined for all Manitobans. Rates were then stratified by FN status and by age group. Age groups were 0 to 18, 19 to 28, 29 to 38, 39 to 48, 49 to 58, 59 to 68, and older than 68 years. We included the 0- to 18-year age group as ambulatory visits in the MPRDR are coded with ICD-9-CM codes that do not reflect updated classification of juvenile chronic arthritis. This whole population approach was more inclusive and respectful of the younger provincial FN demographic. Using a 5-year run-in period to eliminate prevalent cases, incident RA cases were identified and then stratified by FN status, age group, and sex. Prevalence and incidence rates were modeled over time using Poisson regression models adjusted for age group, sex, and regional health authority (based on postal code of residence). All models fit well with the ratio of the model deviance to degrees of freedom close to 1 (range, 0.90–1.35). Comparisons between FN and non-FN populations are reported using relative rates (RRs) with 95% CIs.
Age at diagnosis, defined as the age at first RA code in the incident RA cohorts, was reported as mean years (SD) and duration followed (for incident RA) as months (SD). Health care utilization (physician visits) was identified by provider (any provider, nonrheumatology specialist provider, rheumatology specialist provider) and reported as mean number of visits/RA individual during the 10-year period. Age, duration followed, and physician visits were compared between FN and non-FN populations using Student t tests and reported with 95% CIs. Statistical significance for all comparisons was assessed using the criterion of α = 0.05. All data management, programming, and analyses were performed using SAS (SAS Institute Inc, Cary, NC).
The study was approved by the Ethics Review Board (H2010:273), the MCHP (2012-026), and the Heath Information Research Governance Committee (2010/2011-29) and was conducted with the support of the Assembly of Manitoba Chiefs Secretariat Inc. (now the First Nations Health and Social Secretariat of Manitoba).
RESULTS
Validation of RA Definition
The validation cohort included 2281 RA patients (432 FN) among 9325 total patients (1210 FN). Seventy-three percent were female. Definition 1 had the best specificity when applied to the entire Manitoba population (κ = 65.9 [95% CI, 64.1–67.6]; sensitivity, 77.1 [95% CI, 75.4–78.8]; specificity, 90.3 [95% CI, 89.6–91.0]; PPV, 72.0 [95% CI, 70.3–73.8]; NPV, 92.4 [95% CI, 91.8–93.0]; and Youden J of 67.4). This definition also had the best specificity of those tested when applied to the FN population, although it was less accurate than for the non-FN population (κ = 61.4 [95% CI, 56.8–66.0]; sensitivity, 74.8 [95% CI, 70.8–78.9]; specificity, 86.4 [95% CI, 84.0–88.8]; PPV, 75.7 [95% CI, 71.7–79.7]; NPV, 85.9 [95% CI, 83.4–88.3]; and Youden J of 61.3). The validity of the alternative definitions to identify RA in the total and FN populations in Manitoba is reported in Supplementary Table 2, https://links.lww.com/RHU/A143
We examined those cases identified by definition 1 as RA, who had a non-RA criterion-standard diagnosis in the clinical database, that is, FPs. Of the 681 FPs, 250 had other inflammatory arthropathy diagnoses, 91 had noninflammatory diagnoses, and 93 had connective tissue disease. A total of 247 non-RA cases did not have a known rheumatologic diagnosis or could not be classified as the number of individuals with the diagnosis was censored because of health privacy legislation in Manitoba. For more than 400 of the FP claims, a diagnosis of RA was assigned by a rheumatologist at a subsequent visit (after the date of initial UMRC diagnosis). This later diagnosis suggests that some of these FPs may represent an evolving arthropathy. A large number of these patients were prescribed disease-modifying antirheumatic drugs, corticosteroids, and even biologics, which suggests that many evolved into RA or had another true inflammatory disease. For most of the FPs, RA claims continued to persist in the administrative data set, even if another diagnosis was given by a rheumatologist (Supplementary Table 3, https://links.lww.com/RHU/A143
Prevalence
Between 2000 and 2010, we identified 8095 prevalent RA cases (1095 FN, 7000 non-FN) in Manitoba. During this period, the total Manitoba population increased from 1,151,811 in 2000 to 1,223,221 in 2010, and the total FN population increased from 105,549 to 125,289. The crude prevalence of RA in Manitoba for 2009/2010 was 0.65%, (FN 0.85%, non-FN 0.63%). The adjusted RA prevalence in 2009/2010 for FN females was 1.64% (95% CI, 1.45–1.86); FN males, 0.61% (95% CI, 0.52–0.72); non-FN females, 0.59% (95% CI, 0.53–0.65); and non-FN males, 0.23% (95% CI, 0.20–0.25). Over the study period 2000–2010, the adjusted prevalence of RA increased for both FN and non-FN (Fig. 1 A), as did the crude prevalence rate of RA (Supplementary Table 4, https://links.lww.com/RHU/A143
FIGURE 1: Age-adjusted RA prevalence. A, Annual age-adjusted prevalence (%) of RA in FN and non-FN populations. B, Annual age-specific risk RRs of prevalence for FN versus non-FN populations. All FN versus non-FN comparisons, p < 0.0001.
In 2009/2010, the adjusted prevalence of RA was 2.5-fold higher for FN than non-FN population (Fig. 1 B, Supplementary Table 5, https://links.lww.com/RHU/A143
The overall percentage of prevalent RA patients who were female in Manitoba was 73.76% (in 2009/2010: FN RA 77.34% female, non-FN RA 73.41% female; RR, 1.24 (95% CI, 1.06–1.44; p = 0.007). The crude and adjusted prevalence of RA increased for both sexes. The adjusted prevalence of RA was higher in females than males in all years.
Incidence
Between 2000 and 2010, we identified 4159 incident RA cases (FN 566, non-FN 3593). In 2009/2010, the crude RA incidence was 13.4/100,000 population (FN 21.6/100,000; non-FN 12.5/100,000). Crude RA incidence declined over the study period but remained higher for FN than non-FN populations in each year (Supplementary Table 5, https://links.lww.com/RHU/A143 Table ). The adjusted RA incidence in 2009/2010 was 2.4-fold higher for females than males (RR all males vs. all females, 0.41; 95% CI, 0.25–0.67; p < 0.0001) and 2.2-fold higher for FN compared with non-FN (RR 2.2; 95% CI, 1.26–4.99). Adjusted RA incidence declined for most years between 2000 and 2010. No RA cases were diagnosed in FN older than 68 years in 2004 or 2008. Adjusted incidence rates remained higher for FN than non-FN populations especially for ages 19 to 48 years, which had RRs of RA diagnosis 2.5- to 4.6-fold higher in FN than non-FN populations (Fig. 2 , Supplementary Table 4, https://links.lww.com/RHU/A143
TABLE: Relative Rates (RRs) and 95% Confidence Intervals (95% CI) of Rheumatoid Arthritis (RA) Incidence for FN Versus Non-FN Populations 2000–2010
FIGURE 2: Age-adjusted RA incidence. A, Annual age-adjusted incidence/100,000 population. B, Age-specific RRs of incident RA for FN versus non-FN populations in 2 time periods. All FN versus non-FN comparisons, p < 0.0001.
Age at Diagnosis
The largest percentage of FN were diagnosed between ages 29 and 48 years, whereas the largest proportion of non-FN were diagnosed between ages 49 and 68 years (Fig. 3 ). The First Nations population in the incident RA cohort was on average 13 years younger at RA diagnosis than the non-FN population (FN 39.6 [95% CI, 38.4–40.8 years] vs. non-FN 53.3 [95% CI, 52.7–53.9 years]; P < 0.0001). This was true for females (FN 39.3 [95% CI, 37.9–40.6 years] vs. non-FN 52.4 [95% CI, 51.7–53.2 years]) and males (FN 40.7 [95% CI, 37.9–43.3 years] vs. non-FN 55.3 [95% CI, 54.3–56.4 years]; both p < 0.0001). Age at diagnosis did not change over the study period.
FIGURE 3: Age at RA diagnosis. A, Annual mean age at diagnosis. B, Percentage of RA patients in each age group based on age at diagnosis for 2 time periods. Data censored for FN and non-FN populations at ages 59 to 68 years and older than 68 years (2000–2004).
Physician Visits
Using the prevalent RA cohort, we compared total physician visits, nonrheumatology specialist visits, and rheumatology specialist visits between FN and non-FN populations over the study period. First Nations with RA had more total physician visits compared with non-FN with RA (mean visits/person 109.5 [95% CI, 104.9–114.1] vs. 98.6 [95% CI, 97.21–100.1]; p < 0.0001) and more nonrheumatology specialist visits (86.7 [95% CI, 82.5–91.0] vs. 67.2 [95% CI, 66.1–68.3]; p < 0.0001). However, FN with RA had fewer rheumatology specialist visits (mean visits/person 6.9 [95% CI, 6.4–7.4] vs. 8.2 [95% CI, 8.0–8.4]; p < 0.0001). While we did not have person-years of follow-up, FN and non-FN populations in the incident cohort (2000–2010) had a similar duration of follow-up (FN 66.2 months [95% CI, 63.5–68.8], non-FN 64.6 months [95% CI, 63.5–65.7]; p = 0.29).
DISCUSSION
The prevalence of RA in Manitoba increased steadily in the period 2000 to 2010, with 0.65% of the Manitoba population living with RA including 164 Manitobans diagnosed in that year. In 2009/2010, the adjusted prevalence of RA was 2.6 times higher in FN than in non-FN, whereas the adjusted incidence of RA was 2.2 times higher in FN than in non-FN. This increased RR of prevalent and incident RA for FN compared with non-FN was particularly striking for young females of childbearing age (ages 19–48 years). Over the study period, the adjusted RA prevalence increased and the adjusted RA incidence decreased, but both remained higher for FN than non-FN. Compared with the non-FN population, the FN population was younger at diagnosis and had fewer rheumatology visits.
The overall 2009/2010 prevalence of RA in Manitoba was comparable to reports from other Canadian provinces.4,17,20 20 21
Whereas some studies have also reported declining RA incidence,22,23 21 24,25 26,27
Consistent with a similar Alberta (Canada) population-based study using administrative health data,4
A particularly important finding is that the increased RRs of prevalent and incident RA for FN versus non-FN were highest for females in the childbearing years. The postpartum and perimenopause life stages are known to be high-risk periods for the development of RA. We have previously demonstrated early age at first pregnancy, and that the postpartum period increased the risk of RA onset, whereas greater parity reduced RA risk in FN populations.28 29 30
The reasons for the increased prevalence, incidence, and disease severity of RA in FN people are likely multifactorial. The high frequency of human leukocyte antigen shared epitope alleles in the general FN population, combined with high rates of smoking, parity, and other environmental exposures reported to be associated with RA, may contribute to the increased risk of developing disease.8
The notable exception to the increased prevalence of RA in the FN population was seen in the younger age group. While this analysis was undertaken for a more inclusive understanding of RA demographically, this finding must be interpreted with caution. The administrative definition of RA was validated only with adult patients. The ICD-9-CM codes do not reflect updated classification of juvenile chronic arthritis. The MPRDR still uses ICD-9-CM codes for ambulatory care visits. Therefore, while it is assumed that prevalence and incidence in individuals younger than 18 years would be slightly underestimated using our definition, it is not known whether this would differentially affect FN and non-FN population estimates of pediatric onset arthritis. Administrative case definitions incorporating updates to the classification criteria for childhood arthritis31 32 33 34–36
In terms of health care access, Manitoba FN people overall have more primary care visits and hospitalizations but have fewer specialist visits compared with the non-FN population.13 9 4,37 38 39 40 41,42
Overall, our study has several strengths including its large size, population-based design, and inclusive identification of FN. Like other studies, it is subject to the limitations of administrative data including imprecise diagnosis coding. Our case definition was validated for this data set, performed well compared with other published definitions available at the time of the analysis (and published recently43 17,20
Our administrative data do not contain patient-reported measures contributing to the overall burden of disease. However, we have previously reported severe disease in the FN RA population,9 44
CONCLUSIONS
Compared with the general population, FN people in the province of Manitoba have a disproportionate burden of RA, with higher prevalence and incidence rates particularly for young females, a demographic group that is increasing in size both regionally and nationally. Despite the overall decline of RA incidence, the burden of RA will likely increase significantly for FN Manitobans, along with an increased need for rheumatology services to address disparities in accessing rheumatology care. Strategies to identify and address modifiable factors contributing to this excess RA burden are urgently needed. Moreover, rheumatology resources must be optimized and novel methods of culturally appropriate care delivery explored to improve outcomes for this population group.
KEYPOINTS
The prevalence of RA is increasing and the incidence of RA is decreasing in the Canadian province of Manitoba.
The FN population has a higher prevalence and incidence of RA and an earlier age at RA onset compared with the non-FN population.
The FN population has fewer rheumatology visits demonstrating a gap in rheumatology care delivery, which needs to be addressed.
ACKNOWLEDGMENTS
Drs. El-Gabalawy, Canvin, Chalmers, Barager, Robinson, and Dhindsa contributed to the clinical data set used to validate the administrative definition. The authors thank Dr. Ruth Ann Marrie for helpful critical review of the manuscript. The authors acknowledge the MCHP for use of data contained in the MPRDR under project 2010/2011-29. The results and conclusions are those of the authors, and no official endorsement by the MCHP, Manitoba Health, Seniors and Active Living, or other data providers is intended or should be inferred.
REFERENCES
1. Ferucci ED, Schumacher MC, Lanier AP, et al. Arthritis prevalence and associations in American Indian and Alaska Native people.
Arthritis Rheum . 2008;59:1128–1136.
2. Peschken CA, Esdaile JM. Rheumatic diseases in North America's indigenous peoples.
Semin Arthritis Rheum . 1999;28:368–391.
3. McDougall C, Hurd K, Barnabe C. Systematic review of rheumatic disease epidemiology in the indigenous populations of Canada, the United States, Australia, and New Zealand.
Semin Arthritis Rheum . 2017;46:675–686.
4. Barnabe C, Jones CA, Bernatsky S, et al. Inflammatory arthritis prevalence and health services use in the First Nations and non-First Nations populations of Alberta, Canada.
Arthritis Care Res (Hoboken) . 2017;69:467–474.
5. Barnabe C, Elias B, Bartlett J, et al. Arthritis in Aboriginal Manitobans: evidence for a high burden of disease.
J Rheumatol . 2008;35:1145–1150.
6. Branch C.
Words First: An Evolving Terminology Relating to Aboriginal Peoples in Canada . Ottawa, Ontario, Canada: Indian and Northern Affairs, Government of Canada; 2002.
7. Statistics Canada. Aboriginal Peoples Highlight Tables, 2016 Census. 2017. Updated October 11, 2017; cited October 28, 2017. Available at:
http://www12.statcan.gc.ca/census-recensement/2016/dp-pd/hlt-fst/abo-aut/Table . Accessed October 28, 2017.
8. El-Gabalawy HS, Robinson DB, Hart D, et al. Immunogenetic risks of anti-cyclical citrullinated peptide antibodies in a North American Native population with rheumatoid arthritis and their first-degree relatives.
J Rheumatol . 2009;36:1130–1135.
9. Peschken CA, Hitchon CA, Robinson DB, et al. Rheumatoid arthritis in a North American Native population: longitudinal followup and comparison with a white population.
J Rheumatol . 2010;37:1589–1595.
10. Manitoba Centre for Health Policy. Manitoba Population Research Data Repository—Overview. 2018. Updated 2018; cited 2018 March 28, 2018. Available at:
http://umanitoba.ca/faculties/health_sciences/medicine/units/chs/departmental_units/mchp/resources/repository/index.html . Accessed March 28, 2018.
11. Elias B, Busby K, Martens P. One little, too little: counting Canada's indigenous people for improved health reporting.
Soc Sci Med . 2015;138:179–186.
12. Jebamani LS, Burchill CA, Martens PJ. Using data linkage to identify First Nations Manitobans: technical, ethical, and political issues.
Can J Public Health . 2005;96(suppl 1):S28–S32.
13. Martens PJ, Sanderson D, Jebamani L. Health services use of Manitoba First Nations people: is it related to underlying need?
Can J Public Health . 2005;96(suppl 1):S39–S44.
14. Marrie RA, Yu N, Blanchard J, et al. The rising prevalence and changing age distribution of multiple sclerosis in Manitoba.
Neurology . 2010;74:465–471.
15. Bernstein CN, Blanchard JF, Rawsthorne P, et al. Epidemiology of Crohn's disease and ulcerative colitis in a central Canadian province: a population-based study.
Am J Epidemiol . 1999;149:916–924.
16. MacLean CH, Louie R, Leake B, et al. Quality of care for patients with rheumatoid arthritis.
JAMA . 2000;284:984–992.
17. Lacaille D, Anis AH, Guh DP, et al. Gaps in care for rheumatoid arthritis: a population study.
Arthritis Rheum . 2005;53:241–248.
18. Lix L, Yogendran M, Burchill C, Metge C, McKeen N, Moore D, et al. Defining and validating chronic diseases: an administrative data approach. Winnipeg, Manitoba, Canada: Manitoba Centre for Health Policy; 2006.
19. Youden WJ. Index for rating diagnostic tests.
Cancer . 1950;3:32–35.
20. Hanly JG, Thompson K, Skedgel C. The use of administrative health care databases to identify patients with rheumatoid arthritis.
Open Access Rheumatol Res Rev . 2015;7:69–75.
21. Jean S, Hudson M, Gamache P, et al. Temporal trends in prevalence, incidence, and mortality for rheumatoid arthritis in Quebec, Canada: a population-based study.
Clin Rheumatol . 2017;36:2667–2671.
22. Doran MF, Pond GR, Crowson CS, et al. Trends in incidence and mortality in rheumatoid arthritis in Rochester, Minnesota, over a forty-year period.
Arthritis Rheum . 2002;46:625–631.
23. Symmons DP, Barrett EM, Bankhead CR, et al. The incidence of rheumatoid arthritis in the United Kingdom: results from the Norfolk Arthritis Register.
Br J Rheumatol . 1994;33:735–739.
24. Myasoedova E, Crowson CS, Kremers HM, et al. Is the incidence of rheumatoid arthritis rising?: Results from Olmsted County, Minnesota, 1955–2007.
Arthritis Rheum . 2010;62:1576–1582.
25. Pedersen JK, Svendsen AJ, Horslev-Petersen K. Incidence of rheumatoid arthritis in the southern part of Denmark from 1995 to 2001.
Open Rheumatol J . 2007;1:18–23.
26. Tables for population growth: Canada, provinces and territories, 2010. Government of Canada; 2018 [updated 2018; cited March 23, 2018]. Available at:
https://www.statcan.gc.ca/pub/91-209-x/2011001/article/11508/tbl/pop-tbl-eng.htm#a1 . Accessed March 23, 2018.
27. Chagnon J.
Migration: International, 2010 and 2011. Statistics Canada . Ottawa, Canada: Minister of Industry; 2013.
28. Peschken CA, Robinson DB, Hitchon CA, et al. Pregnancy and the risk of rheumatoid arthritis in a highly predisposed North American Native population.
J Rheumatol . 2012;39:2253–2260.
29. O-Donnell V, Wallace S. Women in Canada. A gender-based statistical report. First Nations Metis and Inuit women.
Stat Canada . 2011.
30. Krause ML, Makol A. Management of rheumatoid arthritis during pregnancy: challenges and solutions.
Open Access Rheumatol . 2016;8:23–36.
31. Petty RE, Southwood TR, Manners P, et al. International League of Associations for Rheumatology classification of juvenile idiopathic arthritis: second revision, Edmonton, 2001.
J Rheumatol . 2004;31:390–392.
32. Shiff NJ, Oen K, Rabbani R, et al. Validation of administrative case ascertainment algorithms for chronic childhood arthritis in Manitoba, Canada.
Rheumatol Int . 2017;37:1575–1584.
33. Shiff NJ, Lix LM, Oen K, et al. Chronic inflammatory arthritis prevalence estimates for children and adolescents in three Canadian provinces.
Rheumatol Int . 2015;35:345–350.
34. Jarvis JN, Cleland SY. Rheumatic disease in Native American children: opportunities and challenge.
Curr Rheumatol Rep . 2003;5:471–476.
35. Rosenberg AM, Petty RE, Oen KG, et al. Rheumatic diseases in Western Canadian Indian children.
J Rheumatol . 1982;9:589–592.
36. Khodra B, Stevens A, Ferucci E. Prevalence of juvenile idiopathic arthritis in the Alaska Native population [abstract].
Arthritis Rheum . 2017;69.
37. Marrone S. Understanding barriers to health care: a review of disparities in health care services among indigenous populations.
Int J Circumpolar Health . 2007;66:188–198.
38. Glazier RH, Dalby DM, Badley EM, et al. Determinants of physician confidence in the primary care management of musculoskeletal disorders.
J Rheumatol . 1996;23:351–356.
39. Hindhede AL, Bonde A, Schipperijn J, et al. How do socio-economic factors and distance predict access to prevention and rehabilitation services in a Danish municipality?
Prim Health Care Res Dev . 2016;17:578–585.
40. Thurston WE, Coupal S, Jones CA, et al. Discordant indigenous and provider frames explain challenges in improving access to arthritis care: a qualitative study using constructivist grounded theory.
Int J Equity Health . 2014;13:46.
41. Hanly JG. Manpower in Canadian academic rheumatology units: current status and future trends. Canadian Council of Academic Rheumatologists.
J Rheumatol . 2001;28:1944–1951.
42. Deal CL, Hooker R, Harrington T, et al. The United States rheumatology workforce: supply and demand, 2005–2025.
Arthritis Rheum . 2007;56:722–729.
43. Hanly JG, Thompson K, Skedgel C. A longitudinal study of ambulatory physician encounters, emergency room visits, and hospitalizations by patients with rheumatoid arthritis: a 13-year population health study.
J Rheumatol . 2017;44:1421–1428.
44. Pelaez-Ballestas I, Granados Y, Quintana R, et al. Epidemiology and socioeconomic impact of the rheumatic diseases on indigenous people: an invisible syndemic public health problem.
Ann Rheum Dis . 2018;77:1397–1404.
