ETHNICITY AND RHEUMATOID ARTHRITIS: A SYSTEMATIC REVIEW

 

Hanan S Abozaid1, 2, Nihal A. Fathi2, David L Scott1, Sophia Steer1

 

  1. Academic Rheumatology, King’s college London, London SE5 9RJ, UK
  2. Rheumatology department, Sohag university, Egypt

 

Background

Several patient-related factors impact on the clinical phenotype and outcome of rheumatoid arthritis (RA). The effects of gender and deprivation are well known. The effects of ethnicity have received less attention. Differences in the clinical expression of disease between different ethnic populations may reflect different environmental, social or genetic factors or all. Understanding these factors will be useful in elucidating basic disease mechanisms and may change management. We systematically reviewed studies examining the effects of ethnicity in RA to define the impact of this key factor.

Methods:

We systematically searched electronic databases and published bibliographies (up to March 2009) to identify cohort, case control and cross sectional studies which compared two or more ethnic populations of adults with RA and evaluated one or more standard outcomes. These spanned clinical (joint counts, extra articular features); laboratory (ESR/CRP and rheumatoid factor), functional (HAQ) and radiological (erosions) assessments. Two authors independently selected studies; extracted data, and assessed study quality, results were presented separately by each outcome using the Cochrane software RevMan 5.

Results

From 2246 citations we reviewed 30 potentially relevant studies; 17 met our entry criteria and were included; they enrolled 10,002 patients. 11 studies compared African American/African Caribbean (AA/AC) patients with White/Europian(WE) populations, including 6 comparisons within single centres. 3 studies compared Asian with WE populations and the remaining studies compared diverse populations, including different areas of Europe, different Jewish populations and Arab WE populations.

 

The 6 studies comparing AA/AC patients with WE populations within single centres showed significant differences between groups. HAQ scores were lower in WE (standardised mean difference 0.30 (95%CI 0.29, 0.31), though there was significant heterogeneity between individual studies (p=0.01). Other clinical outcomes, including swollen and tender joint counts and ESR showed some differences but the findings were inconsistent across studies. There were no consistent differences in rheumatoid factor positivity, erosive disease and extra-articular features such as nodules. We identified several important confounding factors including differences in socioeconomic status and access to treatments.

 

Conclusions

There is evidence that the impact of RA is worse in patients from some ethnic backgrounds, particularly AA/AC patients, with higher HAQ scores in these patients. We consider ethnicity is a potentially important factor in the outcome of RA. However, it is uncertain whether this reflects differences in the clinical phenotype, which might result from genetic differences, or is a consequence of socioeconomic factors and relative deprivation.