OBJECTIVES: To determine the geographical distribution of cases of systemic lupus erythematosus (SLE) in a defined geographical area in the East Midlands, UK, and, in particular, to search for spatial variation in cases that may implicate the role of environmental factors in SLE aetiology. METHODS: Six methods of case ascertainment were used. The postcode of the patient's domicile at time of first definite symptom of SLE was used for analysis which included case mapping, probability mapping by electoral ward, and variogram analysis. RESULTS: The study area population of 613,700 contained 200 SLE patients, 188 of whom experienced their first symptom whilst residing in the area. Case mapping revealed 12 SLE patients residing within an area of one square mile, including four men and six patients with RNP antibodies. The use of probability mapping showed five wards in close proximity to each other to have a greater number of SLE cases than would be expected by chance (p < 0.1). The 'cluster' of patients seen on the case map fell into two wards which showed a significant excess of cases only when combined (p = 0.006). The variogram of the incidence rates for each ward did not confirm any structure or pattern to the distribution of cases for the whole area. CONCLUSIONS: Some areas have a greater than expected prevalence of SLE. The normal result from variogram analysis suggests that the cause(s) for these excess number of cases does not have an effect across the whole study area.
OBJECTIVES: To determine the geographical distribution of cases of systemic lupus erythematosus (SLE) in a defined geographical area in the East Midlands, UK, and, in particular, to search for spatial variation in cases that may implicate the role of environmental factors in SLE aetiology. METHODS: Six methods of case ascertainment were used. The postcode of the patient's domicile at time of first definite symptom of SLE was used for analysis which included case mapping, probability mapping by electoral ward, and variogram analysis. RESULTS: The study area population of 613,700 contained 200 SLEpatients, 188 of whom experienced their first symptom whilst residing in the area. Case mapping revealed 12 SLEpatients residing within an area of one square mile, including four men and six patients with RNP antibodies. The use of probability mapping showed five wards in close proximity to each other to have a greater number of SLE cases than would be expected by chance (p < 0.1). The 'cluster' of patients seen on the case map fell into two wards which showed a significant excess of cases only when combined (p = 0.006). The variogram of the incidence rates for each ward did not confirm any structure or pattern to the distribution of cases for the whole area. CONCLUSIONS: Some areas have a greater than expected prevalence of SLE. The normal result from variogram analysis suggests that the cause(s) for these excess number of cases does not have an effect across the whole study area.
Authors: E M Tan; A S Cohen; J F Fries; A T Masi; D J McShane; N F Rothfield; J G Schaller; N Talal; R J Winchester Journal: Arthritis Rheum Date: 1982-11