Clustering of systemic lupus erythematosus mortality in southwestern Spain

Abstract

Objective To analyse time trends in systemic lupus erythematosus (SLE) mortality and explore possible provincial clustering of SLE mortality in Spain (2001–2020). Methods We conducted an ecological study using deaths registered in SLE at the Spanish National Institute of Statistics between 2001 and 2020. Jointpoint regression models have been used to evaluate temporal trends. To analyse the spatial pattern of SLE mortality in men and women in Spain, crude rates, age-standardised mortality rates (ASMRs), smooth relative risk (RR) and posterior probabilities (PP) for RR greater than one for the period 2001–2020 were calculated. The Global Moran I index was used to assess the existence of global spatial autocorrelation. Local indicators of spatial association (LISA) and Kulldorff’s spatial scan statistic were used to identify clusters. Results Over the 20 years analyzed in this study, the SLE average ASMR for the period was 2.7 for women and 0.7 for men, with a sex ratio (female/male) of 3.8. In men, no province showed a RR>1. Conversely, in women, eight provinces showed values of RR> 1 with a PP greater than 0.8 (Seville, Cadiz, Huelva and Murcia in the south, Barcelona, Zaragoza, Huesca and Leon in the north). In men, neither of the two methods detected a clustering of provinces. However, in women, both methods identified a cluster of provinces located in the southwest of the country (Huelva, Cádiz, Seville and Malaga) as a cluster with significant excess mortality. In the second cluster (centred on the province of Huelva) obtained with the Kulldorff method, two more provinces were added (Badajoz and Cordoba, also located in the southwest). Conclusions We detected a cluster of provinces with an excess risk of female SLE mortality in the southwest of Spain.