Gastric cancer incidence and geographical variations: the influence of gender and rural and socioeconomic factors, Zaragoza (Spain)

Aguilar I, Compés L, Feja C, Rabanaque MJ, Martos C


BACKGROUND: Socioeconomic status (SES) and rural factors have been shown to be associated with gastric cancer epidemiology. The aim of this study was to identify geographical variations in gastric cancer incidence in Zaragoza province (Spain) during the period 1993-2002, and their association with SES and rural factors.

METHODS: Incident cases were extracted from the population-based Zaragoza Cancer Registry. The geographical analysis unit was the census tract (CT) in Zaragoza city (N = 462) and the municipalities for the rest of the province (N = 292). Four indexes were applied: two deprivation and two rurality indexes, included in a Bayesian risk model discretized in quartiles. Standardized incidence ratios (SIRs) were calculated using the incidence rates in Spain. SIRs were adjusted by a Bayesian generalized linear mixed model (GLMM).

RESULTS: From 1993 to 2002, 1,309 cases of gastric cancer were registered in Zaragoza city and 578 in the rest of the province. High risk was observed in CTs for the peripheral areas of the city. The incidence risk in men was 2 (95 % confidence interval [CI] 1.22-2.98) times higher in the most deprived CTs compared with the least deprived CTs, but no statistically significant differences were found in women. Municipalities with higher risk were observed in the north of the province, but no significant association was found with SES. Regarding the rurality index, a positive trend was observed in women, but it was statistically significant only for the most rural quartile (2.49, 95 % CI 1.07-4.92).

CONCLUSIONS: Geographical differences in gastric cancer incidence were detected. Although these differences could be partially explained by the deprivation index for men in Zaragoza city, deprivation index cannot explain geographical differences for women. In the rest of the province, the rurality index 1991 could explain, at least for women, geographical differences. It is still necessary to develop a deprivation index suitable for small municipalities.