Cancer incidence in young people in Saudi Arabia: relation to socioeconomic status and population mixing

This study describes cancer incidence in under 24 year olds, particularly leukaemias, lymphomas and central nervous system tumours. It also describes the socioeconomic status (SES) of the geographically delimited Governorates in Saudi Arabia, by deriving two indices – the first time this has been done in the country. It also sought to determine whether SES and Hajj (occurring in Makkah) as a measure of population mixing has an association with the incidence of these cancers.
During 1994 to 2008, 17,150 cases were identified from the Saudi Cancer Registry. Census data were accessed for 2004 and included 29 indicators. A continuous SES index was constructed using exploratory factor analysis (EFA) and a categorical index using latent class analysis (LCA). Incidence rate ratios (IRRs) were calculated for cancers in Makkah compared to other Governorates by year to assess the effect of Hajj, and for all Governorates to assess the effect of SES.
The Hajj had no significant effect on the incidence for all cancer groups. The continuous index produced by EFA consisted of scores ranging from 100 to 0, for affluent to deprived Governorates. The LCA found a four-class model as the best model fit. Class 1 was termed ‘affluent’, Class 2 ‘upper-middle’, Class 3 ‘lower-middle’ and Class 4 ‘deprived’. The urbanised Governorates were affluent, whereas the rural Governorates were on average more deprived. For SES, an elevated risk was found for acute lymphoblastic leukaemia in the affluent class (IRR=1.38, 95%CI=1.23-1.54), and was reduced in the deprived class (IRR=0.17, 95%CI=0.10-0.29). Similar associations were observed for all cancer groups.
The findings are not supportive of the PM hypothesis, but give support to the delayed infection hypothesis, suggesting that delayed exposure to infections may prevent immune system modulation, although results may be exacerbated by poor case-ascertainment/under-diagnosis in deprived areas. Similarities between the two indices suggest validity.