Cervical cancer is the third most common cancer among women world-wide with an estimated 471,000 cases and 230,000 deaths occurring per year, more than 80% of which occur in developing countries.1 An overview of 21 published studies reported a consistent increasing trend in cervical cancer incidence with various indicators of decreasing social class, including level of education, income and occupation.2 The main explanations advanced for the excess risk among lower socio-economic groups are related to sexual behaviour, corresponding to a greater chance to acquire and/or become chronic carriers of human papillomavirus (HPV), which is a necessary cause of cervical cancer. Less access to early diagnosis or cytological screening for cervical cancer is also likely to result in a higher incidence and mortality.
One problem highlighted by the overview of previous studies was that the majority have been based on record linkage between mortality or cancer registry incidence data and regional measures of social class based on census data. These routine data are usually not collected at the individual level and are liable to substantial misclassification. In addition, the studies are primarily from Western Europe and North America, whereas the major burden of cervical cancer is in developing countries. To overcome some of these deficiencies, we have attempted to identify all previously reported case-control studies of cervical cancer or dysplasia and screen them for information on socio-economic characteristics. Furthermore, we have investigated the relationship between cervical cancer and socio-demographic characteristics separately for stage of disease, geographical region, age and histological type.