The association between CCS and migration
The association between CCS and Ko 143 status was not explained away by the range of confounding factors included in our model, one of which was a detailed measure of SES. We could not account for important factors that could be more prevalent among women of foreign origin and that might account for their having the lowest CCS rate, namely a lack of or incomplete health insurance, unfamiliarity with the health-care system and/or with CCS, poor health literacy, linguistic barriers, having experienced migration-related discrimination, and different socioculturally based health beliefs and attitudes towards medical care, as well as low levels of trust in doctors and the health-care system [8,16,17,, , ]. In addition, we used a crude measure to account for gynaecological follow-up, a major determinant of CCS, and residual confounding is therefore likely to remain. Our study participates in the elucidation of the role of BMI in the association between migration status and CCS. We found that BMI did not contribute to the association between migration origin and CCS, with identical estimators when this characteristic was or was not included in the regression model. Differences in the distribution of BMI according to migration status might not be large enough to explain participation in CCS. Indeed, migrant women form a heterogeneous group with respect to BMI [11,12]. For instance, a Swedish study reported higher obesity rates among all migrant women than among women who were native of the host country, but the rates were especially high among those from Arabic-speaking countries and Poland . This heterogeneity was also reported in a French study conducted in the Paris region, where, compared to women with two French parents, higher overweight and obesity rates were found only among women from Northern and Sub-Saharan Africa but not among women from other countries or those with one foreign parent . Our findings are consistent with this French survey. However, women of African origin accounted for only 1.3% (naturalized) and 0.7% (foreigner) of the women included in our sample, which could explain why BMI did not contribute to the association between migration status and CCS among all the women in our data. Association between migration status and CSS was more pronounced among obese than among normal-weight or overweight women. Obesity seemed to potentiate the lack of participation in CCS among the women of foreign nationality. Discrimination and the health care provider-woman relationship may explain this finding. First, obese migrant women face double discrimination in access to care and CCS because of their high BMI and their migrant origin. This double discrimination may potentiate the effects of each type of discrimination, in particular, because this group is also more vulnerable due to its low SES. Second, woman’s satisfaction with and adherence to medical recommendations are closely linked to the health care provider-patient communication and interaction. Obese women report a perceived lack of respect from health care providers regarding their weight and associated health behaviours or want to avoid unwanted weight loss advice, both of which are barriers to CCS . Furthermore health care providers report technical difficulties and reluctance in performing pelvic examination and Pap tests on obese women. Migrant women report low levels of trust in physicians and the health-care system, which is partly due to sociocultural discordance between the physician and patient [16,17,40]. As for discrimination in care, these two factors could negatively affect each other and account for the extremely low CCS rate reported among obese migrant women.