Cervical cancer screening among immigrants in Norway : Challenges, possibilities and the effect of an intervention

Abstract

The aims of this thesis were: (i) to determine immigrants’ participation rate to CCS compared to non-immigrants and predictors to take the CCS-test, (ii) to obtain knowledge of HCPs perceptions regarding CCS test among immigrants and how they overcome barriers, strategies being used, if any, and (iii) to measure the effect of the intervention conducted at general practices as a cluster randomized controlled trial. Being a mixed method study, data from different sources, both quantitative and qualitative, were collected and analyzed for this thesis. Norway has well established national registers which we took advantage of. The first paper was a cross sectional register-based study using the National Population Registry (NPR), the Norwegian Health Economics Administration Database (HELFO), the GP database (“Fastlegedatabasen”), and the Medical Birth Registry (MBR). We grouped the immigrants by world’s geographic region, carried out descriptive analyses and constructed several logistic regression models. The main outcome variable was whether the woman was registered with a CCS-test or not. This study was part of a bigger registry study “Immigrant health in Norway” and thus data already available from 2008. The second paper presents a qualitative study conducted in 2016, where we used thematic analysis to study three focus groups among general practitioners, and four semi-structured personal interviews; two among gynaecologists and two among midwives. Based on the results of these interviews, a literature review and findings from focus groups conducted among Somali and Pakistani women (focus groups among immigrant women were conducted parallelly by other members of our research group, and are not a part of this thesis), we developed an intervention to increase the participation of immigrants to CCS that was tested using a cluster-randomized study design. The intervention targeted general practices in the clusters and immigrant women were the units of analysis. The intervention consisted of (i) an educational session for GPs; about immigrants’ lower attendance to CCS, some groups having higher prevalence of cervical cancer, and urging GPs to ask about CCS when immigrant women came for consultations for other reasons, (ii) distribution of a mouse pad for GPs in order to remind them of the intervention in their everyday work, and (iii) a poster with a message in four languages (Somali, Polish, Urdu and English) to be placed in the waiting rooms. The intervention was implemented from January 2017 to June 2017, and its effect is presented in the third paper. The main outcome variable was status of screening of the women by January 2018 measured by means of data linked from NPR, GP-database and Norwegian Cancer Registry. Our study confirms that there is lower participation to CCS program in Norway among immigrants compared to non-immigrants. Higher income, residence in rural areas, and having a female GP were associated positively with CCS-test for both immigrants and non-immigrants. The focus groups and interviews among HCPs revealed several challenges related to CCS. While some barriers were common for both immigrants and non-immigrants such as GPs’ understanding of routines and responsibilities for prevention, others were aspects specific for immigrants related to organization of appointments, language, health literacy levels, culture and gender. Some HCPs described several strategies that they already tried to implement to address the existing barriers, such as having longer consultations (organization), using interpreters (language), using anatomy models to explain (health literacy) or dealing differently with the expression of pain (culture). The intervention had a statistically significant effect, both measured as relative effect (OR (95% CI)) 1.24 (1.11-1.38), and as absolute effect (RD (95%CI)) 2.6 (1.1-4.0) adjusted for baseline screening. In addition, in subgroup analyses, the intervention particularly increased participation among women who were not screened at baseline of the intervention and among some specific immigrant groups. Our study presents a feasible intervention in general practice that can increase the participation of immigrants to CCS. However, the effect is clinically small, and we do not know how long it might last or its cost-effectiveness. Thus, we suggest further research including; piloting of measures that facilitate other primary care providers outside general practices, for example midwives, to perform the task, long-term evaluations and studying cost-effectiveness. Migrant health is public health. Public health studies targeting ethnic diversity are necessary to make effective and good policies for achieving health equity. Our randomized controlled study may also be used as a model to increase CCS among immigrants in other HIC. This could also be piloted among primary care providers in HIC to target other health issues where the gap between the majority population and immigrants should be bridged.