Search

Background The attendance to cervical cancer screening is low among immigrants in many high-income countries. Although several interventions have been experimentally tested,implementation remains a challenge. Several factors are an impediment, including the lack of methodological descriptions of the development and implementation of such interventions. In this paper,we present in detail the development, methodological challenges and practical implementation of a community based intervention aimed to increase the participation of immigrant women in cervical cancer screening in Norway. Methods This study was initially designed as a cluster randomized trial to be carried out in four geographical areas near Oslo between Feb-October 2017. Participants were immigrant women aged 25–69 years from Pakistan and Somalia. This paper describes the theoretical background for the development of the intervention,followed by challenges,the changes in the original design and solutions adopted related to the study design,recruitment and implementation of the intervention. The intervention was developed based on two theoretical frameworks, the Ecological and the Heron’s six categories intervention framework. An oral 20–25 min presentation in the language of participants encompassing topics of cervical cancer and screening was given according to the needs detected in focus groups conducted at the beginning of the study,followed by an opportunity to raise questions and answering a short questionnaire. Results Contrary to the initial study design, this had to be converted into a non-randomised trial due to the difficulties associated with randomization of immigrant families who are finely scattered in heavily populated towns and a high risk of contamination. We therefore adopted a pragmatic approach and recruited women in the intervention areas through a variety of channels and institutions. Neighboring areas were considered to be non-randomised controls. Female researchers with Pakistani and Somali background invited as many women as possible in the intervention areas. Among the women who were invited to participate,42% of the Pakistani and 78% of Somali attended the meetings. Conclusion Despite the careful development of a culturally adapted health intervention in collaboration with the community; randomization and recruitment of immigrants for community trials remains challenging. Nevertheless, sharing strategies to overcome specific challenges related to promoting health interventions for immigrants, can be of potential help to scale-up interventions and for building new research projects. Trial registration NCT03155581. Retrospectively registered, on 16 May 2017. Electronic supplementary material The online version of this article (10.1186/s12874-019-0795-8) contains supplementary material, which is available to authorized users.

Introduction Mammographic density (MD) is one of the strongest known breast cancer risk factors. Estrogen and progestin therapy (EPT) has been associated with increases in MD. Dense breast tissue is characterized by increased stromal tissue and (to a lesser degree) increased numbers of breast epithelial cells. It is possible that genetic factors modify the association between EPT and MD, and that certain genetic variants are particularly important in determining MD in hormone users. We evaluated the association between MD and 340 tagging single nucleotide polymorphisms (SNPs) from about 30 candidate genes in hormone metabolism/growth factor pathways among women who participated in the Norwegian Breast Cancer Screening Program (NBCSP) in 2004. Methods We assessed MD on 2,036 postmenopausal women aged 50 to 69 years using a computer-assisted method (Madena, University of Southern California) in a cross-sectional study. We used linear regression to determine the association between each SNP and MD, adjusting for potential confounders. The postmenopausal women were stratified into HT users (EPT and estrogen-only) and non-users (never HT). Results For current EPT users, there was an association between a variant in the prolactin gene (PRL; rs10946545) and MD (dominant model, Bonferroni-adjusted P (Pb) = 0.0144). This association remained statistically significant among current users of norethisterone acetate (NETA)-based EPT, a regimen common in Nordic countries. Among current estrogen-only users (ET), there was an association between rs4670813 in the cytochrome P450 gene (CYP1B1) and MD (dominant model, Pb = 0.0396). In never HT users, rs769177 in the tumor necrosis factor (TNF) gene and rs1968752 in the region of the sulfotransferase gene (SULT1A1/SULT1A2), were significantly associated with MD (Pb = 0.0202; Pb = 0.0349). Conclusions We found some evidence that variants in the PRL gene were associated with MD in current EPT and NETA users. In never HT users, variants in the TNF and SULT1A1/SULT1A2 genes were significantly associated with MD. These findings may suggest that several genes in the hormone metabolism and growth factor pathways are implicated in determining MD.