Setting/study population

Participants were Somali women and men who took part in a cross-sectional study conducted in 2014 in Oslo using a respondent-driven sampling (RDS) technique. The RDS technique consists of an enhanced snowball sampling, in which information on who recruited whom and the individual social network size, provided the basis for the calculation of inclusion probability and minimally biased population indicators, as well as the variability of these indicators [21, 22]. The RDS was initiated by a small number of purposely recruited seeds, who were given uniquely coded coupons and incentives to recruit their peers to participate in the study. Eligible recruits who participated in the study and redeemed their coupons created the first wave of respondents. The first wave respondents were given incentives to further recruit their peers [21], who then formed the second wave, with the cycle continuing until the proposed sample size was achieved. RDS is largely used in behavioral studies [23], and it is widely acknowledged for its potential in studying immigrant communities whose sampling frame is unavailable [24–27].

Inclusion criteria

Eligibility criteria for the study was being of Somali origin, permanently or legally residing in the Oslo municipality, between the ages 16 to 25 years, willing to participate and able to provide informed consent.

Recruitment process

Four eligible seeds, comprising socially well-connected individuals, were selected based on a diversity of gender, location, years of stay in Norway and age. After providing informed consent, the seeds underwent an interview and were educated on how to further recruit other eligible young Somalis. They were given two uniquely coded coupons to help recruit their peers in their social network. The reason for using only two coupons was to elongate the recruitment waves so that the diversion of subsequent waves from the initial seeds was increased. Each seed proceeded to recruit two persons into their network, which became the first wave, with the first-wave participants further recruiting their peers, which then became the second wave. Each participant was compensated with 100 Norwegian kroner (NOK) for participating in the study and received an additional 100 NOK for each recruited peer who met the eligibility criteria and participated in the study. With 95% level of confidence and 5% margin of error, we obtained a desired sample size of 325. In this paper, the focus will be on the women participants.

Variables

Data analysis

Descriptive statistics in the form of frequencies (percentages) and means with standard deviations were used to summarize the data at baseline. Chi-square tests and Fishers’ exact test (where appropriate) were used to establish an association between the FGM/C status and the socio-demographic variables. Generalized linear regression models (GLMs) with the logit function were used to fit binary responses relating to FGM/C and health care-seeking for FGM/C. Possible demographic factors that were associated with the response variables were identified and explained. The modeling process proceeded in two steps; first, simple GLMs were used to identify socio-demographics factors with P ≤ 0.05 for the FGM/C model. Secondly, variables that were significant (P ≤ 0.05) in the univariate analyses together with independent variables that have been previously shown to be associated with FGM/C were used to fit model 2. However, model 2 for the health care-seeking for FGM/C was based on independent variables with P ≤ 0.20. We used P ≤ 0.20 in order not to miss out important variables. Model 1 (full model) was adjusted for all socio-demographic factors. We used the Akaike information criterion (AIC) to select a better fit between model 1 and model 2. The AIC states that given a set of candidate models, the model with the smallest AIC estimate fits the data better. All analyses were performed using SPSS IBM Statistics 24 and the significance level was set at 0.05.

Sample characteristics

Association between FGM/C and socio- demographic characteristics

Health care-seeking for FGM/C

Health care-seeking

According to AIC estimates, women’s age at migration and the stigma of FGM/C practice were associated with health care-seeking for FGM/C among girls and women who have undergone FGM/C. In our study, we cannot determine which health care service is mostly used for FGM/C, or the reasons why only 20.3% of women who have undergone FGM/C have sought care. However, given the lower proportion of women using health care services, might indicate a lack of need or it could be attributed to women’s perceptions and challenges of care [20, 39].

According to the health belief model, an individual’s state of readiness to take action for a health condition is determine by four dimensions. Firstly, the perceived susceptibility to the condition and the probable severity of the condition (defined in terms of physical harm or interfere with social functioning). Secondly, the perception of benefits associated with actions to reduce the level of threat or vulnerability. Thirdly, the assessment of potential barriers, including physical, psychological and financial barriers, and finally, the general health motivations triggering appropriate health behavior, including internal cues as symptoms and external cues like interpersonal interaction and communication [40–42]. Addressing the above concepts of the health belief models both for the women and the health care system could enhance the number of people who use the health care services and assessing the coverage of health services is important to determine the quality of health care services [43].

According to the Norwegian law, it is the duty of parents to ensure that their daughters who have undergone FGM/C receive the necessary care [16]. In as much as FGM/C has a complex socio-cultural perspective, a decision to seek care for FGM/C-related issues is contextual. In an African context, as FGM/C is associated with culture the decision to seek care lies with the woman herself, her husband and/or relatives. In some settings, health seeking for FGM/C may depend on the availability and the skills of care providers, illness characteristic (recognition and severity), the status of the woman, previous experiences and perceived quality of care [39, 42, 44].

Previous studies have reported FGM/C not to be optimal among migrants in receiving countries. This might be attributed to unskilled care providers in host countries, type of services offered, women’s experiences, patients-caregivers interaction and sometimes the costs of care [39, 44, 45]. Health care-seeking may also be affected by a state of vulnerability and traumatization. Care seeking for FGM/C may either enhance or lessen if the women feel embarrassed about their condition, feel inadequate, shy and tense. Revisiting the health services is possible if women were satisfied with the care they received, or their past experiences [46].

According to prior studies, Somalis, Eritrean and Sudanese women with FGM/C in Norway and Sweden, have complained of insufficient attention, support, lack of respect, lack of interpreters, neglect from health care providers and they were uncomfortable with the unpleasant and hurtful comments from caregivers during delivery [20, 34, 47]. Somali women in England, USA and Australia had expressed the feeling of humiliation and avoided questions from health care providers that triggered flashbacks [48, 49]. Other women with FGM/C were very concerned with the breach of their privacy and confidentiality [50]. It is often presumed that it is mandatory for health professionals to inform circumcised women of the care they should expect. However, in Norway, circumcised women complained of lack of information regarding what to expect during delivery [34].

While our data cannot provide the explanations for the findings it is highly likely that the women in our study lack the information and knowledge about access, availability and functioning and navigation of FGM/C related health care services in Norway. Therefore, challenges of care may not only affect the women who are new to Norway, but also those who have been living here for a long period. Furthermore, communication can be a potential barrier to care seeking. Because 96.2% of the circumcised women were born out Norway, language can hinder their access to seeking care, as language and poor communication have been reported by care providers and women to be a barrier to effective care [39, 51]. Lack of a well-functioning referral system [52] and a good social support network may also influence access to care. A good support network has been shown to empower FGM/C affected women in accessing antenatal and intrapartum services in England [48]. While some barriers are potentially easier to overcome, there remain some serious concerns regarding the perceptions and experiences of FGM/C affected women that calls for further research.

A step towards health care-seeking might be a step in the direction of the discontinuation of FGM/C, as 85.7% of FGM/C affected women from Somalia, have previously indicated their interest in the discontinuity of FGM/C practice [53]. We observed FGM/C prevalence of 51.6%, in our study indicating a significant decline among the younger age groups, compared to a higher prevalence of 79.3% among older age group in a prior study [17]. This is consistent with other reports of the discontinuation of FGM/C in Somali communities, both at home and abroad [50, 54, 55]. Readiness to seek help might indicate that these women trust the patient-centered care and/or the Norwegian health care system and that the health personnel are responsive to their needs and values. It could also indicate that the health care centers/ support centers might provide solace as well as access to information and educational resources.

Although fewer numbers of women used the health care services, it is worth highlighting that there was a significant association between women who felt stigmatized by the practice and health care-seeking. Although stigmatization has often been pointed out as a deterrent to seeking medical care among FGM/C affected women [9, 47, 56, 57], it is noteworthy stigma was not a barrier to health care-seeking in the present study. As most (56.7%) of these women have at least secondary education, they may have acquired knowledge of the negative effects of the practice through awareness campaigns in Norway. Good knowledge along with willingness to seek care despite the stigma of the practice could indicate changing attitudes and perceptions towards FGM/C. It may also indicate a readiness for compliance to address their health needs [58].

The age at migration to Norway influenced women’s use of FGM/C health care in our study. Those who arrived at a young age were less likely to seek care, one would expect that they may not need care or may not know if the services are provided. Even if they are aware of the services, they may be faced with difficulty in navigating the health care system. The fear of criminalization has been previously documented as a limitation to accessing much needed quality health services in host countries [9, 57]. However, we are unable to study this association in our study population as most of the affected women were circumcised prior to migration and we did not ask any direct questions related to the fear of criminalization as being a barrier to health seeking.

Strength and limitation of the study

The strength of the study lies in using the RDS methodology, pretesting of data collection tools, recruitment of field workers trusted by the Somali community and adequate training of the field workers. Because the principal investigator and the data collectors spoke the same language, this rules out bias due to translation. While it is impossible to rule out selection bias using our sampling methods, nonetheless, the findings provide valuable information about those seeking care for their health problems. However, our study is limited in providing explanations as to why this is the case. We have not ascertained information such as the degree of FGM/C or the symptoms and that limits our interpretations.