A Review of Health Survey Research for People with Refugee Background Resettled from Africa: Research Gaps and Methodological Issues.

We reviewed 27 studies on adults with a refugee background resettled from Africa published between 1999 and 2017 to appraise their methodological issues for survey research. Out of 27 studies, eleven used a single sampling method (referral = 1, convenience = 10), and 16 relied on multiple sampling methods, many of which were combinations of referral and convenience. The two most salient recruitment strategies found were building trusted relationships with the community (n = 15), and using recruiters who were culturally and linguistically matched to the refugee communities of interest (n = 14). Fifteen studies used existing data collection instruments, while in 13 studies, researchers developed their own data collection instruments. In-person or phone interviews using bilingual interviewers (n = 21) were the most frequently used mode of data collection, followed by a self-administered survey (n = 7). Our review presents methodological gaps in current refugee health studies, such as limited use of probability sampling approach due to system barriers, lack of information in community engagement and recruitment processes, and insufficient considerations of unique culture and experiences of refugee communities when developing or adapting the instruments. Efforts can be made to guide and facilitate appropriate reporting and development of more scientifically robust survey methodologies for refugee health studies, as well as to improve registration system infrastructure that may help identify these hidden populations more effectively.

Predictors of timing of pregnancy discovery.

OBJECTIVE: Earlier pregnancy discovery is important in the context of prenatal and abortion care. We evaluated characteristics associated with later pregnancy discovery among women seeking abortion care. STUDY DESIGN: Data come from a survey of women seeking abortion care at four family planning facilities in Utah. The participants completed a survey during the state-mandated abortion information visit they are required to complete prior to having an abortion. The outcome in this study was pregnancy discovery before versus after 6 weeks since respondents' last menstrual period (LMP). We used logistic regression to estimate the relationship between sociodemographic and health-related independent variables of interest and pregnancy discovery before versus after 6 weeks. RESULTS: Among the 458 women in the sample, 28% discovered their pregnancy later than 6 weeks since LMP. Most (n=366, 80%) knew the exact date of their LMP and a significant minority estimated it (n=92, 20%). Those who estimated the date of their LMP had higher odds of later pregnancy discovery than those who knew the exact date (adjusted odds ratio (aOR)=1.81[1.07-3.07]). Those who used illicit drugs weekly, daily, or almost daily had higher odds of later pregnancy discovery (aOR=6.33[2.44, 16.40]). CONCLUSION: Women who did not track their menstrual periods and those who frequently used drugs had higher odds of discovering their pregnancies later. IMPLICATIONS: Women who estimated the date of their LMP and who frequently used drugs may benefit from strategies to help them recognize their pregnancies earlier and link them to care when they discover their pregnancies later.

Influence of clinician referral on Nebraska women's decision-to-abortion time.

OBJECTIVE: To assess the association of clinician referral with decision-to-abortion time. STUDY DESIGN: We conducted a cross-sectional survey of women seeking abortion at all three Nebraska abortion clinics. We defined referral as direct (information for an abortion clinic), inappropriate (information for a clinic that does not provide abortions) or no referral. Women reported when they recognized their pregnancy, decided to seek abortion and contacted a clinician. The primary outcome - decision-to-abortion time - was time from certain decision to abortion. We used multivariate linear regression analysis, controlling for potential confounders. RESULTS: Participants (n=356) were a mean of 26.8±5.3years old, primarily white (62%), unmarried (88%) and urban (87%), with a mean gestational duration of 8(2/7)weeks (S.D.±20days). Forty-six percent (164) had contacted a clinician and 30% (104) had discussed abortion with one before their abortion. Of those, 30% received a direct referral, 6% received an inappropriate referral and 64% received no referral. Decision-to-abortion time did not vary by referral type [mean difference compared with direct referral: inappropriate referral, 1.1days, 95% confidence interval (CI) -13.4 to 15.6, p=.88; no referral, -0.4days, 95% CI -7.0 to 6.3]. The most common reasons cited for delay in obtaining an abortion were an inability to get an earlier appointment (105/263, 40%) and time needed to raise money to pay for the abortion (73/263, 28%). CONCLUSION: While neither occurrence of referral nor type was associated with decision-to-abortion times, women in Nebraska continue to face barriers to timely abortion care. IMPLICATIONS: Additional research is needed to explore whether quality clinician referral improves abortion access and whether increased resources should be dedicated to improving referral patterns.

Parental perception of child bodyweight and health among Mexican-American children with acanthosis nigricans.

Acanthosis nigricans (AN) is a cutaneous marker associated with elevated risk of type 2 diabetes. This study assesses mother-father differences in perception of child's bodyweight and health by Mexican-American parents with AN-positive children. The study used medical records in conjunction with survey data collected between 2011 and 2012 for 309 Mexican-American children with AN in South Texas. Multivariate logit models were estimated to assess mother-father differences in perception of child bodyweight and health controlling for selected child- and parent-level covariates. About 91 % of the children in the sample were obese and 6.5 % were overweight. One fifth of mothers and 38.5 % of fathers in the sample expressed no concern of their children's bodyweight. After adjusting for selected explanatory variables at both the child and parent level, the odds for fathers, relative to mothers, to be concerned about child's bodyweight were 82 % lower (OR = 0.18, p < 0.05). Similar findings also hold for parental awareness of child's AN (OR = 0.19, p < 0.05). Among Mexican-American families with AN-positive children, the lack of concern over child's bodyweight, unawareness of AN, and misconception of child's health on the part of many parents, especially of fathers, constitutes a challenge to diabetes prevention. Health education programs targeting Mexican-American families with AN-positive children might be more cost effective to consider mother-father differences in perception of child health and bodyweight.