Study design and participants
This was a prospective multicenter and observational study, carried out between January 2017 and March 2020 among migrants in 21 centers of the French Office for Immigration and Integration (OFII) including: 3 centers in Ile-de-France (Cergy, Melun, Montrouge), 16 centers outside Ile-de-France (Bordeaux, Dijon, Grenoble, Lille, Limoges, Lyon, Marseille, Montpellier, Nantes, Nice, Orléans, Rennes, Reims, Rouen, Strasbourg, Toulouse) and 2 overseas centers (Cayenne, Pointe-à-Pitre). Individuals were included during the compulsory medical visit at time of the delivery of their first residence permit. Eligible participants were migrants aged 18 years or more and who consented to participate to this study.
Study Intervention And Data Collection
Study intervention and data collection
Participants completed online the anonymous TROD Screen questionnaire. This self-administrated questionnaire was translated into 10 languages (English, Arabic, Chinese, Bengali, Russian, Lingala, Portuguese, Spanish, Turkish, Haitian Creole) and included data related to sociodemographic, personal medical history and sexual behaviors. A few data were retrieved from medical records (year of birth, gender, height, weight and nationality).
A nurse performed the rapid screening test and the doctor or nurse announced the results. In case of a positive result, participant was offered to be referred to a specialized hospital consultation to confirm the diagnosis and initiate adapted treatments.
Variables Definition
The outcome was a positive rapid screening test for HBV, HCV or HIV infections. It was used as the gold standard for calculating the sensitivity and specificity of various combination of participants’ characteristics for predicting HBV, HCV or HIV infection.
A number of independent predictor factors were used for the prediction of HBV, HCV or HIV infection. (1) Sociodemographic characteristics: age, (years), gender (male or female), weight status (Underweight; Normal weight or Overweight/ Obesity), endemic area for each infection knowledge of HIV, HBV and HCV. (2) Personal history: HIV, HBV or HCV screening, HBV vaccination, dental treatment, surgery, abortion, caesarean section or difficult childbirth, history of liver disease, tattoos or piercings, prison, blood transfusion, living with a person infected with viral hepatitis, drug use. (3) Sexual behaviors: geographical origin of sexual partner, number of sexual partners during the last 12 months and sexual practices and orientation.
Statistical analysis
Statistical analyses were performed using R software version R 3.6.3. The participants’ characteristics were described using absolute frequencies, proportions for categorical variables, or means and standard deviation (SD) for continuous variables.
A cross-analysis between explanatory factors and each infection (HBV, HCV and HIV) was performed using a Student's t test or Wilcoxon for the means and a Chi-square or Fisher test for the proportions
Binary logistic regression models were fitted to identify factors associated with HBV, HCV or HIV infection. In the univariate analysis, independent variables with p-value less than 0.25 were included in multivariable logistic regression analysis in order to control potential confounders. The final multivariable model was performed using a stepwise selection procedure, which was based on the likelihood ratio test (p-value < 0.05). The Akaike information criterion (AIC) was to select the final model. Results were reported as adjusted odds ratios (aOR) with 95% confidence intervals (CI).
The discriminating capacity of the TROD Screen questionnaire for HBV, HCV and HIV was evaluated by using the predictive value of the questionnaire, its sensitivity, its specificity as well as the 95% confidence intervals. A ROC curve was used to determine the cutoff score of the questionnaire for each infection.
By assuming, that the HBV prevalence is higher in men than in women, we carried out the analysis on HBV infection by stratifying by sex, in order to obtain risk scores specific to the men and women included in this study.