Dependence-Robust Confidence Intervals for Capture-Recapture Surveys.

Capture-recapture (CRC) surveys are used to estimate the size of a population whose members cannot be enumerated directly. CRC surveys have been used to estimate the number of Coronavirus Disease 2019 (COVID-19) infections, people who use drugs, sex workers, conflict casualties, and trafficking victims. When k-capture samples are obtained, counts of unit captures in subsets of samples are represented naturally by a 2k contingency table in which one element-the number of individuals appearing in none of the samples-remains unobserved. In the absence of additional assumptions, the population size is not identifiable (i.e., point identified). Stringent assumptions about the dependence between samples are often used to achieve point identification. However, real-world CRC surveys often use convenience samples in which the assumed dependence cannot be guaranteed, and population size estimates under these assumptions may lack empirical credibility. In this work, we apply the theory of partial identification to show that weak assumptions or qualitative knowledge about the nature of dependence between samples can be used to characterize a nontrivial confidence set for the true population size. We construct confidence sets under bounds on pairwise capture probabilities using two methods: test inversion bootstrap confidence intervals and profile likelihood confidence intervals. Simulation results demonstrate well-calibrated confidence sets for each method. In an extensive real-world study, we apply the new methodology to the problem of using heterogeneous survey data to estimate the number of people who inject drugs in Brussels, Belgium.

Univariable associations between a history of incarceration and HIV and HCV prevalence among people who inject drugs across 17 countries in Europe 2006 to 2020 - is the precautionary principle applicable?

BackgroundPeople who inject drugs (PWID) are frequently incarcerated, which is associated with multiple negative health outcomes.AimWe aimed to estimate the associations between a history of incarceration and prevalence of HIV and HCV infection among PWID in Europe.MethodsAggregate data from PWID recruited in drug services (excluding prison services) or elsewhere in the community were reported by 17 of 30 countries (16 per virus) collaborating in a European drug monitoring system (2006-2020; n = 52,368 HIV+/-; n = 47,268 HCV+/-). Country-specific odds ratios (OR) and prevalence ratios (PR) were calculated from country totals of HIV and HCV antibody status and self-reported life-time incarceration history, and pooled using meta-analyses. Country-specific and overall population attributable risk (PAR) were estimated using pooled PR.ResultsUnivariable HIV OR ranged between 0.73 and 6.37 (median: 2.1; pooled OR: 1.92; 95% CI: 1.52-2.42). Pooled PR was 1.66 (95% CI 1.38-1.98), giving a PAR of 25.8% (95% CI 16.7-34.0). Univariable anti-HCV OR ranged between 1.06 and 5.04 (median: 2.70; pooled OR: 2.51; 95% CI: 2.17-2.91). Pooled PR was 1.42 (95% CI: 1.28-1.58) and PAR 16.7% (95% CI: 11.8-21.7). Subgroup analyses showed differences in the OR for HCV by geographical region, with lower estimates in southern Europe.ConclusionIn univariable analysis, a history of incarceration was associated with positive HIV and HCV serostatus among PWID in Europe. Applying the precautionary principle would suggest finding alternatives to incarceration of PWID and strengthening health and social services in prison and after release ('throughcare').

COVID-19 and people who use drugs: Impact of the pandemic on general anxiety and depressive disorders among adults in Belgium.

BACKGROUND: During previous pandemics people who use drugs (PWUD) were categorized among the most vulnerable. In the current study, firstly, we wanted to evaluate the impact of the COVID-19 crisis on the prevalence of anxiety and depressive disorders among PWUD. Furthermore, we wanted to compare the prevalence of these disorders with that of members from the general population who did not use drugs. METHODS: We used a matched cohort design based on two separate repeated cross-sectional online surveys (April and November 2020) among PWUD and the general population. Results of GAD-7 and PHQ-9 were used as outcome variables. We calculated absolute and relative risks for matched pairs for both affective disorders, and logistic regression to compare affective disorders over both waves for PWUD. RESULTS: In April, the prevalence of affective disorders was similar for PWUD and the general population. In November, the risks for anxiety disorders increased with 64% for PWUD compared to non-PWUD (RR = 1.64, 95%CI 1.42-1.88), whereas the risks for depressive disorders more than doubled (RR = 2.29, 95%CI 1.97-2.67). Having a job and being male were protective factors for PWUD for both anxiety and depressive disorders. LIMITATIONS: As this study used self-reported data, GAD-7 and PHQ-9 give an indication of the presence of anxiety and depression which might differ from a clinician's judgement. CONCLUSIONS: PWUD might be disproportionally affected by COVID-19. Health care providers should be attentive to substance use as an indicator for increased risk of mental health problems.

A multicentre interventional study to assess blood-borne viral infections in Belgian prisons.

BACKGROUND: Prevalence data on viral hepatitis B (HBV), hepatitis C (HCV), and HIV infection in prison are often scarce or outdated. There is currently no systematic screening for these blood-borne viral infections (BBV) in Belgian prisons. There is an urgency to assess the prevalence of these BBV to inform policymakers and public healthcare. METHODS: This was a multicentre, interventional study to assess the prevalence of BBV using opt-in screening in prisons across Belgium, April 2019 - March 2020. Prisoners were tested using a finger prick and BBV risk factors were assessed using a questionnaire. A generalized linear mixed model was used to investigate the association between the various risk factors and HCV. RESULTS: In total, 886 prisoners from 11 Belgian prisons were screened. Study uptake ranged from 16.9 to 35.4% in long-term facilities. The prevalence of HCV antibodies (Ab), hepatitis B surface antigen (Ag) and HIV Ab/Ag was 5.0% (44/886), 0.8% (7/886), and 0.2% (2/886). The adjusted odds for HCV Ab were highest in prisoners who ever injected (p < 0.001; AOR 24.6 CI 95% (5.5-215.2). The prevalence of detectable HCV RNA in the total cohort was 2.1% (19/886). Thirteen (68.4%) prisoners were redirected for follow-up of their HCV infection. CONCLUSIONS: Opt-in testing for viral hepatitis B, C and HIV was relatively well-accepted in prisons. Compared with the general population, prisoners have a higher prevalence of infection with BBV, especially for HCV. Systematic screening for these BBV should be recommended in all prisons, preferably using opt-out to optimize screening uptake. TRIAL REGISTRATION: Retrospectively registered at clinical trials NCT04366492 April 29, 2020.

Use of health care services by people with substance use disorders in Belgium: a register-based cohort study.

BACKGROUND: The objective of the study was to describe the frequencies of health-care utilization by people with substance use disorder (SUD), including contacts with general practitioners (GP), psychiatrists, emergency departments (ED) and hospital admissions and to compare this frequency to the general population. METHODS: Data from the national register of people who were in treatment for SUD between 2011 and 2014 was linked to health care data from the Belgian health insurance (N = 30,905). Four comparators were matched on age, sex and place of residence to each subject in treatment for SUD (N = 123,620). Cases were further divided in five mutually exclusive categories based on the main SUD (opiates, crack/cocaine, stimulants, cannabis and alcohol). We calculated the average number of contacts with GP, psychiatrists and ED, and hospital admissions per person over a ten year period (2008-2017), computed descriptive statistics for each of the SUD and used negative binomial regression models to compare cases and comparators. RESULTS: Over the ten-year period, people in treatment for SUD overall had on average 60 GP contacts, 3.9 psychiatrist contacts, 7.8 visits to the ED, and 16 hospital admissions. Rate ratios, comparing cases and corresponding comparators, showed that people in treatment for SUD had on average 1.9 more contacts with a GP (95 % CI 1.9-2.0), 7.4 more contacts with a psychiatrist (95 % CI 7.0-7.7), 4.2 more ED visits (95 % CI 4.2-4.3), and 6.4 more hospital admissions (95 % CI 6.3-6.5). CONCLUSIONS: The use of health services for people with SUD is between almost two (GP) and seven times (psychiatrist) higher than for comparators. People in treatment for alcohol use disorders use health care services more frequently than people in treatment for other SUD. The use of health services remained stable in the five years before and after the moment people with SUD entered into treatment for SUD. The higher use of primary health care services by people with SUD might indicate that they have higher health care needs than comparators.

Estimates of people who injected drugs within the last 12 months in Belgium based on a capture-recapture and multiplier method.

BACKGROUND: For Belgium, available estimates of the number of people who inject drugs (PWID) are based on data from more than fifteen years ago and apply only to those who report ever injecting drugs. As a result, no reliable baseline data exist to determine the scale of services for PWID. METHODS: We obtained pseudo-anonymized identifier information from treatment and harm reduction service providers and a fieldwork study between February and April 2019 in Brussels. We estimated the number of PWID, defined as people who injected within the last 12 months, in Brussels using capture-recapture (CRC) methodology. To obtain national estimates, we scaled the proportion of PWID in Brussels to the total number of this population in Belgium based on two existing drug treatment registers, which were then multiplied with the result of the CRC. RESULTS: The total population of PWID is estimated to be 703 (95 %CI 538-935) for Brussels and between 6620 (95 %CI 4711 - 8576) and 7018 (95 %CI 4794 - 9527) for Belgium. CONCLUSIONS: These estimates provide crucial information to ensure that services to PWID are adequately maintained. They clearly indicate the need to maximize efforts to achieve the targets set by WHO for 2030 on the provision of 300 sterile needles and syringes per PWID per year, a 90 % reduction of new HCV infections, and a 65 % reduction of liver-related mortality.

General practice patients starting treatment for substance use problems: observations from two data sources across levels of care.

BACKGROUND: In Belgium, the incidence of treatment episodes for substance use problems is monitored by the Network of Sentinel General Practices (SGP), and at higher, specialist care levels by the Treatment Demand Indicator (TDI) surveillance. Using both data sources, we examine 1) how patients starting specialist treatment for substance use problems on referral by their GP compare to those that were referred by non-GP caregivers; 2) how patients starting GP treatment for substance use problems without receiving concurrent specialist treatment compare to those who did. METHODS: Both surveillances are based on the TDI protocol for reporting data to the European Monitoring Centre for Drugs and Drug Addiction (EMCDDA) on individuals starting treatment as a result of their substance use. Data from 2016 and 2017 were examined using 95% confidence intervals and multivariate logistic regression. RESULTS: According to TDI-data (n = 16,543), determinants of being referred by a GP (versus by a non-GP caregiver) for specialist treatment were age ≥ median (OR 1.25; 95% CI 1.13-1.38), education ≥ secondary level (OR 1.27; 95% CI 1.15-1.41), recent employment (OR 1.71; 1.56-1.88), recent stable accommodation (3.62; 95% CI 3.08-4.26), first treatment episode (OR 1.72; 95% CI 1.57-1.87), recent daily primary substance use (OR 1.46; 95% CI 1.33-1.59) and mono substance use (OR 1.23; 95% CI 1.04-1.48). Type of substance use was a significant determinant with higher odds of using pharmaceuticals (and alcohol) (OR 1.24; 95% CI 1.04-1.48), and lower odds of using cannabis only/primarily (OR 0.73; 95% CI 0.62-0.86), with reference to street drugs minus cannabis only/primarily. According to SGP data (n = 314), determinants of starting GP treatment without concurrent specialist treatment were recent employment (OR 2.58; 95% CI 1.36-4.91), first treatment episode (OR 2.78; 95% CI 1.39-5.55) and living in the Brussels or Walloon region (OR 1.97; 95% CI 1.06-3.66). CONCLUSIONS: This study provides a useful insight into the general practice population treated for substance use problems. It shows that both surveillances consistently found a relatively favourable profile of general practice patients with substance use problems.

Prevalence of HCV among people who inject drugs in Brussels-a respondent-driven sampling survey.

BACKGROUND: In Belgium, people who inject drugs (PWID) are at a high risk of being infected by hepatitis C (HCV) as injecting drug use is the main mode for transmission of HCV in Europe. Estimates about the number of people living with HCV in Belgium are rare and even less is known about the prevalence of HCV among PWID. METHOD: Between 1 February 2019 and 26 April 2019, PWID and high-risk opiate users (HROU) were recruited in Brussels through respondent-driven sampling (RDS). They were invited to a questionnaire and underwent a rapid HCV test. RESULTS: A total of 253 respondents participated in the study, of which 168 were PWID and 238 were HROU, with 153 respondents belonging to both categories. The overall unweighted sample average for HCV antibodies was 41.1%. The weighted population estimates were 43.7% (95% CI 30.6-56.8%) for RDS-II and 43.4% (95% CI 28.9-58.0%) for RDS-SS. CONCLUSIONS: This prevalence is lower than the prevalence estimates reported elsewhere in Europe. However, the data still suggest that serious efforts are needed to reach the goal set by the WHO to reduce HCV by 2030 with 90%.

The practice of female genital mutilation across the world: Data availability and approaches to measurement.

While Female Genital Mutilation (FGM) has been in existence for centuries, the rigorous and systematic documentation of the extent of the practice is a recent undertaking. This paper discusses data availability related to the practice of FGM and reviews the methods used to generate prevalence estimates. The aim is to illustrate strengths and limitations of the available data. The review is organised around two main categories of countries: FGM countries of origin, where representative prevalence data exist, and countries of migration for women and girls who have undergone FGM, for which representative prevalence data are lacking. This second category also includes countries across the world where FGM is only found among small autochthonous populations.

Utilization of gabapentin by people in treatment for substance use disorders in Belgium (2011-2014): a cross-sectional study.

BACKGROUND: Although gabapentin has been licensed in the European Union only for neuropathic pain and epilepsy for patients who have partial seizures, it has also been prescribed in treatment for substance use disorders. Many studies report the potential risk of abuse of gabapentin by people with substance use disorders. The objective of this paper is to determine if people who have been in treatment for substance use disorders bought gabapentin in a time span that could indicate consumption at a dose that exceeded the maximum approved dose of 3600 mg/day. METHODS: This analysis is the result of an observational cross-sectional descriptive study with matching. Two datasets were used and linked at individual level. Subjects were selected based on their first registration in the database of the Treatment Demand Indicator (TDI) between 2011 and 2014, without any exclusion criteria concerning nationality or age. Through linkage with the database of the InterMutualistic Agency (IMA) information on health service use and medication use was determined. In addition, each subject was matched on age, sex and place of residence to four comparators from the general population who were not in specialized treatment. The prevalence of gabapentin purchases in the period between 2008 and 2014 for both populations were compared. Quantification of the amount of gabapentin between two consecutive purchases was used as a proxy for potential abuse. RESULTS: Out of 30,905 patients in treatment for substance use disorders 2.7% had bought at least once gabapentin in a public pharmacy or received it from a hospital pharmacy, compared to 0.7% in the comparison group (n = 122,142). In both populations, more than half of the patients bought only once or twice gabapentin and about 10.0% bought at least once gabapentin in a time span that could indicate potential abuse. A limitation of the study is that it is only based on reimbursed medication without clinical information. CONCLUSION: Through the linkage of the TDI-database and the database of the Belgian health insurance companies, no evidence was found for regular abuse of prescribed gabapentin in Belgium by people in treatment for substance use disorders.

Longitudinal pharmacoepidemiological and health services research for substance users in treatment: protocol of the Belgian TDI-IMA linkage.

BACKGROUND: Not much is known about the health seeking behavior of people with substance use disorders before they enter specialized treatment and afterwards. This paper explains in detail the protocol that has been followed to establish the Belgian TDI-IMA-database, which is linking two separate databases: the Treatment Demand Indicator (TDI) and the database of the Intermutualistic Agency (IMA). The Treatment Demand Indicator is measuring incidence of people with substance use disorders entering drug treatment. The IMA-database covers data, collected in the framework of the compulsory Belgian health care and benefits insurance program, on reimbursed medication and the use of reimbursed health services. The linkage results in pharmacoepidemiological and health service data for people who were in treatment for substance use disorders and for a group of comparators. METHODS: The TDI-database was linked to the IMA-database for the period between 01/01/2008 and 31/12/2017, based on the national identification number of patients who have been in alcohol or drug treatment between 01/01/2011 and 31/12/2014. Through this linkage, pharmacoepidemiological and health service data became available for at least 3 years before the first registered episode in the TDI-database till at least 3 years after the first episode. For each person in TDI four comparators, who were not in specialized treatment, were matched on age, sex and place of residence. DISCUSSION: The TDI-IMA-database allows for an analysis of health seeking behavior and health care pathways of people before and after they entered specialized alcohol and drug treatment. The presented protocol could be used in other European countries to establish a linkage between existing health databases. This will allow for a better understanding of the health care needs of patients with substance use disorders.

Estimates of first-generation women and girls with female genital mutilation in the European Union, Norway and Switzerland.

BACKGROUND: Female genital mutilation (FGM) is the practice of partial or total removal of female genitalia for non-medical reasons. The procedure has no known health benefits but can cause serious immediate and long-term obstetric, gynaecological and sexual health problems. Health workers in Europe are often unaware of the consequences of FGM and lack the knowledge to treat women adequately. OBJECTIVE: Our goal was to estimate the number of first-generation girls and women in the European Union, Norway and Switzerland who have undergone FGM. Before migration from FGM-practicing countries began, FGM was an unknown phenomenon in Europe. METHODS: Secondary analysis of data from the 2011 EU census and extrapolation from age-specific FGM prevalence rates in the immigrants' home countries to these data were used to provide our estimates. Estimates based on census and other demographic data were compared to our results for Belgium. RESULTS: In 2011 over half a million first-generation women and girls in the EU, Norway and Switzerland had undergone FGM before immigration. One in two was living in the UK or France, one in two was born in East-Africa. CONCLUSIONS: For the first time, scientific evidence gives a reliable estimate of the number of first-generation women and girls in Europe coming from countries where FGM is practiced. The use of census data proves reliable for policy makers to guide their actions, e.g., regarding training needs for health workers who might be confronted with women who have undergone FGM, or the need for reconstructive surgery.