Effective questionnaire-based prediction models for type 2 diabetes across several ethnicities: a model development and validation study.

Background: Type 2 diabetes disproportionately affects individuals of non-White ethnicity through a complex interaction of multiple factors. Therefore, early disease detection and prediction are essential and require tools that can be deployed on a large scale. We aimed to tackle this problem by developing questionnaire-based prediction models for type 2 diabetes prevalence and incidence for multiple ethnicities. Methods: In this proof of principle analysis, logistic regression models to predict type 2 diabetes prevalence and incidence, using questionnaire-only variables reflecting health state and lifestyle, were trained on the White population of the UK Biobank (n = 472,696 total, aged 37-73 years, data collected 2006-2010) and validated in five other ethnicities (n = 29,811 total) and externally in Lifelines (n = 168,205 total, aged 0-93 years, collected between 2006 and 2013). In total, 631,748 individuals were included for prevalence prediction and 67,083 individuals for the eight-year incidence prediction. Type 2 diabetes prevalence in the UK Biobank ranged between 6% in the White population to 23.3% in the South Asian population, while in Lifelines, the prevalence was 1.9%. Predictive accuracy was evaluated using the area under the receiver operating characteristic curve (AUC), and a detailed sensitivity analysis was conducted to assess potential clinical utility. We compared the questionnaire-only models to models containing physical measurements and biomarkers as well as to clinical non-laboratory type 2 diabetes risk tools and conducted a reclassification analysis. Findings: Our algorithms accurately predicted type 2 diabetes prevalence (AUC = 0.901) and eight-year incidence (AUC = 0.873) in the White UK Biobank population. Both models replicated well in the Lifelines external validation, with AUCs of 0.917 and 0.817 for prevalence and incidence, respectively. Both models performed consistently well across different ethnicities, with AUCs of 0.855-0.894 for prevalence and 0.819-0.883 for incidence. These models generally outperformed two clinically validated non-laboratory tools and correctly reclassified >3,000 additional cases. Model performance improved with the addition of blood biomarkers but not with the addition of physical measurements. Interpretation: Our findings suggest that easy-to-implement, questionnaire-based models could be used to predict prevalent and incident type 2 diabetes with high accuracy across several ethnicities, providing a highly scalable solution for population-wide risk stratification. Future work should determine the effectiveness of these models in identifying undiagnosed type 2 diabetes, validated in cohorts of different populations and ethnic representation. Funding: University Medical Center Groningen.

Use of Traditional Botanical Medicines During Pregnancy in Rural Rwanda.

OBJECTIVE: To evaluate the perceptions of healthcare and traditional medicine providers regarding the type, indications, side effects, and prevalence of traditional medicine use amongst pregnant women in a rural Rwandan population. METHODS: Six focus groups with physicians, nurses, and community health workers and four individual in-depth interviews with traditional medicine providers were held. Qualitative data was gathered using a structured questionnaire querying perceptions of the type, indications, side effects, and prevalence of use of traditional medicines in pregnancy. RESULTS: The healthcare provider groups perceived a high prevalence of traditional botanical medicine use by pregnant women (50-80%). All three groups reported similar indications for use of the medicines and the socioeconomic status of the pregnant women who use them. The traditional medicine providers and the healthcare providers both perceived that the most commonly used medicine is a mixture of many plants, called Inkuri. The most serious side effect reported was abnormally bright green meconium with a poor neonatal respiratory drive. Thirty-five traditional medicines were identified that are used during pregnancy. CONCLUSION: Perceptions of high prevalence of use of traditional medicines during pregnancy with possible negative perinatal outcomes exist in areas of rural Rwanda.

Food insecurity with hunger is associated with obesity among HIV-infected and at risk women in Bronx, NY.

BACKGROUND: Food insecurity, insufficient quality and quantity of nutritionally adequate food, affects millions of people in the United States (US) yearly, with over 18 million Americans reporting hunger. Food insecurity is associated with obesity in the general population. Due to the increasing prevalence of obesity and risk factors for cardiovascular disease among HIV-infected women, we sought to determine the relationship between food insecurity and obesity in this cohort of urban, HIV-infected and -uninfected but at risk women. METHODS: Using a cross-sectional design, we collected data on food insecurity, body mass index and demographic and clinical data from 231 HIV-infected and 119 HIV-negative women enrolled in Bronx site of the Women's Interagency HIV Study (WIHS). We used multivariate logistic regression to identify factors associated with obesity. RESULTS: Food insecurity was highly prevalent, with almost one third of women (110/350, 31%) reporting food insecurity over the previous six months and over 13% of women reported food insecurity with hunger. Over half the women were obese with a Body Mass Index (BMI) of ≥ 30. In multivariate analyses, women who were food insecure with hunger had higher odds of obesity (Adjusted odds ratio [aOR] =  2.56, 95% Confidence Interval [CI]  =  1.27, 5.20) after adjusting for HIV status, age, race, household status, income, drug and alcohol use. CONCLUSION: Food insecurity with hunger was associated with obesity in this population of HIV-infected and -uninfected, urban women. Both food insecurity and obesity are independent markers for increased mortality; further research is needed to understand this relationship and their role in adverse health outcomes.

Adverse pregnancy outcomes and sexual violence among female sex workers who inject drugs on the United States-Mexico border.

This study examines the prevalence of miscarriage/stillbirth among female sex workers who inject drugs (FSW-IDUs) and measures its associations with physical and sexual violence. Baseline data from 582 FSW-IDUs enrolled in an HIV intervention study in Tijuana and Ciudad Juárez, Mexico were used for current analyses. 30% of participants had experienced at least one miscarriage/stillbirth, 51% had experienced sexual violence, and 49% had experienced physical violence. History of miscarriage/stillbirth was associated with sexual violence (adjusted odds ratio [aOR] = 1.7, p = .02) but not physical violence. Additional reproductive risks associated with miscarriage/stillbirth included high numbers of male clients in the previous month (aOR = 1.1 per 30 clients, p = 0.04), history of abortion (aOR = 3.7, p < .001), and higher number of pregnancies (aOR = 1.4 per additional pregnancy, p < .001). Programs and research with this population should integrate reproductive health and consider gender-based violence.

Structural determinants of food insufficiency, low dietary diversity and BMI: a cross-sectional study of HIV-infected and HIV-negative Rwandan women.

OBJECTIVES: In Sub-Saharan Africa, the overlapping epidemics of undernutrition and HIV infection affect over 200 and 23 million people, respectively, and little is known about the combined prevalence and nutritional effects. The authors sought to determine which structural factors are associated with food insufficiency, low dietary diversity and low body mass index (BMI) in HIV-negative and HIV-infected Sub-Saharan women. STUDY DESIGN: Cross-sectional analysis of a longitudinal cohort. SETTING: Community-based women's organisations. PARTICIPANTS: 161 HIV-negative and 514 HIV-infected Rwandan women. PRIMARY AND SECONDARY OUTCOME MEASURES: Primary outcomes included food insufficiency (reporting 'usually not' or 'never' to 'Do you have enough food?'), low household dietary diversity (Household Dietary Diversity Score ≤3) and BMI <18.5 (kg/m(2)). The authors also measured structural and behavioural factors including: income, household size, literacy and alcohol use. RESULTS: Food insufficiency was prevalent (46%) as was low dietary diversity (43%) and low BMI (15%). Food insufficiency and dietary diversity were associated with low income (adjusted odds ratio (aOR)=2.14 (95% CI 1.30 to 3.52) p<0.01), (aOR=6.51 (95% CI 3.66 to 11.57) p<0.001), respectfully and illiteracy (aOR=2.00 (95% CI 1.31 to 3.04) p<0.01), (aOR=2.10 (95% CI 1.37 to 3.23) p<0.001), respectfully and were not associated with HIV infection. Alcohol use was strongly associated with food insufficiency (aOR=3.23 (95% CI 1.99 to 5.24) p<0.001). Low BMI was inversely associated with HIV infection (aOR≈0.5) and was not correlated with food insufficiency or dietary diversity. CONCLUSIONS: Rwandan women experienced high rates of food insufficiency and low dietary diversity. Extreme poverty, illiteracy and alcohol use, not HIV infection alone, may contribute to food insufficiency in Rwandan women. Food insufficiency, dietary diversity and low BMI do not correlate with one another; therefore, low BMI may not be an adequate screening tool for food insufficiency. Further studies are needed to understand the health effects of not having enough food, low food diversity and low weight in both HIV-negative and HIV-infected women.

Condom access: Associations with consistent condom use among female sex workers in two northern border cities of Mexico.

To determine whether condom access is associated with consistent condom use among female sex workers (FSWs) in Tijuana and Ciudad Juarez, between 2004 and 2006 we administered a questionnaire to 924 FSWs who reported unprotected sex with a client in the past 2 months. Of these women, 43% reported consistent ("often" or "always") condom use, 74% said condoms were available, and 38% reported having access to free condoms. In a logistic regression, factors positively associated with consistent condom use were condom availability (adjusted odds ratio [AOR] = 2.00; 95% confidence interval [CI]: 1.32-3.03), condom affordability (AOR = 1.72; 95% CI: 1.25-2.38) and self-efficacy (AOR = 2.16; 95% CI: 1.54-3.04). Factors inversely associated with consistent condom use included poor financial status (AOR = 0.65; 95% CI: 0.47-0.90), methamphetamine use (AOR = 0.58; 95% CI: 0.40-0.83), alcohol use (AOR = 0.68; 95% CI: 0.49-0.96), and recent injection drug use (AOR = 0.62; 95% CI: 0.39-0.97). While increased condom availability may improve condom use among FSWs in general, interventions to broaden condom use among lower income and drug-using FSWs are critically needed.

Effects of government registration on unprotected sex amongst female sex workers in Tijuana; Mexico.

BACKGROUND: Sex work is partially regulated in Tijuana, but little is known of its health effects. A recent behavioural intervention amongst female sex workers (FSWs) decreased incidence of HIV/STIs by 40%. We evaluated effects of sex worker regulation on condom use amongst FSWs randomized to this intervention. METHODS: FSWs aged ≥18 years who reported unprotected sex with ≥1 client in the last 2 months and whether they were registered with Tijuana's Municipal Health Department underwent a brief, theory-based behavioural intervention to increase condom use. At baseline and 6 months, women underwent interviews and testing for HIV, syphilis, Chlamydia trachomatis and Neisseria gonorrhoeae. Negative binomial regression was used to determine the effect of registration on numbers of unprotected sex acts and cumulative HIV/STI incidence. RESULTS: Of 187 women, 83 (44%) were registered. Lack of registration was associated with higher rates of unprotected sex (rate ratio: 1.7, 95% CI: 1.2-2.3), compared to FSWs who were registered, after controlling for potential confounders. CONCLUSIONS: Registration predicted increased condom use amongst FSWs enrolled in a behavioural intervention. Public health programmes designed to improve condom use amongst FSWs may benefit from understanding the impact of existing regulation systems on HIV risk behaviours.

A comparison of registered and unregistered female sex workers in Tijuana, Mexico.

OBJECTIVE: Sex work is regulated in Tijuana, Mexico, but only half of the city's female sex workers (FSWs) are registered with the municipal health department, which requires regular screening for sexually transmitted infections (STIs) and human immunodeficiency virus (HIV). We examined correlates of registration to determine if it confers measurable health benefits. METHODS: From 2004 to 2006, we interviewed FSWs in Tijuana > or = 18 years of age who reported recent unprotected sex with at least one client and were not knowingly HIV-positive, and tested them for HIV, syphilis, gonorrhea, and chlamydia. Logistic regression identified factors associated with registration. RESULTS: Of 410 FSWs, 44% were registered, 69% had been tested for HIV, 6% were HIV-positive, and 44% tested positive for any STI. Compared with unregistered FSWs, registered FSWs were more likely to have had HIV testing (86% vs. 56%, p < 0.001) and less likely to test positive for any STI (33% vs. 53%, p < 0.001) or HIV (3% vs. 8%, p = 0.039). Factors independently associated with registration included ever having an HIV test (adjusted odds ratio [AOR] = 4.19) and earning > $30 per transaction without a condom (AOR = 2.41), whereas working on the street (AOR = 0.34), injecting cocaine (AOR = 0.06), snorting or smoking methamphetamine (AOR = 0.27), and being born in the Mexican state of Baja California (AOR = 0.35) were inversely associated with registration. CONCLUSION: Registered FSWs were more likely than unregistered FSWs to have had HIV testing and to engage in less drug use, but did not have significantly lower HIV or STI prevalence after adjusting for confounders. Current regulation of FSWs in Tijuana should be further examined to enhance the potential public health benefits of registration.

Shamba Maisha: a pilot study assessing impacts of a micro-irrigation intervention on the health and economic wellbeing of HIV patients.

BACKGROUND: HIV/AIDS negatively impacts poverty alleviation and food security, which reciprocally hinder the rapid scale up and effectiveness of HIV care programs. Nyanza province has the highest HIV prevalence (15.3%), and is the third highest contributor (2.4 million people) to rural poverty in Kenya. Thus, we tested the feasibility of providing a micro-irrigation pump to HIV-positive farmers in order to evaluate its impact on health and economic advancement among HIV-positive patients and their families. METHODS: Thirty HIV-positive patients enrolled in the Family AIDS Care and Education Services (FACES) program in Kisumu, Kenya were provided a micro-financed loan to receive an irrigation pump and farming guidance from KickStart, the developer of the pump. Economic data, CD4 counts, household health and loan repayment history were collected 12 months after the pumps were distributed. RESULTS: Mean annual family income increased by $1,332 over baseline. CD4 counts did not change significantly. Though income increased, only three (10%) participants had paid off more than a quarter of the loan. CONCLUSIONS: We demonstrated the feasibility of an income-generating micro-irrigation intervention among HIV-positive patients and the collection of health and economic data. While family income improved significantly, loan repayment rates were low- likely complicated by the drought that occurred in Kenya during the intervention period.