BackgroundEdentulism (loss of all teeth) is a final marker of disease burden for oral health common among older adults and poorer populations. Yet most evidence is from high-income countries. Oral health has many of the same social and behavioural risk factors as other non-communicable diseases (NCDs) which are increasing rapidly in low- and middle-income countries with ageing populations. The “common risk factor approach” (CRFA) for oral health addresses risk factors shared with NCDs within the broader social and economic environment.MethodsThe aim is to improve understanding of edentulism prevalence, and association between common risk factors and edentulism in adults aged 50 years and above using nationally representative samples from China (N = 11,692), Ghana (N = 4093), India (N = 6409) and South Africa (N = 2985). The data source is the World Health Organization (WHO) Study on global AGEing and adult health (SAGE) Wave 1 (2007–2010). Multivariable logistic regression describes association between edentulism and common risk factors reported in the literature.ResultsPrevalence of edentulism: in China 8.9 %, Ghana 2.9 %, India 15.3 %, and South Africa 8.7 %. Multivariable analysis: in China, rural residents were more likely to be edentulous (OR 1.36; 95 % CI 1.09–1.69) but less likely to be edentulous in Ghana (OR 0.53; 95 % CI 0.31–0.91) and South Africa (OR 0.52; 95 % CI 0.30–0.90). Respondents with university education (OR 0.31; 95 % CI 0.18–0.53) and in the highest wealth quintile (OR 0.68; 95 % CI 0.52–0.90) in China were less likely to be edentulous. In South Africa respondents with secondary education were more likely to be edentulous (OR 2.82; 95 % CI 1.52–5.21) as were those in the highest wealth quintile (OR 2.78; 95 % CI 1.16–6.70). Edentulism was associated with former smokers in China (OR 1.57; 95 % CI 1.10–2.25) non-drinkers in India (OR 1.65; 95 % CI 1.11–2.46), angina in Ghana (OR 2.86; 95 % CI 1.19–6.84) and hypertension in South Africa (OR 2.75; 95 % CI 1.72–4.38). Edentulism was less likely in respondents with adequate nutrition in China (OR 0.68; 95 % CI 0.53–0.87). Adjusting for all other factors, compared with China, respondents in India were 50 % more likely to be edentulous.ConclusionsStrengthening the CRFA should include addressing common determinants of health to reduce health inequalities and improve both oral and overall health.

BackgroundThe 2015 Global Burden of Disease Study estimated that oral conditions affect 3.5 billion people worldwide with a higher burden among older adults and those who are socially and economically disadvantaged. Studies of inequalities in the use of oral health services by those in need have been conducted in high-income countries but evidence from low- and middle-income countries (LMICs) is limited. This study measures and describes socioeconomic inequality in self-reported unmet need for oral health services in adults aged 50 years and over, in China, Ghana and India.MethodsA cross-sectional analysis of national survey data from the WHO SAGE Wave 1 (2007–2010) was conducted. Study samples in China (n = 1591), Ghana (n = 425) and India (n = 1307) were conditioned on self-reported need for oral health services in the previous 12 months. The binary dependent variable, unmet need for oral health services, was derived from questions about self-reported need and service use. Prevalence was estimated by country. Unmet need was measured and compared in terms of relative levels of education and household wealth. The methods were logistic regression and the relative index of inequality (RII). Models were adjusted for age, sex, area of residence, marital status, work status and self-rated health.ResultsThe prevalence of unmet need was 60, 80, and 62% in China, Ghana and India respectively. The adjusted RII for education was statistically significant for China (1.5, 95% CI:1.2–1.9), Ghana (1.4, 95% CI: 1.1–1.7), and India (1.5, 95% CI:1.2–2.0), whereas the adjusted RII for wealth was significant only in Ghana (1.3, 95% CI:1.1–1.6). Male sex was significantly associated with self-reported unmet need for oral health services in India.ConclusionsGiven rapid population ageing, further evidence of socioeconomic inequalities in unmet need for oral health services by older adults in LMICs is needed to inform policies to mitigate inequalities in the availability of oral health services. Oral health is a universal public health issue requiring attention and action on multiple levels and across the public private divide.

BackgroundMost studies in the United States (US) have used income and education as socioeconomic indicators but there is limited information on other indicators, such as wealth. We aimed to assess how two socioeconomic status measures, income and wealth, compare as correlates of socioeconomic disparity in dentist visits among adults in the US.MethodsData from the National Health and Nutrition Examination Survey (NHANES) 2011–2014 were used to calculate self-reported dental visit prevalence for adults aged 20 years and over living in the US. Prevalence ratios using Poisson regressions were conducted separately with income and wealth as independent variables. The dependent variable was not having a dentist visit in the past 12 months. Covariates included sociodemographic factors and untreated dental caries. Parsimonious models, including only statistically significant (p < 0.05) covariates, were derived. The Akaike Information Criterion (AIC) measured the relative statistical quality of the income and wealth models. Analyses were additionally stratified by race/ethnicity in response to statistically significant interactions.ResultsThe prevalence of not having a dentist visit in the past 12 months among adults aged 20 years and over was 39%. Prevalence was highest in the poorest (58%) and lowest wealth (57%) groups. In the parsimonious models, adults in the poorest and lowest wealth groups were close to twice as likely to not have a dentist visit (RR 1.69; 95%CI: 1.51–1.90) and (RR 1.68; 95%CI: 1.52–1.85) respectively. In the income model the risk of not having a dentist visit were 16% higher in the age group 20–44 years compared with the 65+ year age group (RR 1.16; 95%CI: 1.04–1.30) but age was not statistically significant in the wealth model. The AIC scores were lower (better) for the income model. After stratifying by race/ethnicity, age remained a significant indicator for dentist visits for non-Hispanic whites, blacks, and Asians whereas age was not associated with dentist visits in the wealth model.ConclusionsIncome and wealth are both indicators of socioeconomic disparities in dentist visits in the US, but both do not have the same impact in some populations in the US.Electronic supplementary materialThe online version of this article (10.1186/s12903-018-0613-4) contains supplementary material, which is available to authorized users.

Introduction Population‐based biomarker surveys are the gold standard for estimating HIV prevalence but are susceptible to substantial non‐participation (up to 30%). Analytical missing data methods, including inverse‐probability weighting (IPW) and multiple imputation (MI), are biased when data are missing‐not‐at‐random, for example when people living with HIV more frequently decline participation. Heckman‐type selection models can, under certain assumptions, recover unbiased prevalence estimates in such scenarios. Methods We pooled data from 142,706 participants aged 15–49 years from nationally representative cross‐sectional Population‐based HIV Impact Assessments in seven countries in sub‐Saharan Africa, conducted between 2015 and 2018 in Tanzania, Uganda, Malawi, Zambia, Zimbabwe, Lesotho and Eswatini. We compared sex‐stratified HIV prevalence estimates from unadjusted, IPW, MI and selection models, controlling for household and individual‐level predictors of non‐participation, and assessed the sensitivity of selection models to the copula function specifying the correlation between study participation and HIV status. Results In total, 84.1% of participants provided a blood sample to determine HIV serostatus (range: 76% in Malawi to 95% in Uganda). HIV prevalence estimates from selection models diverged from IPW and MI models by up to 5% in Lesotho, without substantial precision loss. In Tanzania, the IPW model yielded lower HIV prevalence estimates among males than the best‐fitting copula selection model (3.8% vs. 7.9%). Conclusions We demonstrate how HIV prevalence estimates from selection models can differ from those obtained under missing‐at‐random assumptions. Further benefits include exploration of plausible relationships between participation and outcome. While selection models require additional assumptions and careful specification, they are an important tool for triangulating prevalence estimates in surveys with substantial missing data due to non‐participation.