Ranking Age-at-Death Distributions Using Dominance: Robust Evaluation of United States Mortality Trends, 2006-2021.

Diverging mortality trends at different ages motivate the monitoring of lifespan inequality alongside life expectancy. Conclusions are ambiguous when life expectancy and lifespan inequality move in the same direction or when inequality measures display inconsistent trends. We propose using nonparametric dominance analysis to obtain a robust ranking of age-at-death distributions. Application to U.S. period life tables for 2006-2021 reveals that, until 2014, more recent years generally dominate earlier years, implying improvement if longer lifespans that are less unequally distributed are considered better. Improvements were more pronounced for non-Hispanic Black and Hispanic individuals than for non-Hispanic White individuals. Since 2014, for all subpopulations-particularly Hispanics-earlier years often dominate more recent years, indicating worsening age-at-death distributions if shorter and more unequal lifespans are considered worse. Dramatic deterioration of the distributions in 2020-2021 during the COVID-19 pandemic is most evident for Hispanic individuals.

Optimizing cardiovascular disease risk screening in a low-resource setting: cost-effectiveness of program modifications in Sri Lanka modelled with nationally representative survey data.

BACKGROUND: While screening for cardiovascular disease (CVD) risk can help low-resource health systems deliver low-cost, effective prevention, evidence is needed to adapt international screening guidelines for maximal impact in local settings. We aimed to establish how the cost-effectiveness of CVD risk screening in Sri Lanka varies with who is screened, how risk is assessed, and what thresholds are used for prescription of medicines. METHODS: We used data for people aged 35 years and over from a 2018/19 nationally representative survey in Sri Lanka. We modelled the costs and quality adjusted life years (QALYs) for 128 screening program scenarios distinguished by a) age group screened, b) risk tool used, c) definition of high CVD risk, d) blood pressure threshold for treatment of high-risks, and e) prescription of statins to all diabetics. We used the current program as the base case. We used a Markov model of a one-year screening program with a lifetime horizon and a public health system perspective. RESULTS: Scenarios that included the WHO-2019 office-based risk tool dominated most others. Switching to this tool and raising the age threshold for screening from 35 to 40 years gave an incremental cost-effectiveness ratio (ICER) of $113/QALY. Lowering the CVD high-risk threshold from 20 to 10% and prescribing antihypertensives at a lower threshold to diabetics and people at high risk of CVD gave an ICER of $1,159/QALY. The findings were sensitive to allowing for disutility of daily medication. CONCLUSIONS: In Sri Lanka, CVD risk screening scenarios that used the WHO-2019 office-based risk tool, screened people above the age of 40, and lowered risk and blood pressure thresholds would likely be cost-effective, generating an additional QALY at less than half a GDP per capita.

Risk factor contributions to socioeconomic inequality in cardiovascular risk in the Philippines: a cross-sectional study of nationally representative survey data.

BACKGROUND: Primary prevention of cardiovascular diseases (CVD) increasingly relies on monitoring global CVD risk scores. Lack of evidence on socioeconomic inequality in these scores and the contributions that specific risk factors make to this inequality impedes effective targeting of CVD prevention. We aimed to address this evidence gap by measuring and decomposing socioeconomic inequality in CVD risk in the Philippines. METHODS: We used data on 8462 individuals aged 40-74 years from the Philippines National Nutrition Survey and the laboratory-based Globorisk equation to predict 10-year risk of a CVD event from sex, age, systolic blood pressure, total cholesterol, high blood glucose, and smoking. We used a household wealth index to proxy socioeconomic status and measured socioeconomic inequality with a concentration index that we decomposed into contributions of the risk factors used to predict CVD risk. We measured socioeconomic inequalities in these risk factors and decomposed them into contributions of more distal risk factors: body mass index, fat share of energy intake, low physical activity, and drinking alcohol. We stratified by sex. RESULTS: Wealthier individuals, particularly males, had greater exposure to all risk factors, with the exception of smoking, and had higher CVD risks. Total cholesterol and high blood glucose accounted for 58% and 34%, respectively, of the socioeconomic inequality in CVD risk among males. For females, the respective estimates were 63% and 69%. Systolic blood pressure accounted for 26% of the higher CVD risk of wealthier males but did not contribute to inequality among females. If smoking prevalence had not been higher among poorer individuals, then the inequality in CVD risk would have been 35% higher for males and 75% higher for females. Among distal risk factors, body mass index and fat intake contributed most to inequalities in total cholesterol, high blood sugar, and, for males, systolic blood pressure. CONCLUSIONS: Wealthier Filipinos have higher predicted CVD risks and greater exposure to all risk factors, except smoking. There is need for a nuanced approach to CVD prevention that targets anti-smoking programmes on the poorer population while targeting diet and exercise interventions on the wealthier.

Sociodemographic and geographic inequalities in diagnosis and treatment of older adults' chronic conditions in India: a nationally representative population-based study.

CONTEXT: Expeditious diagnosis and treatment of chronic conditions are critical to control the burden of non-communicable disease in low- and middle-income countries. We aimed to estimate sociodemographic and geographic inequalities in diagnosis and treatment of chronic conditions among adults aged 45 + in India. METHODS: We used 2017-18 nationally representative data to estimate prevalence of chronic conditions (hypertension, diabetes, lung disease, heart disease, stroke, arthritis, cholesterol, and neurological) reported as diagnosed and percentages of diagnosed conditions that were untreated by sociodemographic characteristics and state. We used concentration indices to measure socioeconomic inequalities in diagnosis and lack of treatment. Fully adjusted inequalities were estimated with multivariable probit and fractional regression models. FINDINGS: About 46.1% (95% CI: 44.9 to 47.3) of adults aged 45 + reported a diagnosis of at least one chronic condition and 27.5% (95% CI: 26.2 to 28.7) of the reported conditions were untreated. The percentage untreated was highest for neurological conditions (53.2%; 95% CI: 50.1 to 59.6) and lowest for diabetes (10.1%; 95% CI: 8.4 to 11.5). Age- and sex-adjusted prevalence of any diagnosed condition was highest in the richest quartile (55.3%; 95% CI: 53.3 to 57.3) and lowest in the poorest (37.7%: 95% CI: 36.1 to 39.3). Conditional on reported diagnosis, the percentage of conditions untreated was highest in the poorest quartile (34.4%: 95% CI: 32.3 to 36.5) and lowest in the richest (21.1%: 95% CI: 19.2 to 23.1). Concentration indices confirmed these patterns. Multivariable models showed that the percentage of untreated conditions was 6.0 points higher (95% CI: 3.3 to 8.6) in the poorest quartile than in the richest. Between state variations in the prevalence of diagnosed conditions and their treatment were large. CONCLUSIONS: Ensuring more equitable treatment of chronic conditions in India requires improved access for poorer, less educated, and rural older people who often remain untreated even once diagnosed.

Missed opportunities for hypertension screening: a cross-sectional study, India.

OBJECTIVE: To assess missed opportunities for hypertension screening at health facilities in India and describe systematic differences in these missed opportunities across states and sociodemographic groups. METHODS: We used nationally representative survey data from the 2017-2018 Longitudinal Ageing Study in India to estimate the proportion of adults aged 45 years or older identified with hypertension and who had not been diagnosed with hypertension despite having visited a health facility during the previous 12 months. We estimated age-sex adjusted proportions of missed opportunities to diagnose hypertension, as well as actual and potential proportions of diagnosis, by sociodemographic characteristics and for each state. FINDINGS: Among those identified as having hypertension, 22.6% (95% confidence interval, CI: 21.3 to 23.8) had not been diagnosed despite having recently visited a health facility. If these opportunities had been realized, the prevalence of diagnosed hypertension would have increased from 54.8% (95% CI: 53.5 to 56.1) to 77.3% (95% CI: 76.2 to 78.5). Missed opportunities for diagnosis were more common among individuals who were poorer (P = 0.001), less educated (P < 0.001), male (P < 0.001), rural (P < 0.001), Hindu (P = 0.001), living alone (P = 0.028) and working (P < 0.001). Missed opportunities for diagnosis were more common at private than at public health facilities (P < 0.001) and varied widely across states (P < 0.001). CONCLUSION: Opportunistic screening for hypertension has the potential to significantly increase detection of the condition and reduce sociodemographic and geographic inequalities in its diagnosis. Such screening could be a first step towards more effective and equitable hypertension treatment and control.

Awareness, treatment, and control of hypertension in adults aged 45 years and over and their spouses in India: A nationally representative cross-sectional study.

BACKGROUND: Lack of nationwide evidence on awareness, treatment, and control (ATC) of hypertension among older adults in India impeded targeted management of this condition. We aimed to estimate rates of hypertension ATC in the older population and to assess differences in these rates across sociodemographic groups and states in India. METHODS AND FINDINGS: We used a nationally representative survey of individuals aged 45 years and over and their spouses in all Indian states (except one) in 2017 to 2018. We identified hypertension by blood pressure (BP) measurement ≥140/90 mm Hg or self-reported diagnosis if also taking medication or observing salt/diet restriction to control BP. We distinguished those who (i) reported diagnosis ("aware"); (ii) reported taking medication or being under salt/diet restriction to control BP ("treated"); and (iii) had measured systolic BP <140 and diastolic BP <90 ("controlled"). We estimated age-sex adjusted hypertension prevalence and rates of ATC by consumption quintile, education, age, sex, urban-rural, caste, religion, marital status, living arrangement, employment status, health insurance, and state. We used concentration indices to measure socioeconomic inequalities and multivariable logistic regression to estimate fully adjusted differences in these outcomes. Study limitations included reliance on BP measurement on a single occasion, missing measurements of BP for some participants, and lack of data on nonadherence to medication. The 64,427 participants in the analysis sample had a median age of 57 years: 58% were female, and 70% were rural dwellers. We estimated hypertension prevalence to be 41.9% (95% CI 41.0 to 42.9). Among those with hypertension, we estimated that 54.4% (95% CI 53.1 to 55.7), 50.8% (95% CI 49.5 to 52.0), and 28.8% (95% CI 27.4 to 30.1) were aware, treated, and controlled, respectively. Across states, adjusted rates of ATC ranged from 27.5% (95% CI 22.2 to 32.8) to 75.9% (95% CI 70.8 to 81.1), from 23.8% (95% CI 17.6 to 30.1) to 74.9% (95% CI 69.8 to 79.9), and from 4.6% (95% CI 1.1 to 8.1) to 41.9% (95% CI 36.8 to 46.9), respectively. Age-sex adjusted rates were lower (p < 0.001) in poorer, less educated, and socially disadvantaged groups, as well as for males, rural residents, and the employed. Among individuals with hypertension, the richest fifth were 8.5 percentage points (pp) (95% CI 5.3 to 11.7; p < 0.001), 8.9 pp (95% CI 5.7 to 12.0; p < 0.001), and 7.1 pp (95% CI 4.2 to 10.1; p < 0.001) more likely to be aware, treated, and controlled, respectively, than the poorest fifth. CONCLUSIONS: Hypertension prevalence was high, and ATC of the condition were low among older adults in India. Inequalities in these indicators pointed to opportunities to target hypertension management more effectively and equitably on socially disadvantaged groups.

Effect of screening mammography on breast cancer mortality: Quasi-experimental evidence from rollout of the Dutch population-based program with 17-year follow-up of a cohort.

There is uncertainty about the magnitude of the effect of screening mammography on breast cancer mortality. The relevance and validity of evidence from dated randomized controlled trials has been questioned, whereas observational studies often lack a valid comparison group. There is no estimate of the effect of one screening invitation only. We exploited the geographic rollout of the Dutch screening mammography program across municipalities to estimate the effects of one additional biennial screening invitation on breast cancer and all-cause mortality. Population administrative data provided vital status and cause of death of a cohort of women aged 49-63 in 1995 over 17 years. Linear probability models were used to estimate the mortality effects. We estimated 154 fewer breast cancer deaths (95% confidence interval: 40-267; p = 0.01) over 17 years in a population of 100,000 women aged 49-63 who received one additional biennial screening invitation, which corresponds to an 9.6% risk reduction for a woman of age 56. The estimated effect on all-cause mortality was negative but not close to statistical significance. Our study shows that one single invitation for breast cancer screening is effective in reducing breast cancer mortality, which is important for health policy. The effect is smaller than previous estimates of the effect of invitation for multiple screens, which further emphasizes the importance of achieving regular participation.

Progressive universalism? The impact of targeted coverage on health care access and expenditures in Peru.

Like other countries seeking a progressive path to universalism, Peru has attempted to reduce inequalities in access to health care by granting the poor entitlement to tax-financed basic care without charge. We identify the impact of this policy by comparing the target population's change in health care utilization with that of poor adults already covered through employment-based insurance. There are positive effects on receipt of ambulatory care and medication that are largest among the elderly and the poorest. The probability of getting formal health care when sick is increased by almost two fifths, but the likelihood of being unable to afford treatment is reduced by more than a quarter. Consistent with the shallow coverage offered, there is no impact on use of inpatient care. Neither is there any effect on average out-of-pocket health care expenditure, but medical spending is reduced by up to 25% in the top quarter of the distribution. Copyright © 2017 John Wiley & Sons, Ltd.

Impact of Performance-Based Financing in a Low-Resource Setting: A Decade of Experience in Cambodia.

This paper exploits the geographic expansion of performance-based financing (PBF) in Cambodia over a decade to estimate its effect on the utilization of maternal and child health services. PBF is estimated to raise the proportion of births occurring in incentivized public health facilities by 7.5 percentage points (25%). A substantial part of this effect arises from switching the location of institutional births from private to public facilities; there is no significant impact on deliveries supervised by a skilled birth attendant, nor is there any significant effect on neonatal mortality, antenatal care and vaccination rates. The impact on births in public facilities is much greater if PBF is accompanied by maternity vouchers that cover user fees, but there is no significant effect among the poorest women. Heterogeneous effects across schemes differing in design suggest that maintaining management authority within a health district while giving explicit service targets to facilities is more effective in raising utilization than contracting management to a non-governmental organization while denying it full autonomy and leaving financial penalties vague. Copyright © 2015 John Wiley & Sons, Ltd.

conindex: Estimation of concentration indices.

Concentration indices are frequently used to measure inequality in one variable over the distribution of another. Most commonly, they are applied to the measurement of socioeconomic-related inequality in health. We introduce a user-written Stata command conindex which provides point estimates and standard errors of a range of concentration indices. The command also graphs concentration curves (and Lorenz curves) and performs statistical inference for the comparison of inequality between groups. The article offers an accessible introduction to the various concentration indices that have been proposed to suit different measurement scales and ethical responses to inequality. The command's capabilities and syntax are demonstrated through analysis of wealth-related inequality in health and healthcare in Cambodia.

Can vouchers deliver? An evaluation of subsidies for maternal health care in Cambodia.

OBJECTIVE: To evaluate the effect of vouchers for maternity care in public health-care facilities on the utilization of maternal health-care services in Cambodia. METHODS: The study involved data from the 2010 Cambodian Demographic and Health Survey, which covered births between 2005 and 2010. The effect of voucher schemes, first implemented in 2007, on the utilization of maternal health-care services was quantified using a difference-in-differences method that compared changes in utilization in districts with voucher schemes with changes in districts without them. FINDINGS: Overall, voucher schemes were associated with an increase of 10.1 percentage points (pp) in the probability of delivery in a public health-care facility; among women from the poorest 40% of households, the increase was 15.6 pp. Vouchers were responsible for about one fifth of the increase observed in institutional deliveries in districts with schemes. Universal voucher schemes had a larger effect on the probability of delivery in a public facility than schemes targeting the poorest women. Both types of schemes increased the probability of receiving postnatal care, but the increase was significant only for non-poor women. Universal, but not targeted, voucher schemes significantly increased the probability of receiving antenatal care. CONCLUSION: Voucher schemes increased deliveries in health centres and, to a lesser extent, improved antenatal and postnatal care. However, schemes that targeted poorer women did not appear to be efficient since these women were more likely than less poor women to be encouraged to give birth in a public health-care facility, even with universal voucher schemes.

Health and health system effects on poverty: A narrative review of global evidence.

Ill-health causes poverty. The effect runs through multiple mechanisms that span lifetimes and cross generations. Health systems can reduce poverty by improving health and weakening links from ill-health to poverty. This paper maps routes through which ill-health can cause poverty and identifies those that are potentially amenable to health policy. The review confirms that ill-health is an important contributor to poverty and it finds that the effect through health-related loss of earnings is often larger than that through medical expenses. Both effects are smaller in countries that are closer to universal health coverage and have higher social safety nets. The paper also reviews evidence from low- and middle-income countries (LMICs) and the United States (US) on the poverty-reduction effectiveness of public health insurance (PubHI) for low-income households. This reveals that PubHI does not always deliver financial protection to its targeted population in LMICs. Countries that have succeeded in achieving this goal often combine extension of coverage with supply-side interventions to build capacity and avoid perverse provider incentives in response to insurance. In the US, PubHI is effective in reducing poverty by shielding low-income households with children from healthcare costs and, consequently, generating long-run improvements in health that increase lifetime earnings. Poverty reduction is a potentially important co-benefit of health systems.

Distributionally sensitive measurement and valuation of population health.

We introduce a measure of population health that is sensitive to inequality in both age-specific health and lifespan and can be calculated from a health-extended period life table. By allowing for inequality aversion, the measure generalises health-adjusted life expectancy without requiring more data. A transformation of change in the (life-years) measure gives a distributionally sensitive monetary valuation of change in population health and disease burden. Application to Sub-Saharan Africa between 1990 and 2019 reveals that the change in population health is sensitive to allowing for lifespan inequality but is less sensitive to age-specific health inequality. Allowing for distributional sensitivity changes relative burdens of diseases, reduces convergence between the burdens of communicable and non-communicable diseases, and so could influence disease prioritisation. It increases the value of health improvements relative to GDP.

Aversion to health inequality - Pure, income-related and income-caused.

We design a novel experiment to identify aversion to pure (univariate) health inequality separately from aversion to income-related and income-caused health inequality. Participants allocate resources to determine health of individuals. Identification comes from random variation in resource productivity and information on income and its causal effect. We gather data (26,286 observations) from a sample of UK adults (n = 337) and estimate pooled and participant-specific social preferences while accounting for noise. The median person has strong aversion to pure health inequality, challenging the health maximisation objective of economic evaluation. Aversion to health inequality is even stronger when it is related to income. However, the median person prioritises health of poorer individuals less than is assumed in the standard measure of income-related health inequality. On average, aversion to that inequality does not become stronger when low income is known to cause ill-health. There is substantial heterogeneity in all three types of inequality aversion.

Health outcomes and healthcare utilization associated with four undiagnosed chronic conditions: evidence from nationally representative survey data in Sri Lanka.

Background: Low awareness of chronic conditions raises the risk of poorer health outcomes and may result in healthcare utilization and spending in response to symptoms of undiagnosed conditions. Little evidence exists, particularly from lower-middle-income countries, on the health and healthcare use of undiagnosed people with an indication of a condition. This study aimed to compare health (physical, mental, and health-related quality of life (HRQoL)) and healthcare (inpatient and outpatient visits and out-of-pocket (OOP) medical spending) outcomes of undiagnosed Sri Lankans with an indication of coronary heart disease (CHD), hypertension, diabetes, and depression with the outcomes of their compatriots who were diagnosed or had no indication of these conditions. Methods: This study used a nationally representative survey of Sri Lankan adults to identify people with an indication of CHD, hypertension, diabetes, or depression, and ascertain if they were diagnosed. Outcomes were self-reported measures of physical and mental functioning (12-Item Short Form Survey (SF-12)), HRQoL (EQ-5D-5L), inpatient and outpatient visits, and OOP spending. For each condition, we estimated the mean of each outcome for respondents with (a) no indication, (b) an indication without diagnosis, and (c) a diagnosis. We adjusted the group differences in these means for socio-demographic covariates using ordinary least squares (OLS) regression for physical and mental function, Tobit regression for HRQoL, and a generalized linear model (GLM) for healthcare visits and OOP spending. Results: An indication of each of CHD and depression, which are typically symptomatic, was associated with a lower adjusted mean of physical (CHD -2.65, 95% CI -3.66, -1.63; depression -5.78, 95% CI -6.91, -4.64) and mental functioning (CHD -2.25, 95% CI -3.38, -1.12; depression -6.70, 95% CI -7.97, -5.43) and, for CHD, more annual outpatient visits (2.13, 95% CI 0.81, 3.44) compared with no indication of the respective condition. There were no such differences for indications of hypertension and diabetes, which are often asymptomatic. Conclusions: Living with undiagnosed CHD and depression was associated with worse health and, for CHD, greater utilization of healthcare. Diagnosis and management of these symptomatic conditions can potentially improve health partly through substitution of effective healthcare for that which primarily responds to symptoms. Supplementary Information: The online version contains supplementary material available at 10.1186/s44263-024-00075-0.

The Prevalence and Epidemiological Features of Ischaemic Heart Disease in Sri Lanka.

Background: There is limited evidence on the prevalence of ischaemic heart disease (IHD) and its association with risk factors and socioeconomic status (SES) in low- and middle-income countries (LMICs). Given the relatively high levels of access to healthcare in Sri Lanka, the association of IHD with SES may be different from that observed in other LMICs. Objectives: To estimate the prevalence of IHD in Sri Lanka, determine its associated risk factors and its association with SES. Methods: We analysed data from 6,513 adults aged ≥18 years examined in the 2018/19 Sri Lanka Health and Ageing Study. We used the Rose angina questionnaire to classify participants as having angina (Angina+) and used self-report or medical records to identify participants with a history of IHD (History+). The association of Angina+ and History+ with age, ethnicity, sector of residence, education level, household SES wealth quintile, area SES wealth quintile, hypertension, diabetes, smoking, total cholesterol, cholesterol-to-HDL ratio, waist-to-hip ratio and body mass index were analysed in unadjusted and adjusted models. Additional analyses were performed to investigate sensitivity to correction for missing data and to benchmark estimates against evidence from other studies. Conclusions: We estimated prevalence of History+ of 3.9% (95% CI 3.3%-4.4%) and Angina+ of 3.0% (95% CI 2.4%-3.5%) in adults aged 18 years and over. The prevalence of Angina+ was higher in women than men (3.9% vs. 1.9%, p < 0.001) whilst prevalence of History+ was lower (3.8% vs. 4.0%, p = 0.8), which may suggest a higher rate of undiagnosed IHD in women. A history of IHD was strongly associated with age, hypertension and diabetes status even after adjusting for sociodemographic factors. Though the prevalence of History+ was higher in the most developed area SES tertile and urban areas, History+ was also associated with less education but not household SES, consistent with patterns emerging from other LMICs.

Prevalence of pain and its treatment among older adults in India: a nationally representative population-based study.

ABSTRACT: There were no estimates of the prevalence of pain and its treatment in the older population of India obtained from face-to-face interviews with a nationally representative sample. We addressed this evidence gap by using data on 63,931 individuals aged 45 years and older from the 2017/2018 Longitudinal Ageing Study in India. We identified pain from an affirmative response to the question: Are you often troubled by pain? We also identified those who reported pain that limited usual activities and who received treatment for pain. We estimated age- and sex-adjusted prevalence of pain, pain limiting usual activity and treatment, and compared these estimates across states and sociodemographic groups. We used a multivariable probit model to estimate full adjusted differences in the probability of each outcome across states and sociodemographic groups. We estimated that 36.6% (95% confidence interval [CI]: 35.3-37.8) of older adults in India were often troubled by pain and 25.2% (95% CI: 24.2-26.1) experienced pain limiting usual activity. We estimated that 73.3% (95% CI: 71.9-74.6) of those often troubled by pain and 76.4% (95% CI: 74.9-78.0) of those with pain that limited usual activity received treatment. There was large variation in each outcome across states. Fully adjusted prevalence of pain and pain limiting usual activity were higher among individuals who were female, older, less educated, rural residents, and poorer. Prevalence of treatment among those troubled by pain was lower among socially disadvantaged groups.

Public health insurance coverage in India before and after PM-JAY: repeated cross-sectional analysis of nationally representative survey data.

INTRODUCTION: The provision of non-contributory public health insurance (NPHI) to marginalised populations is a critical step along the path to universal health coverage. We aimed to assess the extent to which Ayushman Bharat-Pradhan Mantri Jan Arogya Yojana (PM-JAY)-potentially, the world's largest NPHI programme-has succeeded in raising health insurance coverage of the poorest two-fifths of the population of India. METHODS: We used nationally representative data from the National Family Health Survey on 633 699 and 601 509 households in 2015-2016 (pre-PM-JAY) and 2019-2021 (mostly, post PM-JAY), respectively. We stratified by urban/rural and estimated NPHI coverage nationally, and by state, district and socioeconomic categories. We decomposed coverage variance between states, districts, and households and measured socioeconomic inequality in coverage. For Uttar Pradesh, we tested whether coverage increased most in districts where PM-JAY had been implemented before the second survey and whether coverage increased most for targeted poorer households in these districts. RESULTS: We estimated that NPHI coverage increased by 11.7 percentage points (pp) (95% CI 11.0% to 12.4%) and 8.0 pp (95% CI 7.3% to 8.7%) in rural and urban India, respectively. In rural areas, coverage increased most for targeted households and pro-rich inequality decreased. Geographical inequalities in coverage narrowed. Coverage did not increase more in states that implemented PM-JAY. In Uttar Pradesh, the coverage increase was larger by 3.4 pp (95% CI 0.9% to 6.0%) and 4.2 pp (95% CI 1.2% to 7.1%) in rural and urban areas, respectively, in districts exposed to PM-JAY and the increase was 3.5 pp (95% CI 0.9% to 6.1%) larger for targeted households in these districts. CONCLUSION: The introduction of PM-JAY coincided with increased public health insurance coverage and decreased inequality in coverage. But the gains cannot all be plausibly attributed to PM-JAY, and they are insufficient to reach the goal of universal coverage of the poor.

Cost-Effectiveness and Distributional Impact of Opportunistic Screening for People at High-Risk of Cardiovascular Disease in Sri Lanka: A Modelling Study.

Background: While hypertension, diabetes, hypercholesterolemia and high-risk of cardiovascular disease can be easily diagnosed and treated with cost-effective medicines, a large proportion of people remain undiagnosed. We assessed the potential effectiveness, cost, and distributional impact of opportunistically screening for these chronic conditions at outpatient patient departments in Sri Lanka. Methods: We used nationally representative data on biomarkers and healthcare utilization in 2019 to model the screening of people aged 40+ without preexisting CVD and without a reported diagnosis of hypertension, diabetes, or hypercholesterolemia. We modelled an intensive one month program that would screen a proportion of those making an outpatient visit to a public or private clinic and follow-up a proportion of those screened to confirm diagnoses. We also modelled a less intensive one year program. The main outcomes were the new diagnoses of any of the chronic conditions. Program costs were calculated and the socioeconomic distributions of individuals screened, new cases diagnosed, and treatments delivered were estimated. Sensitivity analyses varied the probability of screening and follow-up. Results: Using data on 2,380 survey participants who met the inclusion criteria, we estimated that the one month program would diagnose 8.2% (95% CI: 6.8, 9.6) of those with a chronic condition who would remain undiagnosed without the program. The one year program would diagnose 26.9% (95% CI: 26.5, 27.4) of the otherwise undiagnosed and would have a cost per person newly diagnosed of USD 6.82 (95% CI: 6.61, 7.03) in the public sector and USD 16.92 (95% CI: 16.37, 17.47) in the private sector. New diagnoses would be evenly distributed over the socioeconomic distribution, with public (private) clinics diagnosing a higher proportion of poorer (richer) individuals. Both programs would reduce underdiagnosis among males relative to females. Conclusions: Opportunistic screening for cardiovascular diseases at outpatient clinics in Sri Lanka could be cost-effective and equitable.