Correspondence to Dr Santosh Kumar Sharma; [email protected]
STRENGTHS AND LIMITATIONS OF THIS STUDY
The study provides empirical evidence on comparison of educational inequalities in the self-reported and measured prevalence of high blood pressure (BP) and high blood glucose (BG) in Kerala, India using a large nationally representative sample of 15+.
The findings were based on a cross-sectional survey; thus, no causality was tested.
The broad pattern of inequality in Kerala suggests that for both high BP and BG, whether self-reported or clinically measured, those with less education are at a disadvantage.
Findings suggest that research and programme efforts need to be redoubled to determine what is driving greater vulnerability to non-communicable diseases among those with lower educational attainment on the one hand, and the possible role that improving education access can be on health outcomes, on the other hand.
Introduction
Non-communicable diseases (NCDs) now account for a larger proportion of lower-middle-income countries (LMICs) overall illness burden.1–3 NCDs account for 41 million annual deaths or 71% of all fatalities worldwide.4 More than 15 million individuals aged 30–69 die from an NCD every year; 85% of these ‘premature’ deaths take place in LMICs.4 In India, the burden of NCDs has increased over the past 20 years,5 and in 2019 was responsible for around 65% of all deaths in the country.6 The share of disability-adjusted life-years (DALYs) in India that are believed to be caused by NCDs has increased from 31% of all DALYs in 1990 to 55% in 2016.7 It is believed that this transformation is being mostly driven by the rising prevalence of diabetes and hypertension.2 The 2015 Sustainable Development Goals (SDGs) recognised NCDs as a serious hazard to achieving sustainable development and set a global goal to reduce premature death by one-third over the next 15 years.8
Kerala, a state in southern India, has for some time outperformed other states regarding measures, including life expectancy, mortality and death rates, to place first in the country’s Sustainable Development Goal India Index in 2019.9 However, when compared with other Indian states, the state’s morbidity levels are significantly higher, primarily due to the burden of NCDs.10–13 Earlier research has found the existence of socioeconomic inequalities in NCD testing,14 as well as NCD prevalence,15 in patterns distinct from other parts of the country.16
Education has long been regarded as the most significant socioeconomic predictor of hypertension and diabetes mellitus.17–22 A growing number of studies have shown that the association between educational attainment and health behaviours is significant in NCDs, given the critical role of health behaviours, including self-management and health-related lifestyle.1 19 20 23–26 Higher education has been associated with 1.3% lower diabetes risk, 2.6% heart disease risk and 1.8% lower 5-year mortality rates.27 While Kerala boasts high educational attainment, some concerns have been raised over inequalities in access, particularly in higher education.28 This state is unique in this regard, and the inequality challenge in Kerala deserves further exploration. Given a low mortality and high morbidity scenario in the state, such inequalities can compound vulnerabilities, making health a cause of further economic deprivation.29
While growing or persisting increases in educational disparity in hypertension, diabetes and other cardiovascular risk factors have been thoroughly reported in Western populations over the last several decades,1 20 22 30–34 relevant studies in Asian settings were harder to come by. Several studies in India also only focused on wealth-related disparities in NCDs14 35 36; none we found explored educational inequalities. For the first time, moreover, India’s Demographic and Health Survey (DHS), also called the National Family Health Survey (NFHS), was furnishing information on self-reports and also carrying out blood glucose (BG) and blood pressure (BP) measurements.13 Taking advantage of the richness of NFHS data, our study attempted to assess educational inequalities in measured hypertension, self-reported high BP, measured high BG, self-reported high BG, and BP and BG testing among participants aged 15 years and above in Kerala, India.
Measures
Data
The study used the fifth round of India’s DHS, called the NFHS-5, conducted between 2019 and 2021. NFHS-5 was conducted by the International Institute for Population Sciences, Mumbai, as the national nodal agency, with technical support from ICF (global advisory and technology services provider for many Demographic and Health Surveys) Macro International, USA, under the aegis of the Ministry of Health and Family Welfare, Government of India. COVID-19 temporarily halted the survey,13 although the data from the state of Kerala that we draw on is from the first phase, which preceded the halt. The NFHS-5 sample was designed to provide national, state/union territory and district-level estimates of various indicators that are critical to monitor the SDGs on population, health, nutrition and gender equality, among others. NFHS-5 adopted a two-stage stratified sampling design to select the nationally representative sample from the country. The 2011 Census of India was used to generate the sampling frame for selecting primary sampling units (PSUs). The samples were stratified into urban and rural areas. The census enumeration blocks were PSUs in the urban areas, and villages were PSUs in the rural areas. Additional details about the sampling design and instruments are available.13
A unique feature of the NFHS-5 is the inclusion of testing of the adult population aged 15 years and above for BP and random BG level within the biomarker components. The study used member recoded file of NFHS-5, which covered biomarker information from 2 049 401 individuals (1 013 679 male, 1 035 722 female) aged 15 and above. We focused on these indicators pertaining to self-reported testing from this dataset with an emphasis on the state of Kerala, which restricted our sample to 36 526 individuals, of whom 32 250 of 36 526 completed BP measurement and 31 917 of 36 526 completed BG measurement within the biomarker components.
Patient and public involvement
Patient or public participation in the planning, execution, analysis or publication of our research was not suitable or practical.
Measures
Dependent variables
In the biomarker schedule, self-reported information was collected from both men and women against a series of questions such as ‘Were you told on two or more different occasions by a doctor or other health professional that you had hypertension or high BP or high BG level’, ‘Before this survey, has your BP or BG ever been checked?’ and ‘To lower your BP and BG, are you now taking a prescribed medicine’. BP was measured for eligible men and women aged 15+, using an Omron Blood Pressure Monitor (ref) to determine the prevalence of high BP. BP measurements for each respondent were taken three times with an interval of 5 min between readings. The average of the last two readings was considered after excluding the first reading to avoid white coat hypertension. Respondents whose average systolic BP was ≥140 mm Hg or whose average diastolic BP was ≥90 mm Hg were considered to have elevated BP readings or he/she is currently taking antihypertensive medication to lower his/her BP.37 38
Random BG was measured using a finger-stick blood specimen for all women and men aged 15 and above using the Accu-Chek Performa glucometer with glucose test strips for BG testing. An individual is classified as having high BG if he/she has a random BG level of >140 mg/dL. The BG thresholds are used to ascertain high BG as per the NFHS-5 India report.13
Independent variables
Sociodemographic characteristics such as age (15–44, 45–59, 60–74, 75+), sex (male, female), education (no education, primary, secondary, higher), marital status (never married, currently married, widowed/divorced/separated/deserted), place of residence (urban, rural), wealth quintile (poorest, poorer, middle, richer, richest), religion (Hindu, Muslim, others (comprising Christian, Sikh, Buddhist, Jain and others)) were considered. Risk factors and risk protection indicators were also considered, including consumption of alcohol (no, yes), current smoking or chewing tobacco (no, yes) and coverage of health insurance for any household member (no, yes).
Education level was used as the core dimension of inequality that we were interested in—and was classified into four categories: no education, primary school or less (up to 5 years of schooling), secondary school (6–12 years) and higher (>12 years). Education level is a measure that reflects family social status, mediated by education policies and is a determinant of employment, income and cognitive ability for self-care; also, it is a component of mechanisms associated with health and healthcare inequalities.17 34 39 Kerala happens to be a high-literacy state, and in large part, educational inequalities have been on the decline in the state for many decades now.40 41 Therefore, the remaining inequality is of particular interest.
Statistical analysis
Descriptive analyses included frequencies of the sociodemographic characteristics and of the outcome of interest. Prevalence of outcome was adjusted for age, sex, marital status, place of residence, religion, caste, wealth quintile, alcohol consumption, smoking/chewing tobacco and health insurance using marginal modelling.17 33 42–44
ORs and 95% CIs for the occurrence of the dependent variables were estimated by using logistic regression models for those with primary, secondary and higher education as compared with those reporting no education after adjusting for age, sex, marital status, place of residence, religion, caste, wealth quintile, alcohol consumption, smoking/chewing tobacco and health insurance.43 A variance inflation factor (VIF) was used to assess multicollinearity between the selected predictors before executing the final model, and the value of VIF is 2.9 in both surveys, which is less than 10. We used classification analysis to check the fitness of the regression model, and our AUC (Area under curve) score is 0.7, which suggests that the fitted model was acceptable.
There is no agreement on the best measure for expressing the magnitude of inequalities. The fundamental difference in how the magnitude of inequalities is expressed is whether the measure is absolute or relative. Absolute inequality is derived from the difference between the prevalence/incidence of health outcomes between groups, that is, by subtracting the extreme values. A relative measure of inequality is based on a ratio between the estimates of two extremes of the stratification variables. Despite their simplicity, these measures, which consider only the top and bottom segments of the population under study, but have significant drawbacks. First, these measures are sensitive to changes in the number of individuals in each stratification category. Another limitation is that in some cases, the lowest and highest stratification groups will not have the lowest and highest prevalence, particularly when the overall prevalence is high. More importantly, these basic inequality measures do not account for intermediate population groups. More advanced measures can overcome this limitation by using information from the whole population. Harper and Lynch have recommended the use of the absolute concentration index or the Slope Index of Inequality (SII) as indicators of absolute inequality and the use of the Relative Concentration Index (RCI) as indicators of relative inequality, both of which look at all population subgroups.45–47 These measures of inequality are complementary and have been computed for WHO48 as well as other eminent inequality analyses, for instance, by Barros and Victora.45
In our study, the SII was used to estimate the magnitude and direction of educational inequalities in the occurrence of measured high BP and high BG levels, and self-reported high BP and high BG levels. It is an absolute inequality measure used specifically for ordinal variables of stratification (usually socioeconomic indicators such as income groups, wealth and literacy rates). It represents the absolute difference, in predicted values, of a health indicator between the highest-educated individuals and the least-educated individuals, taking into consideration the entire distribution of the stratification variable using the adequate regression model.45 47 48 The SII was calculated as the difference, in percentage points, between the estimated values for the extreme groups of a given stratification variable. Although the SII was conceived based on a linear regression model, in general, the logistic regression model is more adequate for its calculation because it is usually applied to coverage of indicators and prevalence of health outcomes, avoiding linear predictions out of the boundaries of an expected interval of a proportion (from 0 to 100).45 Regarding the proportions, the absolute differences between the group and the SII vary from −100 to 100 percentage points. Values close to 0 are expected when there is no inequality. Positive values reveal that the health indicator, be it the coverage of an intervention or the prevalence of a health risk, is greater in the most privileged group.
The RCI is the measure of relative inequality, which, like the SII, considers all categories of the stratification variable. The difference with a relative summary measure is that it is unitless and, therefore, can perhaps more easily be used to compare indicators. The value of RCI corresponds to twice times the area between a diagonal line that represents perfect equality among the groups and the curve that expresses the coverage observed for each cumulative percentage of the population studied. When coverage is greater among the highest educated individuals, the area generated is under the diagonal line, and when coverage is higher among the less educated individuals, the opposite is true.49 The RCI is similar to the Gini coefficient—it varies from −1 to+1, assumes 0 as equality, and the further the values are from 0, the highest the relative inequality is.45 46 50 In our study, the results of SII and RCI were multiplied by 100 to facilitate their visualisation in tables and graphs, ranging from −100 to +100.
All the statistical analyses were conducted by Stata V.17 MP version (StataCorp), using the appropriate sampling weight variables in the dataset due to the complex survey design of NFHS surveys. The complex survey design effects were adjusted by using Stata svyset and svy commands.
Results
Table 1 presents the sociodemographic characteristics of the respondents and the prevalence of measured high BP and high BG, and self-reported high BP and high BG, and screening of BP and BG in sociodemographic strata. The sample had a greater proportion of females (53.2%) compared with males (46.8%), 68.2% married and more than half of the study population was aged 45 or older. We found that 4.6% had no education, 60.7% had attained a secondary level of education and 21.1% had attained higher education. Around 56.9% of respondents belonged to the Hindu religion, and 51.2% belonged to the other backward caste group. About three-fifths (60.1%) of respondents were covered by the health schemes in Kerala.
Table 1Sociodemographic characteristics and prevalence of measured high BP, self-reported high BP, BP tested, measured high BG level, self-reported high BG and BG tested among participants in Kerala, NFHS-5 (2019–2021)
Background | N | % | Measured high BP* | Self-reported high BP | BP tested | Measured random BG† | Self-reported high BG | BG tested |
Age | ||||||||
15–44 | 18 024 | 49.4 | 10.7 | 3.9 | 76.5 | 8.9 | 3.4 | 64.3 |
45–59 | 9875 | 27.0 | 39.8 | 22.6 | 94.5 | 29.5 | 18.1 | 89.7 |
60–74 | 6659 | 18.2 | 63.0 | 46.9 | 96.7 | 42.4 | 33.8 | 93.3 |
75+ | 1967 | 5.4 | 72.9 | 57.2 | 97.6 | 42.3 | 33.9 | 93.1 |
Sex | ||||||||
Male | 17 110 | 46.8 | 32.9 | 18.1 | 81.2 | 23.6 | 15.0 | 73.1 |
Female | 19 415 | 53.2 | 31.5 | 21.8 | 90.6 | 21.7 | 14.5 | 82.2 |
Education | ||||||||
No education | 1667 | 4.6 | 57.7 | 40.8 | 93.6 | 32.5 | 22.7 | 89.0 |
Primary | 4946 | 13.6 | 55.7 | 39.2 | 94.7 | 36.3 | 26.2 | 88.6 |
Secondary | 22 140 | 60.7 | 29.5 | 17.6 | 85.1 | 21.9 | 13.7 | 77.0 |
Higher | 7706 | 21.1 | 17.6 | 9.9 | 83.2 | 12.9 | 8.1 | 72.3 |
Marital status | ||||||||
Never married | 7561 | 20.7 | 8.5 | 3.0 | 56.1 | 5.6 | 2.1 | 38.8 |
Married | 24 898 | 68.2 | 34.1 | 20.7 | 93.4 | 25.3 | 16.2 | 87.4 |
W/Di/Se/De‡ | 4067 | 11.1 | 60.1 | 45.7 | 96.3 | 35.6 | 27.6 | 91.3 |
Place of residence | ||||||||
Urban | 17 421 | 47.7 | 32.1 | 20.7 | 86.9 | 22.7 | 14.7 | 79.3 |
Rural | 19 105 | 52.3 | 32.1 | 19.7 | 86.1 | 22.4 | 14.7 | 77.2 |
Wealth quintile | ||||||||
Poorest | 296 | 0.8 | 39.0 | 24.5 | 81.7 | 22.7 | 11.1 | 71.6 |
Poorer | 1794 | 4.9 | 35.8 | 20.8 | 81.5 | 22.3 | 12.2 | 70.6 |
Middle | 6649 | 18.2 | 31.8 | 18.7 | 84.6 | 21.6 | 13.0 | 75.3 |
Richer | 13 159 | 36.0 | 31.2 | 19.3 | 86.4 | 21.6 | 13.6 | 78.5 |
Richest | 14 629 | 40.1 | 32.5 | 21.4 | 88.1 | 23.9 | 16.9 | 80.4 |
Religion | ||||||||
Hindu | 20 802 | 57.0 | 33.1 | 20.2 | 86.2 | 22.8 | 13.9 | 77.8 |
Muslim | 9037 | 24.7 | 27.5 | 17.9 | 86.9 | 21.0 | 13.5 | 78.9 |
Others§ | 6686 | 18.3 | 35.2 | 23.1 | 86.9 | 24.0 | 18.6 | 78.4 |
Caste | ||||||||
SC/ST¶ | 4690 | 12.8 | 29.7 | 17.9 | 83.2 | 19.6 | 11.8 | 73.2 |
OBC | 18 687 | 51.2 | 30.4 | 18.7 | 86.7 | 22.0 | 13.9 | 78.8 |
Others | 13 148 | 36.0 | 35.4 | 23.0 | 87.3 | 24.4 | 17.0 | 79.1 |
Drink alcohol | ||||||||
No | 33 061 | 90.5 | 31.3 | 20.4 | 86.6 | 22.1 | 14.5 | 78.2 |
Yes | 3465 | 9.5 | 39.7 | 18.0 | 85.1 | 26.6 | 16.5 | 78.5 |
Smoking/tobacco | ||||||||
No | 33 187 | 90.9 | 31.2 | 20.1 | 86.5 | 22.0 | 14.5 | 78.1 |
Yes | 3339 | 9.1 | 41.3 | 20.6 | 86.2 | 28.0 | 16.7 | 79.3 |
Health insurance | ||||||||
No | 14 578 | 39.9 | 32.3 | 20.4 | 86.5 | 22.8 | 15.6 | 78.3 |
Yes | 21 948 | 60.1 | 32.0 | 20.0 | 86.4 | 22.4 | 14.1 | 78.2 |
Total | 36 526 | 100 | 32.1 | 20.1 | 86.5 | 22.5 | 14.7 | 78.2 |
*An individual is classified as having high BP if they have SBP≥140 mm Hg or DBP≥90 mm Hg at the time of the survey or they are currently taking medicine to lower their BP.
†An individual is classified as having high BG if he/she has a random blood glucose level of more than 140 mg/dL.
‡Widowed/divorced/separated/desserted.
§Christian, Buddha, Jain, Sikh, etc.
¶Scheduled caste/scheduled tribe.
BG, blood glucose; BP, blood pressure; DBP, diastolic blood pressure; NFHS, National Family Health Survey; OBC, other backward caste; SBP, systolic blood pressure.
More females reported having tested for BP and BG compared with males in Kerala. Overall, the measured prevalence of high BP and high BG levels was 32.1% and 22.5%, and the self-reported prevalence of high BP and high BG levels was 20.1% and 14.7% in Kerala. About 86.5% and 78.2% of respondents ever tested for high BP and high BG levels. This prevalence increased with age for all six outcomes. The measured prevalence of high BP and high BG level was slightly higher among males as compared with females, whereas the self-reported prevalence of high BP was higher among females (21.8%) as compared with males (18.1%).
An educational gradient was observed for all six outcomes, with a higher prevalence among the least educated group. The largest relative prevalence discrepancy in Kerala, between the least and the most educated groups, was observed for measured high BP (57.7% and 17.6%), resulting in a difference of 40 percentage points. Detailed data can be found in online supplemental figure S1. For self-reported high BP, measured high BG and self-reported high BG, these prevalence differences were: 30.9%, 19.6% and 14.6%, respectively.
The prevalence of all six outcomes was higher among ever-married respondents in Kerala. It was observed that the prevalence of measured high BP, high BG and self-reported high BG was higher for respondents who consumed alcohol and smoked/chewed tobacco. There was no significant difference in the prevalence of measured high BP, self-reported high BP, self-reported high BG, measured high BG, BP and BG testing between those who have health insurance and those who do not.
Table 2 presents the educational inequalities in all outcomes after adjusting for age, sex, marital status, place of residence, religion, caste, wealth quintile, alcohol consumption, smoking/chewing tobacco and health insurance in Kerala, as it is likely that educational attainment among older generations could be lower than in younger generations, given the growing focus on increasing literacy in the state over the past many decades. After adjusting for covariates, the prevalence of all outcomes was approximately the same as unadjusted (figure 1, online supplemental figure S1). We found a significant association between education and odds of measured high BP and self-reported high BP (both unadjusted and adjusted).
Table 2Educational inequalities in measured high BP, self-reported high BP, measured high BG level and self-reported high BG among participants in Kerala, NFHS-5 2019–2021
Unadjusted prevalence % (95% CI) | % Adjusted prevalence (95% CI) * | Unadjusted OR (95% CI) | Adjusted OR (95% CI) † | |
Self-reported high BP | ||||
No education | 40.8 (37.8,43.9) | 40.1 (39.1, 41.1) | 6.01 (5.13, 7.04) | 1.29 (1.06, 1.55) |
Primary | 39.2 (37.3,41.0) | 38.9 (38.3, 39.6) | 5.72 (5.00, 6.54) | 1.38 (1.19, 1.60) |
Secondary | 17.6 (16.9,18.3) | 17.1 (16.8, 17.5) | 1.91 (1.71, 2.14) | 1.17 (1.03, 1.31) |
Higher | 9.9 (9.0,11.0) | 9.6 (9.1, 10.1) | 1.0 | 1.0 |
Measured high BP‡ | ||||
No education | 57.7 (54.8, 60.6) | 56.9 (55.9, 58.0) | 6.40 (5.55, 7.39) | 1.56 (1.31, 1.86) |
Primary | 55.7 (53.8, 57.5) | 55.5 (54.8, 56.1) | 5.90 (5.23, 6.64) | 1.55 (1.36, 1.77) |
Secondary | 29.5 (28.6, 30.4) | 29.1 (28.6, 29.5) | 1.97 (1.78, 2.17) | 1.26 (1.14, 1.39) |
Higher | 17.6 (16.3, 18.9) | 17.3 (16.7, 18.0) | 1.0 | 1.0 |
Self-reported high BG | ||||
No education | 22.7 (20.2, 25.5) | 22.4 (21.9, 23.0) | 3.36 (2.76, 4.08) | 1.02 (0.82, 1.26) |
Primary | 26.2 (24.6, 27.8) | 26.2 (25.8, 26.6) | 4.05 (3.49, 4.71) | 1.22 (1.03, 1.45) |
Secondary | 13.7 (13.1, 14.3) | 13.6 (13.3, 13.9) | 1.81 (1.58, 2.08) | 1.17 (1.02, 1.34) |
Higher | 8.1 (7.1, 9.1) | 8.0 (7.6, 8.4) | 1.0 | 1.0 |
Measured high BG§ | ||||
No education | 32.5 (29.8, 35.3) | 32.1 (31.6, 32.7) | 3.24 (2.78, 3.78) | 1.16 (0.98, 1.39) |
Primary | 36.2 (34.7, 37.9) | 36.2 (35.8, 36.6) | 3.83 (3.41, 4.30) | 1.35 (1.19, 1.54) |
Secondary | 21.9 (21.2, 22.6) | 21.8 (21.5, 22.1) | 1.89 (1.70, 2.11) | 1.29 (1.16, 1.44) |
Higher | 12.9 (11.8, 14.1) | 12.9 (12.5, 13.3) | 1.0 | 1.0 |
*Prevalence of health outcomes was adjusted for age, sex, marital status, place of residence, religion, caste, wealth quintile, alcohol consumption, smoking/chewing tobacco and health insurance by means of marginal modelling.
†Adjusted for age, sex, marital status, place of residence, religion, caste, wealth quintile, alcohol consumption, smoking/chewing tobacco and health insurance.
‡An individual is classified as having high BP if they have SBP≥140 mm Hg or DBP≥90 mm Hg at the time of the survey, or they are currently taking medicine to lower their blood pressure.
§An individual is classified as having high blood glucose if he/she has a random blood glucose level of more than 140 mg/dL.
BG, blood glucose; BP, blood pressure; DBP, diastolic blood pressure; NFHS, National Family Health Survey; SBP, systolic blood pressure.
Figure 1. Adjusted prevalence of self-reported and measured high BP and high BG by educational categories, NFHS-5 2019-2021. Note: Prevalence of health outcomes was adjusted for age, sex, marital staus, place of residence, religion, caste, wealth quintile, alcohol consumption, smoking/chewing tobacco and health insurance by means of marginal modelling. BG, blood glucose; BP, blood pressure; NFHS, National Family Health Survey.
Respondents with no education had sixfold the odds of having measured high BP and self-reported high BP compared with respondents with higher education levels. The adjusted odds of measured high BP were 56% greater among those with no education as compared with those with higher education (no education vs higher education, adjusted OR (AOR) 1.56 (95% CI 1.31 to 1.86)), while the adjusted odds of self-reported high BP were 29% greater across the same comparison groups (no education vs higher education, AOR 1.29 (95% CI 1.06 to 1.55)). This pattern was not seen with BG: there was no statistically significant association between respondents with no education and adjusted odds of measured high BG (AOR 1.16; 95% CI 0.98 to 1.39) and self-reported high BG occurrence (AOR 1.02; 95% CI 0.82 to 1.26) respectively.
Table 3, online supplemental figures S2 and S3 show the absolute (SII) and relative (RCI) measures of educational inequality for all six outcomes. The negative magnitude of SII and RCI for all self-reported high BP, measured high BP, self-reported high BG and measured high BG suggests a higher concentration of morbidity among those with the lowest educational attainment. Strikingly, absolute and relative educational inequality were higher in measured high BP (SII −45.4% (95% CI –47.3% to –43.4%); RCI −26.6% (95% CI –27.9% to –25.3%)) as compared with self-reported high BP (SII −34.5% (95% CI –36.3% to –32.7%); RCI −19.0% (95% CI –20.1% to –17.9%)). Similarly, measured high BG (SII −26.6% (95% CI –28.6% to –24.7%); RCI −15.7% (95% CI –16.9% to–14.5%)) showed higher absolute and relative educational inequality as compared with self-reported high BG (SII −20.6% (95% CI –22.3% to –18.8%); RCI −11.5% (95% CI –12.5% to –10.5%)).
Table 3Educational inequality in different measures of health outcome
SII (95% CI) | RCI (95% CI) | |
Self-reported high BP | −34.5 (−36.3, -32.7) | −19.0 (−20.1, -17.9) |
Measured high BP* | −45.4 (−47.3, -43.4) | −26.6 (−27.9, -25.3) |
Self-reported high BG | −20.6 (−22.3, -18.8) | −11.5 (−12.5, -10.5) |
Measured random BG † | −26.6 (−28.6, -24.7) | −15.7 (−16.9, -14.5) |
*An individual is classified as having high BP if they have SBP≥140 mm Hg or DBP≥90 mm Hg at the time of the survey or they are currently taking medicine to lower their blood pressure.
†An individual is classified as having high blood glucose if he/she has a random blood glucose level of more than 140 mg/dL.
BG, blood glucose; BP, blood pressure; DBP, diastolic blood pressure; RCI, Relative Concentration Index; SBP, systolic blood pressure; SII, Slope Index of Inequality.
Discussion
In this study, we assessed educational inequality in relation to self-reported and measured high BP and BG, drawing on data for the state of Kerala from India’s DHS. We have three major inequality-related findings:
First, the broad pattern of inequality in Kerala suggests that for both high BP and BG, whether self-reported or clinically measured, those with the lowest educational attainment are at a disadvantage. This was seen in unadjusted and age-sex-adjusted OR measures, as well as using both absolute (SII) and relative (RII) measures of inequality. This has been seen in a number of other studies in India and other LMIC settings, including in Brazil.1 The association between less educated population with high BP and BG has been shown in high-income countries contexts, but not always showing the same pattern, for example, less educated populations (women) having greater BP in the UK51 and Hong Kong,17 but not in Chile.34 Research51 52 suggests complex gendered patterns of education-related inequalities in measured high BP and high BG. Our initial age-sex-adjusted ORs did not suggest this, but further analysis is required since in another study conducted in Kerala, women with low education were more often tested but also more often had high BP.14
Second, in relative terms, the magnitude of education-related inequality in high BP was found to be greater than that of high BG, whether self-reported or measured, and in both absolute and relative terms. Although there is no established threshold of inequality levels suggesting that one or the other should be prioritised, the greater prevalence of high BP in the overall population may be related to this (almost one in three persons reporting high measured BP, as compared with roughly one in five reporting measured high BG). Further research examining the relationship between overall prevalence and magnitudes of inequalities could shed further light on this and indicate whether, indeed, screening emphasis for low-education groups should prioritise BP testing, particularly in resource-constrained settings. It is clear that primary and secondary prevention options must be made more widely available to less educated groups. In Kerala, our ongoing fieldwork has found that BP testing is offered on a routine basis for adults accessing primary healthcare facilities—ensuring that populations without or with only primary-level education are accessing these services should be an area of focus in programming. Qualitative fieldwork (yet unpublished) demonstrates a demand across the state for primary prevention strategies like programmes and spaces for yoga and wellness, ensuring that awareness and access to these programmes extend to those with populations with lower educational attainment—who may be sequestered in certain occupations that constrain their time availability—will be important. Thus far, global evidence on individual-level interventions for less educated populations is limited53; this is clearly an area for future research as well.
Third, we found that the magnitude of both absolute and relative inequality measures was lower for self-reported indicators than measured ones. This means that there may be ‘hidden’ inequalities in NCD risk factors in the state and points to the need to continue to ramp up and routinise testing in the state. The national primary healthcare programme has created guidelines for population-based testing,54 which seems appropriate given this finding. On the research front, our finding be lies in Vellakkal et al’s study global study, which found that inequalities for self-reported measures were positive and greater than clinically measured inequalities.55 At the state level, we found education-related inequalities to be of smaller magnitudes in self-report with larger inequalities in the measured prevalence of high BP and BG. Interestingly, both Vellakal and we draw the conclusion14 36 55 that more reliable estimates of prevalence and of inequality may be derived from measurement.
Finally, as has been seen in other countries, an upstream approach to address health inequalities will be intervention in the dimension of inequality itself: education. While traditionally, Kerala’s education access and attainment have been high, studies have been finding barriers to the pursuit of education among some population subgroups56 57 in northern districts of the state as well as those in disadvantaged caste and tribal status groups. The compounding of these exclusions is important to understand further, as are studies and policy instruments to address them. A number of studies have focused on wealth-related inequality in the detection and control of NCDs and concluded that the major share of SES inequality in NCD screening and control at the moment in middle-income countries is reflected in disparities between rural and urban areas as well as lower and more educated individuals.58–62 There has been little attention has been given to other socioeconomic indicators, such as education, which also contributed to socioeconomic inequality in NCDs. Our analysis, therefore, makes a unique contribution to fill this gap.
At the macroeconomic level, moreover, global evidence suggests that left of centre governments and welfare states tend to be associated with lower inequalities. In Kerala, leftist or liberal progressive governments have been in power for the entire duration of the state’s existence. Schemes, such as Comprehensive Health Insurance Scheme, Aardram Mission, Cancer Suraksha, Hridyam Mission, Karunya Arogya Suraksha Padhathi, Medical Insurance Scheme State Employers and Pensioners, Nutritious Child63–67 and movements such as the Kerala Sasthra Sahithya Parishad, the people’s science campaign, could also be explored in terms of their current impact, and areas needing focus going forward.
Limitations
The study provides empirical evidence on the comparison of educational inequalities in the self-reported and measured prevalence of high BP and high BG in Kerala, India. However, there are some limitations that require mention. Because the study used cross-sectional data, it is not possible to attribute causation and establish associations between explanatory factors and NCDs. Another drawback of the study is that since the estimations of disease prevalence are based on self-reported data, they might be prone to under-reporting or over-reporting. Further, absent global thresholds, interpretation of the magnitude of inequality is difficult. A more nuanced and conclusive interpretation of the magnitude of inequality would require an analysis that compares Kerala to national or global data, which would be an important next analysis to carry out.
Conclusions
Our secondary analysis of 2019–2020 DHS data for Kerala suggests that there exist education-related inequalities in the prevalence of high BP and BG using both self-reported and measured approaches. In a state that has a strong record of addressing social welfare and prioritising the unreached, our findings suggest that research and programme efforts need to be redoubled to determine what is driving greater vulnerability to NCDs among populations with lower educational attainment on the one hand, and the possible role that improving education access can be on health outcomes, on the other hand.
The authors are grateful to the Demographic and Health Survey (DHS) Programme, for assembling and publishing accurate, nationally representative data on a range of health, biomarkers and healthcare utilisation indicators for population in the age range of 15 years and older. The authors are also grateful to NFHS’s project partners, International Institute for Population Sciences (IIPS) and Ministry of Health and Family Welfare, Government of India. We acknowledge the Other TGI researcher at the George Institute for Global Health for their valuable comments and suggestions.
Data availability statement
Data are available on reasonable request. All data used in the study is archived in the public repository of Demographic and Health Survey (DHS). The data can be accessed using: https://dhsprogram.com/data/dataset_admin/index.cfm, which requires registration.
Ethics statements
Patient consent for publication
Not applicable.
Twitter @sun2138
Contributors SKS: data analysis, writing, and manuscript preparation; DN: fund acquisition, conceptualisation, survey design, training, monitoring, supervision, manuscript preparation, and review; JJ: editing and review. All the authors accepts full responsibility for the work and/or the conduct of the study, had access to the data, and controlled the decision to publish
Funding This work was supported by The Wellcome Trust/DBT India Alliance Fellowship (grant number IA/CPHI/16/1/502653) awarded to DN.
Competing interests None declared.
Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.
1 Dulgheroff PT, da Silva LS, Rinaldi AEM, et al. Educational disparities in hypertension, diabetes, obesity and smoking in Brazil: a trend analysis of 578 977 adults from a national survey, 2007-2018. BMJ Open 2021; 11: e046154. doi:10.1136/bmjopen-2020-046154
2 Geldsetzer P, Manne-Goehler J, Theilmann M, et al. Diabetes and hypertension in India: a nationally representative study of 1.3 million adults. JAMA Intern Med 2018; 178: 363–72. doi:10.1001/jamainternmed.2017.8094
3 Mulyanto J, Kringos DS, Kunst AE. Socioeconomic inequalities in the utilisation of hypertension and type 2 diabetes management services in Indonesia. Trop Med Int Health 2019; 24: 1301–10. doi:10.1111/tmi.13303
4 WHO. Non communicable diseases. 2021. Available: https://www.who.int/news-room/fact-sheets/detail/noncommunicable-diseases [Accessed 09 Aug 2022 ].
5 Indian Council of Medical Research, Public Health Foundation of India, Institute for Health Metrics and Evaluation. The India state-level disease burden initiative - public health foundation of India. Health of the Nation’s States India; 2017. Available: https://phfi.org/the-work/research/the-india-state-level-disease-burden-initiative/
6 IHME, GBD. GBD compare | IHME viz hub. 2017. Available: http://vizhub.healthdata.org/gbd-compare [Accessed 09 Aug 2022 ].
7 IHME. GBD compare | IHME viz hub. 2017. Available: http://vizhub.healthdata.org/gbd-compare [Accessed 09 Aug 2022 ].
8 United Nations. Transforming our world: the 2030 agenda for sustainable development | Department of economic and social affairs [ United Nations 2015 ]. 2015. Available: https://sdgs.un.org/2030agenda [Accessed 10 Aug 2022 ].
9 NITI Aayog. SDG india index & dashboard 2019-20 [ Government of India ]. 2019. Available: http://14.139.60.153/handle/123456789/13652 [Accessed 10 Aug 2022 ].
10 MOSPI, GOI. Key indicators of household social consumption on education in India NSS 75TH round (July 2017- June 2018) [ Governmnet of India ]. 2019. Available: https://pib.gov.in/Pressreleaseshare.aspx?PRID=1593251 [Accessed 10 Aug 2022 ].
11 Ottapura Prabhakaran A. Burden of noncommunicable disease in Kerala; 2015.
12 Thankappan KR, Shah B, Mathur P, et al. Risk factor profile for chronic non-communicable diseases: results of a community-based study in Kerala, India. Indian J Med Res 2010; 131: 53–63.
13 IIPS, ICF Macro. National, state and union territory, and district fact sheets 2019-21 national family health survey NFHS5 (English). 2021. Available: https://dhsprogram.com/publications/publication-OF43-Other-Fact-Sheets.cfm [Accessed 01 Aug 2022 ].
14 Negi J, Sankar D H, Nair AB, et al. Intersecting sex-related inequalities in self-reported testing for and prevalence of Non-Communicable Disease (NCD) risk factors in Kerala. BMC Public Health 2022; 22: 544. doi:10.1186/s12889-022-12956-w
15 Menon J, Vijayakumar N, Joseph JK, et al. Below the poverty line and non-communicable diseases in Kerala: the epidemiology of non-communicable diseases in rural areas (ENDIRA) study. Int J Cardiol 2015; 187: 519–24. doi:10.1016/j.ijcard.2015.04.009
16 Prinja S, Jeyashree K, Rana S, et al. Wealth related inequalities in self reported morbidity: positional objectivity or epidemiological transition? Indian J Med Res 2015; 141: 438–45. doi:10.4103/0971-5916.159290
17 Chung GKK, Lai FTT, Yeoh E-K, et al. Educational inequality in physician-diagnosed hypertension widened and persisted among women from 1999 to 2014 in Hong Kong. Sci Rep 2019; 9: 14361. doi:10.1038/s41598-019-50760-6
18 Grossman M. Education and nonmarket outcomes. In: Handbook of the Economics of Education. Elsevier, 2006: 577–633. Available: https://ideas.repec.org/h/eee/educhp/1-10.html
19 Karter AJ, Stevens MR, Brown AF, et al. Educational disparities in health behaviors among patients with diabetes: the translating research into action for diabetes (TRIAD) study. BMC Public Health 2007; 7: 308. doi:10.1186/1471-2458-7-308
20 Lamb KE, Crawford D, Thornton LE, et al. Educational differences in diabetes and diabetes self-management behaviours in who SAGE countries. BMC Public Health 2021; 21: 2108. doi:10.1186/s12889-021-12131-7
21 Leng B, Jin Y, Li G, et al. Socioeconomic status and hypertension: a meta-analysis. J Hypertens 2015; 33: 221–9. doi:10.1097/HJH.0000000000000428
22 Mathisen J, Jensen AKG, Andersen I, et al. Education and incident type 2 diabetes: quantifying the impact of differential exposure and susceptibility to being overweight or obese. Diabetologia 2020; 63: 1764–74. doi:10.1007/s00125-020-05150-3
23 Oshio T, Kan M. Educational level as a predictor of the incidences of non-communicable diseases among middle-aged Japanese: a hazards-model analysis. BMC Public Health 2019; 19: 852. doi:10.1186/s12889-019-7182-6
24 Petrelli A, Sebastiani G, Di Napoli A, et al. Education inequalities in cardiovascular and coronary heart disease in Italy and the role of behavioral and biological risk factors. Nutr Metab Cardiovasc Dis 2022; 32: 918–28. doi:10.1016/j.numecd.2021.10.022
25 Rodin D, Stirbu I, Ekholm O, et al. Educational inequalities in blood pressure and cholesterol screening in nine European countries. J Epidemiol Community Health 2012; 66: 1050–5. doi:10.1136/jech-2011-200273
26 Zeng Y, Wu J, Han Y, et al. Educational disparities in the associations between self-monitoring of blood glucose and glycemic control in type 2 diabetes patients in Xiamen, China. J Diabetes 2018; 10: 715–23. doi:10.1111/1753-0407.12651
27 Cutler DM, Lleras-Muney A. Education and health: evaluating theories and evidence. National Bureau of Economic Research; 2006. Available: https://www.nber.org/papers/w12352 [Accessed 10 Aug 2022 ].
28 Kumar NA, George KK. Kerala’s education system: from inclusion to exclusion? Econ Polit Wkly 2009; 44: 55–61.
29 Cholakkal I, Radhakrishnan N. Education complements health status of households in Kerala: a study of Malappuram district. IOSRJHSS 2014; 19: 50–5. doi:10.9790/0837-19155055
30 Dray-Spira R, Gary TL, Brancati FL. Socioeconomic position and cardiovascular disease in adults with and without diabetes: United States trends, 1997-2005. J Gen Intern Med 2008; 23: 1634–41. doi:10.1007/s11606-008-0727-5
31 Redondo A, Benach J, Subirana I, et al. Trends in the prevalence, awareness, treatment, and control of cardiovascular risk factors across educational level in the 1995-2005 period. Ann Epidemiol 2011; 21: 555–63. doi:10.1016/j.annepidem.2011.02.008
32 Eggen AE, Mathiesen EB, Wilsgaard T, et al. Trends in cardiovascular risk factors across levels of education in a general population: is the educational gap increasing? The Tromsø study 1994-2008. J Epidemiol Community Health 2014; 68: 712–9. doi:10.1136/jech-2013-203428
33 Bidulescu A, Ferguson TS, Hambleton I, et al. Educational health disparities in hypertension and diabetes mellitus among African descent populations in the Caribbean and the USA: a comparative analysis from the Spanish town cohort (Jamaica) and the Jackson heart study (USA). Int J Equity Health 2017; 16: 33. doi:10.1186/s12939-017-0527-9
34 Nazzal C, Shea S, Castro-Diehl C, et al. Educational inequalities in cardiovascular risk factor and blood pressure control in the elderly: comparison of MESA cohort and Chilean NHS survey outcome measures. Glob Heart 2018; 13: 19–26. doi:10.1016/j.gheart.2017.09.001
35 Sharma SK, Vishwakarma D, Puri P. Gender disparities in the burden of non-communicable diseases in India: evidence from the cross-sectional study. Clin Epidemiology Glob Health 2020; 8: 544–9. doi:10.1016/j.cegh.2019.11.011
36 Vellakkal S, Subramanian SV, Millett C, et al. Socioeconomic inequalities in non-communicable diseases prevalence in India: disparities between self-reported diagnoses and standardized measures. PLOS ONE 2013; 8: e68219. doi:10.1371/journal.pone.0068219
37 IIPS, ICF. National family health survey, 2015-16. Mumbai International Institute for Population Sciences; 2017. Available: https://dhsprogram.com/methodology/survey/survey-display-355.cfm
38 WHO. A global brief on hypertension: silent killer, global public health crisis: World Health Day 2013. 2013. Available: https://www.who.int/publications-detail-redirect/a-global-brief-on-hypertension-silent-killer-global-public-health-crisis-world-health-day-2013
39 Galobardes B, Shaw M, Lawlor DA, et al. Indicators of socioeconomic position (part 1). J Epidemiol Community Health 2006; 60: 7–12. doi:10.1136/jech.2004.023531
40 Muraleedharan M, Chandak AO. Emerging challenges in the health systems of Kerala, India: qualitative analysis of literature reviews. JHR 2022; 36: 242–54. doi:10.1108/JHR-04-2020-0091
41 Paul K, Chaudhary M, Chowdhary R, et al. Changes in levels of morbidity and hospitalisation in kerala: a district level analysis (1995–2014). Clin Epidemiology Glob Health 2020; 8: 21–8. doi:10.1016/j.cegh.2019.03.011
42 Alves RFS, Faerstein E. Educational inequality in the occurrence of abdominal obesity:Pró-Saúde study. Rev Saude Publica 2015; 49: 65. doi:10.1590/S0034-8910.2015049005786
43 Alves RFS, Faerstein E. Educational inequalities in hypertension: complex patterns in intersections with gender and race in Brazil. Int J Equity Health 2016; 15: 146. doi:10.1186/s12939-016-0441-6
44 Bartolini L, Caranci N, Gnavi R, et al. Educational inequalities in the prevalence and outcomes of diabetes in the Emilian longitudinal study. Nutr Metab Cardiovasc Dis 2020; 30: 1525–34. doi:10.1016/j.numecd.2020.04.032
45 Barros AJD, Victora CG. Measuring coverage in MNCH: determining and interpreting inequalities in coverage of maternal, newborn, and child health interventions. PLOS Med 2013; 10: e1001390. doi:10.1371/journal.pmed.1001390
46 Harper S, Lynch J. Methods for measuring cancer disparities - relevant to healthy people 2010 objectives - SEER publications. Bethesda (Maryland) National Cancer Institute; 2005. Available: https://seer.cancer.gov/archive/publications/disparities/index.html
47 Harper S, Lynch J. Selected comparisons of measures of health disparities - SEER publications. Bethesda (Maryland) National Cancer Institute; 2007. Available: https://seer.cancer.gov/publications/disparities2/index.html
48 World Health Organization. Handbook on health inequality monitoring with a special focus on low- and middle-income countries. Geneva; 2013. Available: https://www.who.int/data/inequality-monitor/handbook [Accessed 12 Dec 2022 ].
49 da Silva ICM, Restrepo-Mendez MC, Costa JC, et al. Measurement of social inequalities in health: concepts and methodological approaches in the Brazilian context. Epidemiol E Serv Saude Rev Sist Unico Saude Bras 2018; 27: e000100017.
50 Victora CG, Vaughan JP, Barros FC, et al. Explaining trends in inequities: evidence from Brazilian child health studies. Lancet 2000; 356: 1093–8. doi:10.1016/S0140-6736(00)02741-0
51 Bann D, Fluharty M, Hardy R, et al. Socioeconomic inequalities in blood pressure: co-ordinated analysis of 147,775 participants from repeated birth cohort and cross-sectional datasets, 1989 to 2016. BMC Med 2020; 18: 338. doi:10.1186/s12916-020-01800-w
52 Neufcourt L, Deguen S, Bayat S, et al. Gender differences in the association between socioeconomic status and hypertension in France: a cross-sectional analysis of the CONSTANCES cohort. PLoS ONE 2020; 15: e0231878. doi:10.1371/journal.pone.0231878
53 Vilhelmsson A, Östergren P-O. Reducing health inequalities with interventions targeting behavioral factors among individuals with low levels of education-a rapid review. PLOS ONE 2018; 13: e0195774. doi:10.1371/journal.pone.0195774
54 NHM, MoHFW. Operational guidelines prevention, screening and control of common non-communicable diseases: hypertension, diabetes, common cancers (oral, breast, cervix). National Health Mission; Ministry of Health & Social Welfare India; 2016. Available: https://nhsrcindia.org/sites/default/files/2021-06/Operational%20Guideline%20Comprehensive%20Primary%20Health%20Care.pdf [Accessed 07 Sep 2022 ].
55 Vellakkal S, Millett C, Basu S, et al. Are estimates of socioeconomic inequalities in chronic disease artefactually narrowed by self-reported measures of prevalence in low-income and middle-income countries? Findings from the WHO-SAGE survey. J Epidemiol Community Health 2015; 69: 218–25. doi:10.1136/jech-2014-204621
56 John KT, Ali RTM, Rejikumar G. Bowed, bent and broken: investigating enrolments of scheduled castes/tribes to technical higher education programmes in Kerala. Contemporary Voice of Dalit 2022; 2455328X221091621. doi:10.1177/2455328X221091621
57 Tripathi H. Higher education in Kerala: development, equity and issues of access. South Asia Res 2019; 39: 42S–60S. doi:10.1177/0262728019872054
58 García MF, Hessel P, Rodríguez-Lesmes P. Wealth and inequality gradients for the detection and control of hypertension in older individuals in middle-income economies around 2007-2015. PLOS ONE 2022; 17: e0269118. doi:10.1371/journal.pone.0269118
59 Dieteren CM, O’Donnell O, Bonfrer I. Prevalence and inequality in persistent undiagnosed, untreated, and uncontrolled hypertension: evidence from a cohort of older mexicans. PLOS Glob Public Health 2021; 1: e0000114. doi:10.1371/journal.pgph.0000114
60 Redmond N, Baer HJ, Hicks LS. Health behaviors and racial disparity in blood pressure control in the National health and nutrition examination survey. Hypertension 2011; 57: 383–9. doi:10.1161/HYPERTENSIONAHA.110.161950
61 Berry KM, Parker W-A, Mchiza ZJ, et al. Quantifying unmet need for hypertension care in South Africa through a care cascade: evidence from the SANHANES, 2011-2012. BMJ Glob Health 2017; 2: e000348. doi:10.1136/bmjgh-2017-000348
62 Scholes S, Conolly A, Mindell JS. Income-based inequalities in hypertension and in undiagnosed hypertension: analysis of health survey for England data. J Hypertens 2020; 38: 912–24. doi:10.1097/HJH.0000000000002350
63 Government of Kerala. Cancer suraksha. Kerala Social Security Mission. Available: https://socialsecuritymission.gov.in/scheme_info.php?id=MQ== [Accessed 08 Sep 2022 ].
64 State Health Agency Kerala (SHA). Karunya Arogya Suraksha Padhathi–sha. Available: https://sha.kerala.gov.in/karunya-arogya-suraksha-padhathi/ [Accessed 08 Sep 2022 ].
65 Neelambaran A. Kerala launches ‘nutritious childhood’ scheme to eradicate malnutrition among children [ NewsClick ]. 2022. Available: https://www.newsclick.in/Kerala-Launches-Nutritious-Childhood-Scheme-Eradicate-Malnutrition-Among-Children [Accessed 08 Sep 2022 ].
66 MEDISEP. Medical insurance for state employess and pensioners. 2022. Available: https://medisep.kerala.gov.in/
67 National Health Mission (NHM). Government of Kerala. Aardram–National Health Mission. 2022. Available: https://arogyakeralam.gov.in/2020/04/01/aardram/
You have requested "on-the-fly" machine translation of selected content from our databases. This functionality is provided solely for your convenience and is in no way intended to replace human translation. Show full disclaimer
Neither ProQuest nor its licensors make any representations or warranties with respect to the translations. The translations are automatically generated "AS IS" and "AS AVAILABLE" and are not retained in our systems. PROQUEST AND ITS LICENSORS SPECIFICALLY DISCLAIM ANY AND ALL EXPRESS OR IMPLIED WARRANTIES, INCLUDING WITHOUT LIMITATION, ANY WARRANTIES FOR AVAILABILITY, ACCURACY, TIMELINESS, COMPLETENESS, NON-INFRINGMENT, MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Your use of the translations is subject to all use restrictions contained in your Electronic Products License Agreement and by using the translation functionality you agree to forgo any and all claims against ProQuest or its licensors for your use of the translation functionality and any output derived there from. Hide full disclaimer
© 2023 Author(s) (or their employer(s)) 2023. Re-use permitted under CC BY-NC. No commercial re-use. See rights and permissions. Published by BMJ. http://creativecommons.org/licenses/by-nc/4.0/ This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/ . Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.
Abstract
Objective
This study assesses educational inequalities in measured as well as self-reported high blood pressure (BP) and high blood glucose (BG) in the southern Indian state of Kerala, which is known to have high chronic disease morbidity.
Design
The present findings are drawn from a large-scale, nationally representative cross-sectional study.
Settings and participants
India’s Demographic and Health Survey (conducted in 2019–2021) had data on 36 526 individuals aged 15 years and above in the state of Kerala, India.
Primary and secondary outcomes measures
Measured high BP and BG; self-reported high BP and BG; as well as self-reported BP and BG testing. Descriptive statistics, bivariate analysis, along with multivariate statistics, were used. Educational inequalities were assessed through absolute and relative complex measures of inequality, namely the Slope Index of Inequality (SII) and Relative Concentration Index (RCI), respectively, with 95% CIs.
Results
The largest margin of inequality in Kerala, between the least and the most educated groups, was observed for measured high BP (57.7% and 17.6%). Measured high BP (SII −45.4% (95% CI –47.3% to –43.4%); RCI −26.6% (95% CI –27.9% to –25.3%)), self-reported high BP (SII −34.5% (95% CI –36.3% to –32.7%); RCI −19.0% (95% CI –20.1% to –17.9%)). High BG levels were concentrated among those with lower educational attainment (SII −26.6% (95% CI –28.6% to –24.7%); RCI −15.7% (95% CI –16.9% to –14.5%)), represented by negative SII and RCI values.
Conclusions
The study findings suggest that research and programme efforts need to be redoubled to determine what is driving greater vulnerability to non-communicable diseases among population with lower educational attainment on the one hand and the possible role that improving education access can be on health outcomes, on the other hand. Further research should explore relevant intersections with low education.
You have requested "on-the-fly" machine translation of selected content from our databases. This functionality is provided solely for your convenience and is in no way intended to replace human translation. Show full disclaimer
Neither ProQuest nor its licensors make any representations or warranties with respect to the translations. The translations are automatically generated "AS IS" and "AS AVAILABLE" and are not retained in our systems. PROQUEST AND ITS LICENSORS SPECIFICALLY DISCLAIM ANY AND ALL EXPRESS OR IMPLIED WARRANTIES, INCLUDING WITHOUT LIMITATION, ANY WARRANTIES FOR AVAILABILITY, ACCURACY, TIMELINESS, COMPLETENESS, NON-INFRINGMENT, MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Your use of the translations is subject to all use restrictions contained in your Electronic Products License Agreement and by using the translation functionality you agree to forgo any and all claims against ProQuest or its licensors for your use of the translation functionality and any output derived there from. Hide full disclaimer
Details


1 Healthier Societies, The George Institute for Global Health India, New Delhi, Delhi, India
2 Healthier Societies, The George Institute for Global Health India, New Delhi, Delhi, India; Faculty of Medicine, University of New South Wales, Sydney, New South Wales, Australia; Prasanna School of Public Health, Manipal Academy of Higher Education, Manipal, India