Hypertension Diagnosis, Treatment, and Control in India

Key Points Question What are the gaps in the hypertension care continuum (diagnosis, treatment, and control), and how do they vary by state, district, and sociodemographic groups in India? Findings In this 2019-2021 national survey study of 1.7 million respondents, 28.1% had hypertension, of whom 36.9% received a diagnosis; 44.7% of those who received a diagnosis (17.7% of the total with hypertension) reported taking medication, and 52.5% of those treated (8.5% of the total with hypertension) achieved blood pressure control. Most of the variability in the hypertension care continuum was observed within, not between, states. Meaning This study suggests that differences between states and between districts in states indicate the need for targeted, decentralized solutions to improve the hypertension care continuum in India.


Introduction
Hypertension is associated with 12.8% of all deaths globally. 1Many countries have implemented large-scale programs to diagnose and manage hypertension and other chronic diseases, with varying success. 2,3[6] To address the burden of noncommunicable diseases, India launched the National Programme for Prevention and Control of Cancer, Diabetes, Cardiovascular Diseases and Stroke in 2010 (now known as the National Programme for Prevention and Control of Non-Communicable Diseases) under the National Health Mission for 100 districts across 21 states. 7However, few data are currently available to assess the success of and opportunities for improved control of high blood pressure at subnational levels. 8[15] Our primary objective was to describe the national, state-level, and district-level hypertension care continuum (prevalence, diagnosis, treatment, and control) in India, the world's most populous country.Our secondary objective was to represent these data through a publicly available dashboard for stakeholders to help identify priorities for reducing the burden of hypertension in India and tracking the progress of national initiatives.

Study Population
The Fifth National Family Health Survey (NFHS) is a nationally representative survey conducted in 2 phases from June 17, 2019, to March 21, 2020, and from November 21, 2020, to April 30, 2021, in 707 districts from 28 states and 8 union territories and powered to provide estimates at the district level. 16Using a multistage stratified approach, 636 699 households within primary sampling units were randomly sampled from a list of households in which eligible participants (female participants, aged 15-49 years; male participants, aged 15-54 years) resided. 17 We restricted our analysis to nonpregnant female participants and male participants who had a valid measurement of blood pressure (eFigure 1 in Supplement 1).The analytic sample consisted of

Data Collection
Trained examiners measured systolic and diastolic blood pressure 3 times at 5-minute intervals using standardized protocols and validated, self-calibrating electronic blood pressure monitors (OMRON Healthcare Inc) after a 5-minute period when the participant was asked to sit comfortably. 17The respondent was also asked to avoid eating, smoking, and exercising for 30 minutes before the measurement.The cuff size of the blood pressure monitor was based on the circumference of the bare upper arm measured using a Gulick tape measure.Blood pressure was measured on the left arm, positioned so that it was at heart level with the cuff placed over bare skin or over thin clothes.
Consistent with 2016 Indian Council of Medical Research (ICMR) guidelines, we took the lowest of the first 2 measurements if their difference in systolic blood pressure was less than or equal to 5 mm Hg and the lowest of the 3 measurements otherwise. 18The blood pressure measurements in the households selected were high for both women (urban, 88.3%; rural, 93.0%) and men (urban, 79.3%; rural, 85.7%).Details on the interexaminer and intraexaminer reliability of blood pressure measurement are unavailable. 16,17rticipants were also asked the question: "Before this survey, were you ever told you had high blood pressure by a doctor, nurse, or health practitioner on 2 or more occasions?"Medication status was asked only to those who self-reported a diagnosis of hypertension.

Hypertension Care Continuum: Diagnosis, Treatment, and Control
We defined hypertension as self-reported or, among those without a prior diagnosis, as measured blood pressure of 140/90 mm Hg or more. 18We defined the hypertension care continuum using the following metrics: proportion of individuals who self-reported a diagnosis (ie, self-reported diagnosed hypertension prior to the survey among the total with hypertension) and proportion of those individuals treated (ie, self-reporting medication use).We defined the proportion of individuals with blood pressure control among those treated (<140/90 mm Hg for those aged 18-79 years and <150/90 mm Hg for those aged 80 years or older) based on ICMR guidelines for the management of hypertension. 18We also provided age-standardized estimates of treatment and control among all of those with hypertension.The definitions are summarized in eTable 2 in Supplement 1.

Sociodemographic Variables
We estimated care continuum metrics by 3 individual-level sociodemographic factors: sex (male or female), age (18-39, 40-64, or Ն65 years), and schooling (none or missing, primary [up to 4th grade], secondary [up to 10th grade], or postsecondary).We also stratified by 2 household sociodemographic factors: rural residence (vs urban) and regional wealth quintile (urban and rural) from the household wealth index as provided by the NFHS. 19

Statistical Analysis
We report survey-weighted estimates accounting for the complex survey design and cluster-robust 95% CIs. 16Individual and household characteristics of the analytic sample were assessed by strata of residence (urban or rural) and sex.
Continuum performance indicators were estimated for the national sample, for states stratified by sociodemographic factors (residence, sex, age category, schooling, and regional wealth quintile), and for districts.Age-standardized estimates of the continuum indicators were computed for different strata at the national level based on the distribution of the total sample because different strata of schooling and wealth have different age distributions.We also calculated weighted estimates at the state and district levels that were not age standardized but would be relevant for local decision-making.We compared the estimates with those obtained when taking the mean of the last 2 blood pressure measurements as a sensitivity analysis.

JAMA Network Open | Global Health
To assess whether the differences in the care continuum were greater between or within states (ie, between districts), we partitioned the variance in the care continuum at both levels using variance partition coefficients from generalized linear mixed models with state-level and district-level intercepts, adjusting for sociodemographic variables. 20To illustrate the variability between and within states, we present examples of 2 states from regions with moderate to high burdens of hypertension, Karnataka from South India and Meghalaya from North East India.
To further aid policy and priority decision-making, we developed a dashboard 21 to visually depict the disparities in the hypertension care continuum using Shiny by RStudio.[24][25]

Results
The analytic sample of 1 691 036 adult respondents was 52.6% women and 47.4% men, with a mean (SD) age of 41.6 (16.5) years (Table 1).Nationally, more than three-fourths of the population lived in rural areas from 2019 to 2021.More than half the respondents were younger than 40 years of age,  third measurements of blood pressure (eTable 5 in Supplement 1) instead of the lowest measurements (eTable 6 and eFigure 3 in Supplement 1).

State-Level Care Continuum
The prevalence of hypertension was similar among the southern states (Kerala, Tamil Nadu, Karnataka, Telangana, and Andhra Pradesh), union territories (Andaman and Nicobar Islands, Lakshadweep, and Puducherry), and Goa compared with other parts of the country (Figure 2; median percentage of states: southern states, 29.9% [IQR, 29.1%-31.4%]vs rest of India, 26.8% [24.4%-32.0%]).A higher prevalence of hypertension was observed in urban vs rural areas for all states (eFigure 4 in Supplement 1).
The proportions of participants with diagnosed hypertension were similar between southern states and the rest of India (eFigure 5 in Supplement 1).However, the proportion of participants with treated hypertension and the proportion of participants with controlled hypertension were higher among the southern states.Disparities in diagnosis, treatment, and control between sociodemographic groups within each state beyond the state-level heterogeneity observed in

District-Level Care Continuum
There was considerable within-state variation in the hypertension care continuum (Figure 3) such that 94.7% of the variance in the proportion of participants who received a diagnosis (range, 6.3%-77.5%),93.6% of the variance in the proportion of participants treated among those who received a diagnosis (range, 8.7%-97.1%),and 97.3% of the variance in the proportion of participants with blood pressure control among those treated (range, 2.7%-76.6%)were at the district level or below, with the remaining at the state level (between states).We visualized this variability between and within states from all regions in eFigure 6 in Supplement 1.
We illustrated this variability in Meghalaya and Karnataka.In Meghalaya, the 5 districts of Garo  All columns are survey-weighted percentages in total population.We performed age standardization to the distribution of the within-sample total population separately for total population, population with hypertension, diagnosed population, and treated population.This procedure harmonizes the age distribution within each category (total, hypertension, diagnosed, and treated).The values should therefore not be sequentially multiplied to determine prevalence within total population.The values inside the bars are proportions of diagnosed hypertension among patients with hypertension, treated among diagnosed hypertension, and controlled among treated hypertension (from Table 2).Values above the bars are relative to all patients with hypertension (100 − % diagnosed among hypertension, 100 − % treated among hypertension, and 100 − % controlled among hypertension).All values are survey-weighted percentages (not age standardized).Undiagnosed are among those with hypertension.Untreated and uncontrolled are among those diagnosed with hypertension and among those treated, respectively.We report weighted estimates at the state level that were not age standardized and relevant for local decision-making in this article.prevalence, 41.1% [range, 36.7%-45.5%])and Khasi Hills (median prevalence, 29.4% [range, 27.7%-42.3%])(eFigure 7A in Supplement 1).
In Karnataka, there was substantial between-district heterogeneity in treatment among those who received a diagnosis but less heterogeneity in blood pressure control among those treated between districts with similar prevalences.Chikmagalur (31.5% [95% CI, 28.8%-34.3%]),Udupi Household and individual characteristics were collected using standardized instruments after obtaining written informed consent.The survey additionally collected data on blood pressure among all adults (aged Ն18 years) who were living in the same household as eligible participants.The overall approached sample consisted of 1 895 297 adults aged 18 to 98 years.This secondary analysis of publicly available deidentified data was exempted from ethical approval by the institutional review board of Emory University.This survey study followed the American Association for Public Opinion Research (AAPOR) reporting guideline and the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.

Figure 2
Figure2are published on the interactive Hypertension Care Continuum dashboard.21

Table 1 .
Characteristics of Participants in the Analytic Sample for Estimating Care Cascade of Hypertension in India (N = 1 691 036)and almost 90% were aged 18 to 64 years.The mean (SD) systolic blood pressure and the mean (SD) diastolic blood pressure were 120.4(18.3)mm Hg and 79.6 (10.5) mm Hg, respectively, for women and 124.7 (15.9) mm Hg and 81.6 (10.4) mm Hg, respectively, for men.

is g a r h Goa Haryana Haryana Himachal Himachal Pradesh Pradesh Tamil Tamil Nadu Nadu Uttar Pradesh Uttar Pradesh West West Bengal Bengal Andhra Andhra Pradesh Pradesh Puducherry Maharashtra Maharashtra
Mauer N, Geldsetzer P, Manne-Goehler J, et al.Longitudinal evidence on treatment discontinuation, adherence, and loss of hypertension control in four middle-income countries.Sci Transl Med.2022;14(652): eabi9522.doi:10.1126/scitranslmed.abi952245.Datta BK, Ansa BE, Husain MJ.An analytical model of population level uncontrolled hypertension management: a care cascade approach.J Hum Hypertens.2022;36(8):726-731.doi:10.1038/s41371-021-00572-x46.Lamprea-Montealegre JA, Zelnick LR, Hall YN, Bansal N, de Boer IH.Prevalence of hypertension and cardiovascular risk according to blood pressure thresholds used for diagnosis.Hypertension.2018;72(3):602-609.doi:10.1161/HYPERTENSIONAHA.118.1160947.Oksanen T, Kivimäki M, Pentti J, Virtanen M, Klaukka T, Vahtera J. Self-report as an indicator of incident disease.Ann Epidemiol.2010;20(7):547-554.doi:10.1016/j.annepidem.2010.03.017Characteristics of Participants in Analytic Sample Versus Those Excluded, N = 1,895,297 eTable 2. Definitions of Disease and Care Continuum of Hypertension 3. Crude Estimates of Socio-Demographic Variations in Care Continuum in India Relative to Population With Hypertension, N = 1,691,036 eTable 4. Socio-Demographic Variations in Care Continuum in India Relative to Population With Hypertension, N = 1,691,036 eTable 5. Average Values of Systolic and Diastolic Blood Pressure for Each of the Three Measurements, N = 1,691,036 eTable 6. Socio-Demographic Disparities in Hypertension Care Continuum in India When Using Average of Last 2 Blood Pressure Measurements, N = 1,691,036 eFigure 1. Flowchart of Analytic Sample eFigure 2. Distribution of Diagnosed Hypertension by Treatment and Control Status eFigure 3. Distribution of Diagnosed Hypertension by Treatment and Control Status When Using Average of Last 2 Blood Pressure Measurements eFigure 4. State-Level Care Continuum eFigure 5. Heatmap for Care Continuum eFigure 6. Distribution of Between-and Within-State Variability in Hypertension Care Continua for Selected States eFigure 7. Disparities Within States at the District Level eMethods.National Family Health Survey-5, 2019-21: Survey Design and Data Collection All values are survey-weighted percentages (not age standardized).Undiagnosed are among those with hypertension.Untreated and uncontrolled are among those diagnosed with hypertension and among those treated, respectively.We report weighted estimates at the district-level that were not age standardized and relevant for local decision-making in this article.44.