Abstract
Background:
Hypertension in women of reproductive age group is of special concern because of the vulnerability of women to pregnancy-induced hypertension apart from socio-cultural vulnerability.
Aim:
The objective of the study was to identify the predictors for hypertension among Indian women and to develop a risk score which would provide an opportunity for early detection and appropriate action.
Material and Methods:
This study was based on the data collected in National Family Health Survey in 2015–2016. Women in India of 15–49 years were the study population. Data were analysed using SPSS v17. Logistic regression analysis was carried and expressed as odds ratio with 95% confidence intervals to identify predictors of hypertension. The risk score for hypertension was developed after shrinkage of variables and by using regression coefficients obtained by standard Logistic Regression Model.
Results:
Among 6,87,230 women between 15 and 49 years, 77,788 (11.3%) were hypertensive. The study results revealed that there was an increasing trend in the prevalence of hypertension (26.5%) with increasing age, and with increasing weight (23.4%). Urban areas (12.3% vs 10.9%), alcoholics (19.2%) and various forms of tobacco users (14.8%) had more prevalence of hypertension.
Conclusion:
Age, residing in urban area, consuming tobacco products, consumption of alcohol, non-vegetarian diet and overweight, were found to be the significant predictor variables, and were used to develop the Risk Prediction score using logistic regression model.
Keywords: Hypertension, hypertension among women, nfhs-4
Introduction
According to the WHO, there are five important risk factors for non-communicable diseases (NCDs) and hypertension is one of them.[1] Hypertension contributes to higher disability-adjusted life years (DALYs) due to cardiovascular disease as compared to other metabolic risk factors such as High Total Cholesterol, High Fasting Blood Sugar, and High body mass index (BMI).[1] Globally, it has been reported that hypertensive heart disease accounts for 1.16 million deaths and 21.5 million DALYs per year with an annual prevalence of 18.6 million cases.[2] The prevalence of hypertension is increasing in India from 24% in 1980 to 26% in 2015 in comparison to developed nations like the United States which has shown a decline from 23% in 1980s to 12% in 2015.[1,2,3]
NCDs and mental health conditions are leading causes of mortality and morbidity for women globally.[4] The Global Burden of Diseases study has revealed that the leading causes of death for women are ischaemic heart disease (IHD) and stroke, both of them are associated with hypertension. Also, NCDs accounted for seven out of the ten leading causes of death and included hypertensive heart disease, stroke, chronic obstructive pulmonary disease, Alzheimer’s disease, diabetes, kidney disease and lung cancer.[5] Hypertension in women of reproductive age group is of special interest because of the vulnerability of women to pregnancy-induced hypertension, which is associated with poor maternal and foetal outcomes.[6,7,8] Women apart from their biological vulnerability are also socially and culturally vulnerable to hypertension – be it in poor health seeking behaviour, or suboptimal health system responsiveness. Conventionally, reproductive health needs have been prioritised among women, but with the epidemiological transition and the globalization of NCDs, attention needs to be drawn to addressing NCDs and their risk factors among women in reproductive age group.[9]
Several studies globally had documented risk factors for hypertension, including increasing age, family history, obesity, physical inactivity, smoking and poor dietary practices.[10,11,12,13] However very few risk prediction models for hypertension have been developed which are mostly from the United States and Europe, and a few from Asian countries.[14] Risk prediction models would be critical for risk communication and shared decision making for the intensity of prevention measures. They would also help us to project the burden of hypertension and resource allocation at population level. Development of a composite risk score and its validation for major NCDs is being attempted by Nangia R et al.[15] through their protocol paper which describes the methodology for development and validation of the composite risk score. However, the score development is based on data from the WHO-STEPS survey of only one State of India. The external validation is also attempted through local population which limits its generalizability. Another risk score for hypertension was developed by Sathish T et al.[16] for rural population in India which cannot be generalised to urban population. The available risk scores have been developed for general population, while disaggregated data for specific groups or sub populations are not being analysed. The National Family Health Survey conducted in India provides us an opportunity to study data of a large representative sample of population of women and develops a risk score by identifying the predictors for hypertension among them.
Materials and Methods
Data source: The current study is based on the data collected from women belonging to the reproductive age group (aged 15–49 years) from a nationally representative health survey namely the National Family Health Survey – 4. The survey was conducted in all 29 states and 6 union territories in India during the years 2015–2016.[17] [Individual Recode: IAIR74SV.ZIP file accessed from Demography and Health Survey (DHS)].
Inclusion and exclusion criteria: Among the selected households, those women who belong to the households and women who have come as visitors to the particular household and have stayed overnight were included. Women for whom all the readings of blood pressure data were not available were excluded.
Sampling Technique: Multistage sampling covers 640 districts in India. Primary sampling unit was a village in rural areas and census enumeration block in urban areas. Data were collected using Computer-Assisted Personal Interviewing on mini laptops.
Operational definitions
Dependent variable
Hypertension: For the purpose of the study, the dependent (outcome) variable was hypertension. Out of the three recordings of the blood pressure which were recorded with a gap of 5 min, the first reading was eliminated and average of the second and third reading was considered. Hypertension was defined as systolic blood pressure of at least 140 mm Hg or diastolic blood pressure of at least 90 mm Hg.[18] In addition, women currently on antihypertensive medication were also included as having hypertension.[17]
Independent variables
Use of Tobacco (Current use of tobacco in any form – either in smoke form or non-smoke form such as smoking cigarettes/bidis/cigars, a pipe, hookah or chewing of paan masala or other tobacco forms such as gutkha/khaini/paan with tobacco/other chewing tobaccos), Use of alcohol (Alcohol taken in any form currently), Age (In completed years), Education (Classification was based on years of completed schooling. Primary: Up to five years of completed schooling, Secondary: 6–12 years of completed schooling, Higher: More than 12 years of completed schooling), Region (The country has a wide geographical distribution and for the purpose of the study it is divided into six regions – Northern, Southern, Western, Eastern, North Eastern, and Central regions), Wealth Index (It is used as a proxy indicator for socioeconomic scale. The wealth index is, based on scores on ownership of consumer goods and household characteristics, such as availability of basic facilities like clean drinking water, owning of televisions and non-motor two-wheeler, type of housing, drinking water access and facilities for sanitation. All households are classified into wealth quintiles ranging from Richest to Poorest with in between classes as Richer, Middle and Poorer), Body Mass Index (A value of <18.5 was considered as underweight, 18.5–24.9 was considered as normal and >25 was considered as overweight and obese), Assessment of dietary intake (The frequency of foods consumed was categorised as daily, weekly, occasionally or never in relation to intake of milk or curd, pulses or beans, dark green leafy vegetables, fruits, eggs, fish, chicken or meat. Non-vegetarian were those who reported consuming fruits, vegetables, pulses or beans, animal products (chicken or meat, fish, eggs, milk or curd) either daily, weekly or occasionally).[19]
Statistical analysis
Descriptive analysis was initially run to understand the baseline characteristics of data and reported in frequencies and percentages. Continuous values were expressed using mean ± standard deviation (SD). Bivariate analysis was conducted using Pearson’s χ2 test. Later Logistic regression model (enter and forward likelihood ratio method) was adapted, model significance was measured by omnibus Chi-square (52252.39), -2 log likelihood (431967.20) with pseudo-R square value of 0.14. Odds ratio with corresponding 95% confidence intervals were calculated to derive the significant predictors for hypertension. Predictors with P value < 0.05 were considered as statistically significant. The risk score for developing hypertension was developed after shrinkage of variables by adopting above techniques and by using regression coefficients obtained by standard logistic regression model as.[20]
Individual risk for developing Hypertension = exp (Individual risk score) ÷ [1 + exp (Individual risk score)]
Six predictors were chosen for fitting the risk model based on likelihood ratios using standard and stepwise logistic regression model.
Individual risk score = intercept + (bage × age) + (bBMI × BMI) + (bresidence × residence) + (btobacco × tobacco) + (balcohol × alcohol) + (bfood habits × food habits).
b = Regression coefficients
The statistical package for data analysis was Statistical Package for Social Sciences for windows, version 17.0 (IBM SPSS Inc., Chicago, IL).
Ethical considerations
Institutional Ethics Committee, Post Graduate Institute of Medical Education and Research, Chandigarh (PGI/IEC/2021/001139) has approved the study.
Results
Out of a total sample of 6,99,686 of women of childbearing age (15–49 years), blood pressure data was incomplete for 12,456 (1.8%) women. Hence, data were analyzed for 6,87,230 women (98.2%). Among them, 11.3% were hypertensive. There was a significantly increasing trend in the prevalence of hypertension as age increased from 15 to 49 years. Among the women who had no education, prevalence of hypertension was 14.4% which decreased gradually to 8.5% as the educational status increased (p < 0.001). Among the States, Arunachal Pradesh (19.1%) had reported the highest prevalence followed by Sikkim (18.5%) and Assam (18.1%). Least prevalence was observed in Bihar (7.9%) and Rajasthan (8.1%). A significantly (p < 0.001) higher prevalence was observed in urban women (12.3%), Sikhs (16.2%), widowed (20.8%) and richest quintile (13.1%). Hypertension was reported in 13.9% of women who smoke cigarettes, 21% among those who smoke pipes, 14.5% among those who chew tobacco, 16% among those who use snuff, 15.6% among those who smokes cigars, 13.1% among those who use gutkha/paan masala with tobacco, 16.3% among those who use paan with tobacco and 14% among those who smokes other products. Overall prevalence of hypertension was 14.8% among women who use tobacco which was significantly higher (p < 0.001) than non-tobacco users (11%). A significantly higher prevalence of hypertension was observed among women who were alcoholics (19.2%) as compared to non-alcoholics (11.1%). Also, it was observed that among alcoholics, those who consume almost every day had more prevalence (21.8%) as compared to those who consume once a week (20.2%) and less than once a week (17.6%). Prevalence of hypertension was significantly higher among non-vegetarians’ diet (11.6%) as compared to vegetarians (10.5%) and among those who never consumed fried food (13.8%) as compared to those who consume it daily (13%). Also, it was significantly higher in those who never take aerated drinks (12.9%) as compared to who take it daily (12.2%). Prevalence of hypertension also had a significantly increasing trend with change in body mass index from underweight to overweight [Table 1].
Table 1.
Sociodemographic characteristics of hypertensive and non-hypertensive woman in the age group of 15-49 years, NFHS-4 (2015-2016)
| Variable | Hypertensive n (%) | Non-Hypertensive n (%) | Chi Square | P |
|---|---|---|---|---|
| Age group | 40748.49 | <0.001 | ||
| 15-19 | 3767 (3.1) | 118344 (96.9) | ||
| 20-24 | 5837 (4.8) | 114880 (95.2) | ||
| 25-29 | 8293 (7.3) | 104830 (92.7) | ||
| 30-34 | 10768 (11.3) | 84667 (88.7) | ||
| 35-39 | 14220 (16.0) | 74775 (84.0) | ||
| 40-44 | 15950 (21.2) | 59451 (78.8) | ||
| 45-49 | 18953 (26.5) | 52495 (73.5) | ||
| State (regions) | 3065.966 | <0.001 | ||
| Northern | 16590 (12.0) | 121145 (88.0) | ||
| Southern | 10549 (11.7) | 79280 (88.3) | ||
| Western | 6158 (11.2) | 48722 (88.8) | ||
| Eastern | 11960 (9.6) | 112371 (90.4) | ||
| Northeastern | 15300 (15.8) | 81552 (84.2) | ||
| Central | 17231 (9.4) | 166372 (90.6) | ||
| Residence | 291.776 | <0.001 | ||
| Urban | 24557 (12.3) | 174419 (87.7) | ||
| Rural | 53231 (10.9) | 435023 (89.1) | ||
| Religion | 1297.145 | <0.001 | ||
| Hindu | 54002 (10.6) | 456363 (89.4) | ||
| Muslim | 11741 (12.7) | 80810 (87.3) | ||
| Christian | 6892 (13.5) | 44238 (86.5) | ||
| Sikh | 2454 (16.2) | 12674 (83.8) | ||
| Buddhist | 1299 (14.7) | 7517 (85.3) | ||
| Jain | 146 (14.7) | 845 (85.3) | ||
| Jewish | 1 (14.3) | 6 (85.7) | ||
| Parsi/Zoroastrian | 0 (0.0) | 7 (100.0) | ||
| Other | 1176 (15.0) | 6670 (85.0) | ||
| No religion | 77 (19.8) | 312 (80.2) | ||
| Caste | 720.427 | <0.001 | ||
| Schedule Caste | 12634 (10.3) | 110207 (89.7) | ||
| Schedule Tribe | 15335 (12.3) | 109687 (87.7) | ||
| OBC | 27785 (10.3) | 241583 (89.7) | ||
| None of them | 17273 (12.5) | 120891 (87.5) | ||
| Don’t know | 447 (14.0) | 2743 (86.0) | ||
| Wealth index | 1152.76 | <0.001 | ||
| Poorest | 12451 (9.5) | 118987 (90.5) | ||
| Poorer | 15459 (10.5) | 132086 (89.5) | ||
| Middle | 16191 (11.2) | 128811 (88.8) | ||
| Richer | 16975 (12.5) | 118930 (87.5) | ||
| Richest | 16712 (13.1) | 110628 (86.9) | ||
| Education level | 3785.442 | <0.001 | ||
| No education | 27879 (14.4) | 165828 (85.6) | ||
| Primary | 11708 (13.4) | 75390 (86.6) | ||
| Secondary | 31623 (9.6) | 297300 (90.4) | ||
| Higher | 6578 (8.5) | 70924 (91.5) | ||
| Current marital status | 11754.71 | <0.001 | ||
| Never in union | 7318 (4.4) | 160218 (95.6) | ||
| Married | 65266 (13.3) | 426700 (86.7) | ||
| Widowed | 4164 (20.8) | 15875 (79.2) | ||
| Divorced | 450 (14.8) | 2595 (85.2) | ||
| Separated | 590 (12.7) | 4054 (87.3) | ||
| BMI | 23685.103 | <0.001 | ||
| Underweight | 9000 (6.0) | 141313 (94.0) | ||
| Normal | 39202 (9.6) | 370430 (90.4) | ||
| Overweight | 29392 (23.4) | 96432 (76.6) | ||
| Tobacco user | 795.164 | <0.001 | ||
| Yes | 8704 (14.8) | 49916 (85.2) | ||
| No | 69084 (11.0) | 559526 (89.0) | ||
| Drinks alcohol | 1070.964 | <0.001 | ||
| Yes | 3267 (19.2) | 13784 (80.8) | ||
| No | 74521 (11.1) | 595658 (88.9) | ||
| Frequency of alcohol intake | 27.028 | <0.001 | ||
| Almost every day | 473 (21.8) | 1697 (78.2) | ||
| About once a week | 1348 (20.2) | 5325 (79.8) | ||
| Less than once a week | 1446 (17.6) | 6762 (82.4) | ||
| Non-vegetarian | 159.852 | <0.001 | ||
| Yes | 58128 (11.6) | 442362 (88.4) | ||
| No | 19660 (10.5) | 167080 (89.5) |
Logistic regression analysis reveals that independent variables like increasing age, residing in urban area, people belonging to richest wealth index, education up to primary level, consuming tobacco products and alcohol, eating non-vegetarian diet, belonging to Muslim community, residing in Northeastern states, overweight BMI, widowed/divorced were significant predictors of hypertension [Table 2].
Table 2.
A logistic regression analysis to identify predictors for hypertensive among women in the age group of 15-49 years, NFHS 2015-2016
| Variable | OR | P | 95% CI | |
|---|---|---|---|---|
|
| ||||
| Lower | Upper | |||
| Age | ||||
| 15-19 | ||||
| 20-24 | 1.52* | 0.001 | 1.45 | 1.58 |
| 25-29 | 2.13* | 0.001 | 2.04 | 2.23 |
| 30-34 | 3.19* | 0.001 | 3.05 | 3.33 |
| 35-39 | 4.60* | 0.001 | 4.41 | 4.81 |
| 40-44 | 6.42* | 0.001 | 6.15 | 6.71 |
| 45-49 | 8.55* | 0.001 | 8.19 | 8.93 |
| Place of residence | ||||
| Rural | ||||
| Urban | 1.04* | 0.001 | 1.02 | 1.07 |
| Wealth index | ||||
| Poorest | ||||
| Poorer | 1.00 | 0.727 | 0.97 | 1.02 |
| Middle | 0.99 | 0.725 | 0.97 | 1.02 |
| Richer | 1.04* | 0.001 | 1.01 | 1.08 |
| Richest | 1.03* | 0.001 | 1.01 | 1.07 |
| Tobacco consumption | ||||
| No | ||||
| Yes | 1.05* | 0.07 | 1.01 | 1.10 |
| Alcohol consumption | ||||
| No | ||||
| Yes | 1.39* | 0.001 | 1.33 | 1.45 |
| Dietary (Non-vegetarian) | ||||
| No | ||||
| Non-vegetarian | 1.06* | 0.001 | 1.03 | 1.08 |
| Religion | ||||
| Hindu | ||||
| Muslim | 1.22* | 0.001 | 1.19 | 1.25 |
| Others | 1.00 | 0.767 | 0.97 | 1.02 |
| Caste | ||||
| Others | ||||
| SC | 0.97* | 0.009 | 0.94 | 0.99 |
| ST | 0.98 | 0.163 | 0.95 | 1.01 |
| OBC | 0.92* | 0.001 | 0.90 | 0.93 |
| State zones | ||||
| Northern | ||||
| Southern | 0.85* | 0.001 | 0.83 | 0.88 |
| Western | 0.93* | 0.001 | 0.90 | 0.97 |
| Eastern | 0.88* | 0.001 | 0.85 | 0.91 |
| North-eastern | 1.43* | 0.001 | 1.39 | 1.48 |
| Central | 0.86* | 0.001 | 0.84 | 0.89 |
| BMI | ||||
| Underweight | ||||
| Normal | 1.30* | 0.001 | 1.27 | 1.33 |
| Overweight | 2.84* | 0.001 | 2.77 | 2.92 |
| Marital status | ||||
| Married | ||||
| Unmarried/divorced/widowed/separated | 1.03* | 0.014 | 1.01 | 1.06 |
| Highest educational level | ||||
| No education | ||||
| Primary | 1.06* | 0.001 | 1.03 | 1.08 |
| Secondary | 0.97* | 0.003 | 0.95 | 0.99 |
| Higher | 0.80* | 0.001 | 0.77 | 0.83 |
Prediction score was calculated by using logistic regression model. To predict risk individually, the fitted risk model can be used to calculate the risk score [Table 3]. For example, the risk score for a 44-year-old female patient with a BMI of 24 kg/m2 who belongs to rural area who doesn’t smoke, drinks alcohol and consumes non vegetarian then the risk score would be calculated as: −5.968 + 0.074 × 44 + (0.059 × 24) + (0.012 × 0) + (0.027 × 0) + (0.424 × 1) +(0.144 × 1) = −5.968 + 5.24 =-0.728. Therefore, her predicted risk for developing hypertension would be: Exp (-0.728) ÷ {1 + exp (-0.728)} = 32.4%, when compared to a woman without these risk factors.
Table 3.
Results of Logistic Regression model (Exp B)
| Covariates | Beta coefficients | Value to be substituted |
|---|---|---|
| Respondent’s current age | 0.074 | Actual age |
| BMI | 0.059 | Actual BMI |
| Uses tobacco | 0.027 | 1 - Smoker; 0 - Non smoker |
| Place of residence | 0.012 | 1 - Urban; 0 - Rural |
| Drinks alcohol | 0.424 | 1 - Alcoholic; 0 - Non alcoholic |
| Non-veg | 0.144 | 1 - Non vegetarian; 0 - Vegetarian |
| Constant | -5.968 |
Discussion
The current study indicates that about 1 in 10 women in the age group of 15–49 years suffer from hypertension; the overall prevalence was found to be 11.3%. This would amount to an estimated number of around 282.5 million of hypertensive women in reproductive age requiring healthcare services.
Our study finding of 11.3% is consistent with the study done by Ghosh S et al.[18] However, the reported prevalence was lower when compared to the prevalence from other studies: Agrawal V et al.[21] (systolic, 17.0% and diastolic, 14.4%), Bhansali A et al (22.6%),[22] Talukdar et al. (18.69%).[23] A significant positive trend in prevalence of HTN was observed among females ranging from 6.9% to 13.1% in urban and from 3.5% to 8.8% in rural areas by P Devi et al.[24] in a systematic review of studies published from 1969 to July 2011. The reason for the lower prevalence in the present study could be due to the inclusion of different age groups and settings along with different criteria for defining hypertension.
There was an increase in the prevalence of hypertension from 3.1% in the 15–19 years age group to 26.5% in the 45–49 years age group. In addition to the biological changes, the changes in the lifestyle leading to increasing prevalence of risk factors as well could have contributed to these findings. Similar finding of a linear dose-response relationship of increased prevalence as age advances has been reported from earlier studies. The crude hypertension prevalence was found to be 23.6% among women ranging from 9.2% among women aged 18 to 25 years to 48.6% among women older than 65 years.[25] The relatively small percentages of prevalence of hypertensive subjects as compared to percentages of non-hypertensive subjects in the 15–19 years should not be ignored as India has a large population of young adults which is 236 million with a sex ratio of 908 meaning that the number of subjects that do have hypertension is alarmingly high.[26]
Interstate variations with higher prevalence than the national average from the Northeastern states (15.8%) and lower prevalence from central states (9.4%) were observed. Other studies have shown that in addition to the Northeastern states, Punjab, Himachal Pradesh and Kerala have also reported higher rates of hypertensive population which is around 35%.[25] The regional differences either in the dietary patterns or the prevalence of other risk factors like substance use could once again have contributed to higher prevalence which needs further exploration. The present study reported higher prevalence of hypertension among urban women than among rural women as was found in the previous studies in the National Capital Region (42.2% urban and 28.9% rural NCR).[27,28] The pressure for double income among the lower-income group and middle-income groups as well as the prioritisation of the health of the other members in the family in terms of health expenditure in the urban areas is well known.
The prevalence of hypertension was highest among Sikhs (16.2%) as compared to other religions among which Hindus (10.6%), Muslims (12.7%), Christian (13.5%), Buddhist (14.7%) while the previous studies showed highest prevalence among Muslims (68.7%) followed by Hindus (21%) and Christians (10.3%).[29] The religious and cultural practices and the influence of gender on healthcare seeking behaviours could have contributed to these differences. Women belonging to scheduled tribes reported to have the highest prevalence (12.3%) while the women from scheduled caste and other backward classes reported equal prevalence (10.3%) in this study. This may be due to the limited access of healthcare facilities in tribal areas.
The poorest of the sections reported a prevalence of hypertension of 9.5% as compared to the richest in whom the prevalence is reported as 13.5% which is consistent from previous studies from Bengaluru which showed a prevalence of 52% in above poverty line and 10% in below poverty line.[30] The higher prevalence of hypertension among the richest may be due high prevalence of obesity and sedentary lifestyle among them. The study reported a higher prevalence of 14.4% among illiterates as compared to prevalence among highly educated which is 8.5% as compared to a previous study from North India which showed a prevalence of 23% among illiterates and 27% among highly educated.[29] The high prevalence among the illiterates may be a reflection of the poor health seeking behaviour among them due to lack of awareness.
Among those women who consume fried food daily, prevalence of hypertension was 13% and women who consume aerated drinks daily had prevalence of 12.2%. However, 13.8% of those who never consume fried food and 12.9% of those who never take aerated drinks had hypertension. The variation in the dietary patterns across age groups could have confounded the effect which needs to be explored further. The study showed a prevalence of hypertension of 23.4% among overweight individuals which is far higher than the 6% prevalence among underweight which is in line with the findings from previous studies from Mumbai which showed a prevalence of 72.4% in obese individuals and 33.8% in people with low BMI.[31] This confirms the contribution of obesity, an established modifiable risk factor for hypertension.
A prevalence of 14.8% was reported among women who use tobacco while a prevalence of 11.0% is reported among non-tobacco users in the present study as compared to an overall prevalence of 18.8% among people using tobacco from previous studies. Datta and Husain et al.[32] further observed an association between tobacco use and uncontrolled hypertension (OR 1.10) among women aged 20–35 years. This emphasises the need for eliciting the history of tobacco use among women and counselling them towards better lifestyle. The prevalence of hypertension was found to be 19.2% among those who consume alcohol when compared to the non-drinkers (11.1%). The ICMR–INDIAB study by Bhansali A et al.[22] reported hypertensive subjects had a significantly higher prevalence of alcohol consumption, than normotensive subjects across four states. The changing lifestyles include increase in alcohol consumption among women even in developing countries. Restriction of alcohol consumption may be an important driving factor for prevention of hypertension.
This study had shown that the major predictors of hypertension among women of reproductive age to be increasing age, increase in body mass index, tobacco usage, residing in urban area, alcoholism, and non-vegetarian diet. This is similar to a study by Talukdar D et al.[23] who found age more than 35 years, current smoking, prehypertension, and central obesity to be significantly associated with incident hypertension. A systematic review of studies on risk prediction models for hypertension[14] showed that the most included predictors in the final prediction models were age, sex, body mass index, systolic blood pressure, and diastolic blood pressure, parental history of hypertension, and cigarette smoking. Similarly, a prospective cohort study by Paynter NP et al.[33] which included only women; had shown that a model based on clinical information including age, blood pressure, ethnicity, and body mass index predicted incident hypertension better than a model based on blood pressure alone. One interesting finding in this study was that the use of tobacco in any form increased the risk for hypertension among women.
Strengths and limitations
The study is done using NFHS data which is a nationally representative sample with robust estimates and is first of its kind with desegregated data for women in India which fills in a critical gap about hypertension and its determinants among them. The prediction score developed can be applied by individuals and clinicians for shared decision-making.
The limitation of the study is that the inclusion of pregnant women in the sample may perhaps lead to overestimate of the proportion with high blood pressure among females in the reproductive age group.
Future research
The use of “prediction scores” can be validated by further community-based studies. Longitudinal studies to identify the regional differences among the risk factors and their strength of association could lead to tailor-made solutions for the specific populations. The incremental increase in hypertension among the study population indicates the need for exploring innovative health promotion strategies and early screening approaches to prevent the need for pharmacological intervention in the future.
Conclusion
Hypertension is found to be a significant public health problem among women in the age group of 15–49 years with more than 1 in 10 women suffering from the disease. The major predictor variables for hypertension are age, body mass index, tobacco usage, residents of urban area, alcoholics, and non-vegetarians.
Financial support and sponsorship
Nil.
Conflicts of interest
There are no conflicts of interest.
Acknowledgements
The authors would like to acknowledge and thank Resource Center for Cardiovascular Health (RCCVH) established under Postgraduate Institute of Medical Education and Research (PGIMER), Chandigarh, for providing technical support towards writing the manuscript. We are also grateful to Global Health Advocacy Incubator (GHAI) for supporting the study (Grant number-INDIA-RIIR-20) and Demographic and Health Surveys (DHS) Program for providing the data set (survey ref no. 155509 downloaded on June 3, 2021) which helped in the development of the manuscript.
References
- 1.Ezzati M, Lopez AD, Rodgers A, Vander Hoorn S, Murray CJL. Comparative risk assessment collaborating group. Selected major risk factors and global and regional burden of disease. Lancet Lond Engl. 2002;360:1347–60. doi: 10.1016/S0140-6736(02)11403-6. [DOI] [PubMed] [Google Scholar]
- 2.Roth GA, Mensah GA, Johnson CO, Addolorato G, Ammirati E, Baddour LM, et al. Global burden of cardiovascular diseases and risk factors, 1990–2019:Update from the GBD 2019 Study. J Am Coll Cardiol. 2020;76:2982–3021. doi: 10.1016/j.jacc.2020.11.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Prabhakaran D, Singh K, Roth GA, Banerjee A, Pagidipati NJ, Huffman MD. Cardiovascular Diseases in India Compared with the United States. J Am Coll Cardiol. 2018;72:79–95. doi: 10.1016/j.jacc.2018.04.042. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Nothing for us, without us. Opportunities for meaningful engagement of people living with NCDs:Meeting report. Geneva: World Health Organization; 2021. Licence:CC BY-NC-SA 3.0 IGO. [Google Scholar]
- 5.Peters SAE, Woodward M, Jha V, Kennedy S, Norton R. Women's health:A new global agenda. BMJ Glob Health. 2016;1:e000080. doi: 10.1136/bmjgh-2016-000080. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Lamba S. Assessment of risk factors of pregnancy induced hypertension in a hospital:A case control study. Int J Clin Obstet Gynaecol. 2019;3:09–12. [Google Scholar]
- 7.Tessema KF, Gebremeskel F, Getahun F, Chufamo N, Misker D. Individual and obstetric risk factors of preeclampsia among singleton pregnancy in hospitals of Southern Ethiopia. Int J Hypertens. 2021;2021:7430827. doi: 10.1155/2021/7430827. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Kaneko H, Itoh H, Kamon T, Fujiu K, Morita K, Michihata N, et al. Association of cardiovascular health metrics with subsequent cardiovascular disease in young adults. J Am Coll Cardiol. 2020;76:2414–6. doi: 10.1016/j.jacc.2020.09.545. [DOI] [PubMed] [Google Scholar]
- 9.Non-Communicable Diseases:A priority for women's health and development. NCD Alliance 2011. [Last accessed on 2021 Jun 15]. Available from:https://www.who.int/pmnch/topics/maternal/2011_women_ncd_report.pdf.pdf .
- 10.Shukla AN, Madan T, Thakkar BM, Parmar MM, Shah KH. Prevalence and Predictors of Undiagnosed Hypertension in an Apparently Healthy Western Indian Population. Hindawi Publishing Corporation Advances in Epidemiology. 2015:5. Article ID 649184. Available from:http://dx.doi.org/10.1155/2015/649184. [Google Scholar]
- 11.Singh S, Shankar R, Singh GP. Prevalence and associated risk factors of hypertension:A cross-sectional study in Urban Varanasi. Hindawi International Journal of Hypertension Volume 2017, Article ID 5491838. :10. doi: 10.1155/2017/5491838. Available from:https://doi.org/10.1155/2017/5491838. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Hasan M, Sutradhar I, Akter T, Das Gupta R, Joshi H, Haider MR, et al. Prevalence and determinants of hypertension among adult population in Nepal:Data from Nepal Demographic and Health Survey 2016. PLoS One. 2018;13:e0198028. doi: 10.1371/journal.pone.0198028. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Tyagi R, Dhall M, Kapoor S. Bio-Social predictors of hypertension among premenopausal and postmenopausal women. SAGE Open. 2015 doi:10.1177/2158244015574227. [Google Scholar]
- 14.Echouffo-Tcheugui JB, Batty GD, Kivimäki M, Kengne AP. Risk models to predict hypertension:A systematic review. PLoS One. 2013;8:e67370. doi: 10.1371/journal.pone.0067370. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Nangia R, Thakur JS, Bhalla AK, Duseja A. Development and validation of composite risk score to assess risks of major noncommunicable diseases in Northern Indian populations:A research protocol. Int J NonCommun Dis. 2020;5:207–10. [Google Scholar]
- 16.Sathish T, Kannan S, Sarma PS, Razum O, Thrift AG, Thankappan KR. A risk score to predict hypertension in primary care settings in Rural India. Asia Pac J Public Health. 2016;28:26S–31S. doi: 10.1177/1010539515604701. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.National Family Health Survey-4, (2015-16). Ministry of Health and Family Welfare, Government of India, International Institute for Population Sciences. [Last accessed on 2021 Jul 19]. Available from:http://rchiips.org/nfhs/factsheet_nfhs-4.shtml .
- 18.Ghosh S, Kumar M. Prevalence and associated risk factors of hypertension among persons aged 15–49 in India:A cross-sectional study. BMJ Open. 2019;9:e029714. doi: 10.1136/bmjopen-2019-029714. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Agrawal S, Millett CJ, Dhillon PK, Subramanian SV, Ebrahim S. Type of vegetarian diet, obesity and diabetes in adult Indian population. Nutr J. 2014;13:89. doi: 10.1186/1475-2891-13-89. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Pavlou M, Ambler G, Seaman SR, Guttmann O, Elliott P, King M, et al. How to develop a more accurate risk prediction model when there are few events. BMJ. 2015;351:h3868. doi: 10.1136/bmj.h3868. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Agrawal V, Bhalwar R, Basannar D. Prevalence and determinants of hypertension in a rural community. Med J Armed Forces India. 2008;64:21–5. doi: 10.1016/S0377-1237(08)80139-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Bhansali A, Dhandania VK, Deepa M, Anjana RM, Joshi SR, Joshi PP, et al. Prevalence of and risk factors for hypertension in urban and rural India:The ICMR-INDIAB study. J Hum Hypertens. 2015;29:204–9. doi: 10.1038/jhh.2014.57. [DOI] [PubMed] [Google Scholar]
- 23.Talukdar D, Tripathi M, Tripathi V, Teelucksingh S. Prevalence and associated factors of undiagnosed hypertension among women aged 15–49 years in India:An analysis of National Family Health Survey-4 data. J Hum Hypertens. 2021;35:726–40. doi: 10.1038/s41371-020-0384-7. [DOI] [PubMed] [Google Scholar]
- 24.Devi P, Rao M, Sigamani A, Faruqui A, Jose M, Gupta R, et al. Prevalence, risk factors and awareness of hypertension in India:A systematic review. J Hum Hypertens. 2013;27:281–7. doi: 10.1038/jhh.2012.33. [DOI] [PubMed] [Google Scholar]
- 25.Geldsetzer P, Manne-Goehler J, Theilmann M, Davies JI, Awasthi A, Vollmer S, et al. Diabetes and Hypertension in India. JAMA Intern Med. 2018;178:363–72. doi: 10.1001/jamainternmed.2017.8094. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Youth in India-2017. Ministry of Statistics and Programme Implementation, Government of India. [Last accessed on 2021 Aug 04]. Available from:http://mospi.nic.in/sites/default/files/publication_reports/Youth_in_India-2017.pdf .
- 27.Roy A, Praveen PA, Amarchand R, Ramakrishnan L, Gupta R, Kondal D, et al. Changes in hypertension prevalence, awareness, treatment and control rates over 20 years in National Capital Region of India:Results from a repeat cross-sectional study. BMJ Open. 2017;7:e015639. doi: 10.1136/bmjopen-2016-015639. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Kaur M. Blood pressure trends and hypertension among rural and urban Jat women of Haryana, India. Coll Antropol. 2012;36:139–44. [PubMed] [Google Scholar]
- 29.Tripathy JP, Thakur JS, Jeet G, Chawla S, Jain S. Alarmingly high prevalence of hypertension and pre-hypertension in North India-results from a large cross-sectional STEPS survey. PLoS One. 2017;12:e0188619. doi: 10.1371/journal.pone.0188619. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Gowda MJ, Bhojani U, Devadasan N, Beerenahally TS. The rising burden of chronic conditions among urban poor:A three-year follow-up survey in Bengaluru, India. BMC Health Serv Res. 2015;15:330. doi: 10.1186/s12913-015-0999-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Gupta P, Gupta R, Pednekar M. Hypertension prevalence and blood pressure trends in 8⇍subjects in Mumbai, India. J Hum Hypertens. 2004;18:907–10. doi: 10.1038/sj.jhh.1001763. [DOI] [PubMed] [Google Scholar]
- 32.Datta BK, Husain MJ. Uncontrolled hypertension among tobacco-users:Women of prime childbearing age at risk in India. BMC Womens Health. 2021;21:146. doi: 10.1186/s12905-021-01280-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Paynter NP, Cook NR, Everett BM, Sesso HD, Buring JE, Ridker PM. Prediction of incident hypertension risk in women with currently normal blood pressure. Am J Med. 2009;122:464–71. doi: 10.1016/j.amjmed.2008.10.034. [DOI] [PMC free article] [PubMed] [Google Scholar]