Prevalence of hypertension, its correlates and awareness among adult tribal population of Kerala state, IndiaII Meshram, N Arlappa, N Balkrishna, KM Rao, A Laxmaiah, GNV Brahmam
Division of Community Studies, National Institute of Nutrition, Indian Council of Medical Research, Hyderabad, Andhra Pradesh, India
Correspondence Address: DOI: 10.4103/0022-3859.105444 PMID: 23298919
Background: Increasing prevalence of hypertension is a public health problem in India. Aims: To study prevalence, correlates, and awareness of hypertension among tribal adult population in Kerala. Setting and Design: A community-based, cross-sectional study was carried out in tribal areas of Kerala by adopting multistage random sampling procedure. Materials and Methods: Data was collected on socio-demographic and behavioral factors, and anthropometric measurements were carried out. Body mass index (BMI) was categorized using the classification recommended for Asians. Waist circumference ≥90 cm for men and ≥80 cm for women was used cut off for defining an abdominal obesity. Bivariate and multivariate analysis was carried out to study association of hypertension with socio-demographic variables, personal habits, and obesity. Results: A total of 4,193 adults (men 1,891, women: 2,302) of ≥20 years of age were covered. The overall prevalence of hypertension was 40% (n=1671). The prevalence of hypertension increases with increase in age among both the genders. Regression analysis showed that the risk of hypertension was significantly (P<0.001) lower among educated and among higher socio-economic status group. Sedentary activity had 1.3 times (CI=1.09-1.60) and alcohol consumption had 1.4 (CI=1.17-1.73) times higher risk of hypertension. The risk of hypertension was 1.7 times higher among overweight/obese subjects. Overall, only 10% (n=164) of the adult population was aware of hypertension status, and about 8% (n=129) were on regular treatment. Conclusion: It was observed that the prevalence of hypertension was higher among tribal adult population of Kerala and was associated with age, gender, education, HHs wealth index, physical inactivity, alcohol consumption, and overweight/obesity.
Keywords: Hypertension, obesity, prevalence, risk behaviors, risk factors
High blood pressure (BP) is a major public health problem in developing countries around the world and is one of the most important modifiable risk factor for cardiovascular diseases (CVDs). An estimated 17.3 million people died from CVDs in 2008, representing 30% of all global deaths, over 80% of CVD deaths take place in low- and middle-income countries and occur almost equally in men and women.  It has been predicted that by 2020, there would be a 111% increase in cardiovascular deaths in India.  As reported by World Health Organization,  hypertension is the third 'killer' disease, accounting for 1 in every 8 deaths worldwide. Analysis showed that about 26% of population globally is suffering from hypertension, and the prevalence is higher among developed as compared to developing countries.  It was predicted that the number of adults with hypertension will increase by about 60% to a total of 1.56 billion (1.54-1.58 billion) by 2025.  Study carried out by Indian Council of Medical Research (ICMR) among Indian population during 1994 observed that the prevalence of hypertension was 29% and 25% among urban and 13% and 10% among rural men and women, respectively.  Hypertension is directly responsible for 57% of all stroke deaths and 24% of all coronary heart disease (CHD) deaths in India.
Demographic and health survey carried out in South Africa during 1996 among urban and non-urban population showed that the risk of hypertension was higher for urban, white, and colored Africans as compared to rural black Africans.  Another study showed prevalence of hypertension was about 35% among United South and Eastern tribal population.  Study carried out among elderly tribal in Orissa observed high prevalence of hypertension (32% and 42% among men and women),  while another study among Lepchas of the Sikkim Himalayas also observed high prevalence of hypertension.  Recent study by Sachdev among tribal population of Rajasthan showed 16% to 30% prevalence of hypertension among different tribes. 
The increasing prevalence of hypertension is attributable to rapid transition of life style practices in developing countries including India,  as well as increased elderly population due to an increase in life expectancy.
Tribal are characterized by "geographical isolation, economic backwardness, depending mostly on hunting and to some extent agriculture, a distinctive language and religion". Tribal constitute about 1.14% of total population of Kerala as per 2001 census. 
Very few studies have been carried out on prevalence of hypertension among tribal population with sufficient sample size. Therefore, the present study was undertaken with the objective to assess prevalence and determinants of hypertension among adult tribal population in Kerala. Study carried out by the National Nutrition Monitoring Bureau (NNMB) during 2007-08 on diet and nutritional status of tribal population and prevalence of hypertension among tribal adults -second repeat survey was used for the present communication. Although repeat survey, blood pressure measurements were carried out for the first time.
The ethical clearance was obtained from Institutional Ethical Committee (IEC), National Institute of Nutrition (NIN), Hyderabad. Written informed consent was obtained from the subjects involved in the study.
Study design and period
It was a community based cross-sectional study, carried out in Integrated Tribal Development Agency (ITDA) areas of Kerala State, India. The study was carried out during March 2007 to August 2008.
Sample size and sampling strategy
Sample size was calculated based on the prevalence of hypertension in rural areas as no systematic studies available on prevalence of hypertension in tribal areas. Considering prevalence of hypertension as 10%,  with 95% confidence interval (CI), and 20% relative precision, with design effect of 2, sample required was 1728≈1800 for each gender. Out of 5,582 adults enlisted in the selected HHs, 4193 were covered for measurement of blood pressure with response rate of 75%. All the individuals of 20 years of age and above and willing to give an information were included in the study, while those less than 20 years of age, not willing to participate, and pregnant and lactating women were excluded.
Selection of villages
A total of 120 villages and 40 households (HHs) from each selected villages were covered. In first stage, villages were selected, and HHs were selected in second stage. Of the 120 villages, 90 were selected from that were covered at baseline (1985-87)  and from first repeat survey (1998-99),  and the remaining 30 villages were selected randomly afresh from the list of villages provided by ITDA by adopting random sampling. Detailed methodology is described in the report by National Nutrition Monitoring Bureau, Report of the tribal surveys, 1985-87 and 1998-99. , All the tribal HHs in each selected villages were enumerated approximately, and the first HH was selected randomly using random number table, starting from the north-east corner of the village. Starting from the first HH, 40 consecutive tribal HHs were covered.
Collection of data
All the field investigators were trained and standardized in the methodology for a period of 3 weeks at NIN. Supervision was carried out by the scientists to ensure quality of data periodically. Pre-testing and standardization of interview schedules was carried out prior to the main survey.
Data was collected from the subjects on pre-tested proforma on various socio-economic and demographic variables by questionnaire method in all the selected HHs, and diet survey was carried out by 24 hr recall method of diet survey in every 4 th HHs selected for anthropometry.  Consumption of salt (g) per day per person was estimated from each individual dietary intake. Based on the consumption level of salt per day, individuals were categorized into 2 groups (≤5 and >5 g/day).
Clinical examination and examinations
Anthropometric measurements such as weight (kg), height, waist, and hip circumference (cm) were carried out on all the adult subjects available at the time of survey and during re-visits to the same households on next day to avoid selection bias with standard equipment and procedure.  Weight was measured in standing position with SECA weighing scale with minimal clothing and without shoes with accuracy of 100 g, while height was measured with anthropometer rod with an accuracy of 0.1 cm. The weighing machine was checked with known weights each day before starting the survey. Waist circumference (WC) was measured at a point midway between lower rib margin and iliac crest. BP was recorded using Diamond Regular mercury sphygmomanometer BP Apparatus (IS: 3390/CL/L-0196043)  on all the available individuals present at the time of survey and in revisits. Three measurements of BP at 5 minute interval in recumbent position were taken, and average of three readings was used for classification of hypertension, although sitting position is preferred for BP measurements, as the facilities (table, chair etc.) are not available in tribal areas. Before an initiation of the survey, the correctness of the instrument was checked by measuring BP of the field staff and was cross-checked by other similar instrument for validation. Individuals diagnosed with hypertension were given referral slip to consult Medical Officer, primary health centre (PHC) for further management. The subjects were also advised to restrict excess salt intake in the diet, cessation from smoking and abstinence from alcohol. Physical activity was assessed in relation to occupation. Those engaged as housewives, landlords, service, business, and pensioners were classified as sedentary, while laborers, owner cultivators, artisans etc. were classified as moderately active. 
Definition and diagnostic criteria
Individual with systolic BP (SBP) ≥140 mmHg, and/or diastolic BP (DBP) ≥90 mm, and or on anti-hypertensive treatment were categorized as hypertensive. Hypertension was also classified as per Joint National Committee (JNC 7) criteria, which excludes subjects already on anti-hypertensive treatment.  BP measurements were available for 4,193 subjects.
Body mass index (BMI) was calculated as weight (kg)/ht.(mt) 2 . Individuals with BMI of <18.5 was classified as 'chronic energy deficiency' (CED), BMI between 18.5-22.99 as 'normal' and ≥23-27.49 as 'overweight' and BMI ≥27.50 as 'obese'. 
Individuals with WC of ≥90 cm for men and ≥80 cm for women were considered cut off points for defining an abdominal obesity. 
Awareness about hypertensive status was based on the subjects report of a prior diagnosis of hypertension (or high BP) made by a health professional. Current use of prescription medication for lowering an elevated BP among hypertensive subjects in our sample was considered as treatment of hypertension; we considered only pharmacological treatment. 
Smokers, alcohol consumers or tobacco chewers were defined as those who ever smoked, consume alcohol or chew tobacco in their lifetime (present as well as in the past).
HH wealth was assessed by an index generated by principal component analysis. ,
Data was analyzed by using SPSS (Window Version 15.0). Prevalence of hypertension was calculated by age groups and gender. Association between hypertension and socio-economic, demographic, and other variables were tested by unadjusted odds ratio (OR). Multivariate regression analysis was carried out between dependant (hypertension) and independent variables (socio-economic and demographic, personal habits and obesity). Significance was tested at 95%, P<0.05 was taken as significance.
HH wealth was assessed by an index generated by principal component analysis, which had been in use as proxy for wealth assessment. , The variables included in the factor analysis were household, socio-economic and demographic variables such as occupation, per capita income, type of house, land holdings, source of drinking water, type of fuel for cooking, electricity and sanitary latrine that are related to wealth status. Each HH asset for which information is collected was assigned a weight or factor score generated through principal components analysis. The resulting asset scores were standardized in relation to a standard normal distribution with a mean of 0 and a SD of 1. The first 2 components explained 46% of the variance in the observed set of variables. These variables weighed the heaviest (>0.5) and were in positive direction, except source of drinking water. The regression scores from the first component analysis were used to create an index that was divided into tertiles as lowest, middle, and highest.
The mean age of the study population was 43.314.9 and 42.414.4 for men and women, respectively. Of the total subjects covered, about 45% were men.
Prevalence of hypertension and association with risk factors
Of the 4,193 adults of ≥20 years of age covered, 1,641 (40%) were classified as hypertensive. As per JNC- 7, about 37% were hypertensive. An information on socio-demographic particulars and prevalence of hypertension in different SES group is given in [Table 1]. The overall prevalence of hypertension was 45% and 36% among men and women, respectively. The prevalence of hypertension increases with age from 31% and 15% among 20-29 years to 67% and 61% among ≥60 years of men and women, respectively.
As per JNC-7 classification, about 25% were having normal BP, 37% were pre-hypertensive, 24.7% in stage I hypertension, while 12.6% were in stage II hypertension.
Bivariate regression analysis showed that the risk of hypertension was significantly (P<0.001) associated with literacy status, sedentary activity, personnel habits such as use of tobacco in any form, alcohol consumption, smoking and duration of tobacco chewing among men. Similarly, among women, the risk of hypertension was significantly (P<0.001) associated with education, HHs wealth index, sedentary activity, alcohol consumption, smoking, duration of tobacco chewing and salt intake. Education was negatively associated with the risk of hypertension with lower risk among educated subjects. It was also observed that the risk of hypertension was significantly (P<0.001) higher among overweight/obese men and women as compared to normal subjects [Table 2].
To know the influence of individual factors associated with hypertension, multivariate regression analysis was carried out by including age groups, gender, HHs wealth index, education, personal habits such as use of tobacco, smoking, alcohol consumption and obesity as independent variable and hypertension as dependant variable combined for both the gender as most of the significant variables in bivariate analysis were similar. It was observed that the risk of hypertension was 2.1, (CI=1.70-2.65), 3.7 (CI=2.89-4.73) and 5.6 times (CI=4. 34 -7.27) higher among 40 - 49, 50 - 59 years and ≥60 years of age group, respectively as compared to 20 -29 years age groups. Education and wealth index were negatively associated with risk of hypertension with lower prevalence among educated and high socio-economic status groups (OR for educated 0.72, CI=0.59-0.90, OR for high SES 0.72, CI =0.56-0.91). Sedentary activity was significantly (P<0.01) associated with risk of hypertension (OR=1.3, CI=1.09-1.60). Significant association was observed between alcohol consumption and hypertension (OR 1.4, CI=1.17-1.73). Abdominal obesity and BMI ≥23 had 2-fold (OR for abdominal obesity=1.7, CI=1.24-2.33, OR for BMI ≥23=1.7, CI=1.35-2.21) increase in risk of hypertension [Table 3].
Hypertension awareness and treatment
Awareness about hypertensive status and its anti-hypertensive pharmaceutical treatment is presented in [Table 4]. Among hypertensive (n=1641), about 10% were aware about the condition and about 8% were on treatment. The awareness and treatment was more among women and among educated subjects.
The present study revealed that the prevalence of hypertension was 45% and 36% among adult tribal men and women, respectively in Kerala. Prevalence of hypertension was significantly higher among overweight/obese as compared to normal subjects. The study also revealed high prevalence of tobacco use in any form and alcohol consumption (76% and 53% among men and 47% and 1% among women, respectively). It was also observed that the level of awareness about the condition was very low (9.8%) among the hypertensive and thus treatment for hypertension.
Strength of the study lies in its adequate sample size and same investigators have collected the data as per methodology. Data on subsample of subjects on dietary intake was also collected by 24 hr recall method that does not give information on consumption of outside/junk foods consumed by an individual. Other limitations of the study are: Because of cross-sectional design, certain bias arises in the study. Most of the risk factors were obtained from the subjects as there was no other means of obtaining the information, and as such could have been subject to recall bias. However, the outcome was measured prospectively by trained investigators, and therefore, the recall bias is less likely to have affected the observed associations. Inter-reliability was assessed by periodic check-up of height, weight, WC, and BP by the technical and scientific staff of NIN.
The high prevalence of hypertension may be due to the effect of epidemiological transition and also high intake of salt in the diet. Also, the prevalence of tobacco use and alcohol consumption was higher among the study population, which may be the risk factor for high prevalence of hypertension. Study carried out among aboriginal Nicobarese tribe during 2010 revealed very high prevalence (50.5%) of hypertension  while the study carried out during 1999 among adult Kurichias, tribe of Kerala showed very less (2.7%) prevalence of hypertension.  Another study by Kusuma et al. carried out in Visakhapatnam district of Andhra Pradesh during 2004 reported a low prevalence of hypertension among a primitive tribal group and a high prevalence of hypertension in an acculturating tribal population; depicting acculturation as a cause for increased BP levels and increased prevalence of hypertension.  Another study carried out by Kusuma et al. in 2008 among tribal population of Orissa showed 25% and 13% prevalence of hypertension among men and women, respectively.  These studies also showed high prevalence of hypertension in tribal population, but the sample size was smaller as compared to the present study. Prevalence of hypertension increases with increase in age and was higher among elderly population as observed by Kerketta et al.
Multivariate regression analysis showed that age groups, gender, education, HHs wealth index, physical inactivity, alcohol consumption, abdominal obesity, and overweight/obesity (BMI ≥23) were risk factors for hypertension. Low prevalence of hypertension among literate subjects may be because of an increased awareness and health seeking behavior among them. An inverse association was observed between hypertension and socio-economic status, which is similar with the study by Grotto et al.  The prevalence of hypertension was significantly higher among overweight/obese subjects, which is similar with findings of other authors. ,, High prevalence of hypertension among those who consume alcohol was similar with the findings by Mukhopadhyay et al. Alcohol consumption is mostly associated smoking, which is also risk factor for hypertension. Sedentary nature of work was observed to be associated with risk of hypertension. A study on lifestyle intervention observed significant decline in BP level among young adults doing regular physical exercise. 
Awareness and treatment for hypertension was very low in general (9.8% and 8%, respectively), which is similar with other study.  The awareness was lower among younger age group, but increased with increase in age, an observation similar to Vimala et al.  Therefore, increasing awareness about the condition and regular BP check-up for an early diagnosis and control of hypertension should be strengthened.
High prevalence of hypertension is of great concern among the tribal population of Kerala. However, in the present study, details about physical activities were not collected, especially during leisure time, except occupational activity. An excess salt intake is one of the important risk factor, but in the present study, salt intakes were computed-based on the 24 hr diet method, which does not include outside/junk food consumption, which contain higher amount of salt. As per the existing literature, among the population, salt-sensitive and non-sensitivity also existed, and these factors have to be addressed. Currently, there is an increase in the consumption of processed foods even among tribal and rural population, and data on these aspects were not collected, which is one of the limitation.
In conclusion, prevalence of hypertension is high in tribal population of Kerala and is higher among middle-aged and elderly, among men, among illiterate, belonging to lower socio-economic group, engaged in sedentary nature of work, habit of alcohol consumption and among overweight/obese. Appropriate strategies are needed for prevention and control of hypertension such as an increasing awareness about regular check-up of BP, regular physical exercise, and abstinence from alcohol and smoking.
The authors acknowledge the financial support by Indian Council of Medical Research (ICMR). We are also thankful to our Director, Dr. Sesikeran, for his support and timely co-operation for the survey. We are also thankful to the field staff of all NNMB Kerala State Unit. We would like to thank staff of NNMB, CRL and technical staff of Division of Community Studies, NIN for their technical help. At last but not the least, we are also thankful to all the participants without whose participation, study could not be completed.
[Table 1], [Table 2], [Table 3], [Table 4]