A community-based study was carried out in Dakshinpuri Extension, Dr. Ambedkar Nagar, a resettlement colony of South Delhi, which is the urban field practice area of Centre for Community Medicine, All India Institute of Medical Sciences since January 2002. This has a population of about 36,000 spread over ten blocks, in which there are approximately 1,500 elderly persons comprising about 4% of the population.

The study population consisted of all persons aged 60 years and above residing in this area for more than 6 months, and willing to participate. Those who were unable to communicate and a reliable informant was not available were excluded. Sample size was calculated by using the formula N = 4pq/d². Taking the prevalence of diabetes to be 13% [8], and 25% relative error, design effect of 1.5 and 10% non-response, the sample size was calculated to be 685. A cluster random sampling was done where each block was one cluster. All elderly persons in the randomly selected three blocks were recruited to the study. The information was collected by a trained investigator through house to house visits. If a person was not available after three visits, he/she was labelled as a non-responder. Data collection was carried out from Jan-June 2015. A structured questionnaire was prepared to collect information about socio-demographic variables including economic dependence. A person was considered economically independent if his/her source of personal income or any monetary benefit from a social scheme was perceived to be sufficient to maintain himself/herself. The elderly person was partially dependent if he/she had some personal income or any monetary benefit from a social scheme but that was not perceived to be sufficient to maintain himself/herself. The person was classified as economically dependent if he/she was totally dependent on other family members [9].

Self-reported diabetes and hypertension were those which have been diagnosed by a qualified person and confirmed by documentary proof.

Fasting blood glucose was measured using automated glucometer (Accu-chek glucometer) under strict aseptic precautions. An overnight fast of at least 8 hours was ensured, by asking the participant to note the time of last meal at night preceding the examination, and blood glucose was measured on empty stomach.

Blood pressure measurement was done using digital sphygmomanometer (Omron automatic blood pressure monitor, model HEM-7130) in right upper arm in the sitting position with feet kept firmly on the ground and arm kept at the level of heart. Systolic and diastolic blood pressure of two readings taken at five minutes interval was recorded. Blood pressures shows diurnal variation, so ideally the blood pressure should have been measured at the similar time of the day; however, as the survey was done through house-to house visits blood pressure was measured in different times of the day. Participants were informed a day prior to the visit for collection of fasting blood sample and were explained about the necessary precautions.

Hypertension

As per Joint National Committee VII criteria (Joint National Committee in measurement, diagnoses and evaluation of cardiovascular disorders), a systolic blood pressure of ≥140 mm Hg, and / or a diastolic blood pressure of ≥ 90 mm Hg measured on two separate occasions, with a minimum interval of at least five minutes between two measurements; or a self-reported history of taking antihypertensive medications, was defined as hypertension. An average of two readings was calculated for estimating the final blood pressure. Isolated hypertension was defined as systolic blood pressure of ≥140 mmHg [10]. Participants who were already hypertensive and taking treatment were considered to have their blood pressure controlled if their systolic blood pressure is ≤ 140 mm Hg, and/or diastolic blood pressure is ≤ 90 mm Hg.

As per WHO-IDF 2005 (World Health Organization-International Diabetes Federation) criteria, diabetes mellitus was defined as fasting plasma glucose of ≥ 126 mg/dl after an overnight fast for at least 8 hours, or a person taking treatment for diabetes mellitus [11]. Participants who were already diabetic and taking treatment were considered to be controlled if the fasting blood glucose was <126 mg/dL.

Ethical clearance from the AIIMS Ethics Committee was obtained. Informed written consent was obtained from all the subjects. Those with medical problems were referred to AIIMS for appropriate management.

Data was entered in MS Excel 2007 and analysed using STATA 11.0. Chi square was used to test the statistical significance of categorical variables. Diabetes and hypertension have multiple risk factors (independent variables) and some factors act synergistically to aggravate the disease conditions. We wanted to find out the effect of independent risk factors on the prevalence of diabetes and hypertension (dependant variables). In order to address the confounding effect of multiple risk factors, multivariate logistic regression analysis was done to measure the association between the dependent variables and independent variables.

A total of 851 elderly persons were approached for the study. Of these, 111 were not available even after three visits, and 30 refused to participate. So, a total of 710 elderly persons were recruited to the study, giving a non-response rate of 17%. Among the study participants, 413 (58.2%) were women more than half (56%) were economically partially independent and 661 (93%) were living with their families [Table/Fig-1].

Fasting blood sugar could be done for 693 participants. The prevalence of diabetes in this study was found to be 24.1% (n=167). There was no statistically significant difference between those without diabetes when analysed by age, gender, education, occupation and economic dependency [Table/Fig-2].

Blood pressure was measured for all the study participants and it was found that 477(67.2%) elderly persons were hypertensive. Hypertension was more in women (68.5%) as compared to men (65.3%), but there was no statistically significant difference by gender, age, education and economic dependency [Table/Fig-3]. Multivariate analysis was done to find out the association of hypertension with socio-economic variables; however no statistically significant difference was seen with respect to age, gender, occupation, marital status or living status. Isolated Systolic Hypertension (ISH) was seen in 25.9% of the study participants, but a subgroup analysis on Isolated Systolic Hypertension by age or gender did not show any significant association [Table/Fig-4].

Out of 167 persons with diabetes, 104 (62.3%) were on treatment and 35 (33.6%) were under control [Table/Fig-5]. Similarly, of the 477 patients detected with hypertension in this study, 197 (41.2%) were under treatment and only one-third of them had their blood pressure under control [Table/Fig-6].

This community based study was conducted among 710 elderly persons residing in an urban resettlement colony of Delhi. The overall prevalence of hypertension was 67%, which was higher than the study by Gupta et al., which reported a much lower prevalence (13%) from Delhi [8]. CURES study in urban Chennai also reported a lower prevalence (50.8%) as compared to this study [6]. Another study from an urban slum in Delhi reported a prevalence of 39% among elderly persons [12]. The reasons for these differences could be because of different time period, sample size and study settings. The prevalence of Isolated Hypertension in this study was 25.9%, which was similar to the CURES study which reported a prevalence of 25% [6]. Socio-demographic variables were not found to be statistically significantly associated in the present study, whereas a study by Ghosh et al., and Meshram et al., showed significant association between age and gender [13,14]. There was a gap in the awareness, treatment and control of hypertension in this study.

A study from an urban slum of Delhi reported similar findings, where amongst those detected with hypertension, 28.6% were aware of their condition; while only 60.7% were on treatment. Half (53%) of those on treatment had their blood pressure under control [12].

In this study, 24% of elderly persons had diabetes. This was similar to a study in Chandigarh, which reported the prevalence of diabetes to be 27.1% [7] and 23.7% from urban south India [15] and prevalence varies from 30% in Maharashtra [16] to 16.3% in Trivandrum [17]. A study from a Delhi in similar settings reported the prevalence of impaired fasting blood glucose to be 19.8% (95% CI 16.3–23.7) and was higher among women. In this study, there was no statistically significant difference between those without diabetes when analysed by age, gender, education. Similar findings were also reported from other studies in urban India [18,19]. Studies from other Asian nations such as Thailand and Hong Kong have reported lower prevalence rates of 14% and 15%, respectively [20,21]. Of the diabetics in the study, 62.3% were on treatment, while only one-third were under control. A similar study in Delhi also showed that one-third of the diabetic participants were aware of their condition; two-thirds of these were on treatment and three-fourths of those on treatment had controlled fasting blood sugar level [22] Similarly, the Indian Council of Medical Research-India Diabetes (ICMR-INDIAB) Study conducted in a representative population of three states of India (Tamil Nadu, Maharashtra, and Jharkhand) and one Union Territory (Chandigarh) observed good glycaemic control only in 31.1% of urban subjects [23].

Blood pressures shows diurnal variation, but blood pressure was measured in different times of the day. This may have introduced a measurement bias. The sample size was calculated using a relative error of 25%, which is high for a cross-sectional study. This could have decreased the precision of the study. Comorbidity in diabetes and hypertension is known but in this study we do not have data regarding those participants with co-existing diabetes and hypertension; factors associated with co-existing morbidities may be different which have not been captured in the study.

This study highlighted a significant burden of non-communicable diseases amongst elderly persons in a middle class community in Delhi. It also showed the need for services at the primary level. In addition to health system strengthening, there is a need to create awareness about NCDs through strong advocacy measures and also help in improving the health status of the elderly persons.