Non-communicable diseases: epidemiology and prevention

In short

Non-communicable diseases (NCDs) are chronic, non-infectious conditions of long latency and multifactorial causation - chiefly cardiovascular disease, cancer, chronic respiratory disease and diabetes, plus mental illness, injuries and musculoskeletal, renal, oral, eye and ear disorders. They now cause roughly two-thirds of all deaths in India (WHO NCD country profile estimates ~60-66%) and strike about a decade earlier than in high-income countries, so NCD epidemiology is about measuring risk in populations and shifting whole distributions, not just treating patients. The syllabus demands the risk-factor framework, the levels and strategies of prevention, screening theory, surveillance systems and India's programme architecture (NP-NCD under NHM, Ayushman Arogya Mandirs, COTPA, FSSAI rules). It is examined in BPH/B.Sc Public Health and MPH papers and tested heavily in interviews for NHM state/district NCD cells, ICMR and NCDIR projects, PHFI, WHO-India, Tata Trusts, Apollo/Narayana population-health teams and pharma/CRO epidemiology roles.

Explained

What makes an NCD different epidemiologically

• Defining features: non-transmissible, multifactorial, long induction and latent period, insidious onset, no single necessary cause, usually incurable but controllable, high case-fatality over years rather than days.
• Web of causation replaces the single-agent model: tobacco, diet, inactivity, alcohol and air pollution act through intermediate (metabolic) factors - raised BP, raised glucose, dyslipidaemia, adiposity - to produce disease.
• Iceberg phenomenon is extreme: for every diagnosed diabetic or hypertensive there are large pools of undiagnosed disease and of prediabetes/high-normal BP; measurement therefore needs surveys, not clinic records.
• Prevalence, not incidence, dominates description because duration is long; incidence requires cohorts or registries (cancer) and is harder to obtain.
• Risk-factor classification: non-modifiable (age, sex, genetics, ethnicity, family history), behavioural (tobacco, harmful alcohol, unhealthy diet, physical inactivity), metabolic/biological (raised BP, glucose, lipids, BMI/waist), environmental (household and ambient air pollution, occupational exposure), and social determinants (income, education, urbanisation, marketing, food environment).
• Asian Indian phenotype: higher body fat and visceral fat at a lower BMI, insulin resistance, central obesity, low HDL, high triglycerides, early CAD - the reason Indian BMI and waist cut-offs are lower than WHO's.
• Developmental origins (Barker / DOHaD): low birth weight with later adiposity ("thin-fat Indian baby", Yajnik) raises adult diabetes and CVD risk - so maternal nutrition is NCD prevention.
• Dual and double burden: NCDs rise while communicable disease and undernutrition persist - India's epidemiological transition is uneven, with states at different stages (ICMR-PHFI-IHME State-level Disease Burden Initiative).

The main NCDs and their Indian epidemiology

• Cardiovascular disease (CVD): the single largest killer - ischaemic heart disease and stroke. Onset 8-10 years earlier than in Western populations; hypertension is the leading attributable risk.
• Diabetes mellitus: ICMR-INDIAB (national, published 2023) found diabetes 11.4% (~101 million) and prediabetes 15.3% (~136 million) in adults, with hypertension 35.5%, generalised obesity 28.6%, abdominal obesity 39.5% and dyslipidaemia in about 4 of 5 adults.
• Hypertension: NFHS-5 (2019-21) measured elevated BP or antihypertensive use in roughly a fifth to a quarter of adults; the rule of halves still applies - only a fraction are diagnosed, treated and controlled.
• Cancer: tracked by the ICMR-NCDIR National Cancer Registry Programme (population- and hospital-based registries). Estimated incidence was about 14.6 lakh cases in 2022 with a projected rise of roughly 12% by 2025; lifetime risk is about 1 in 9. Leading sites: breast and cervix in women; oral cavity and lung in men - tobacco-related cancers are about a quarter to a third of the burden.
• Chronic respiratory disease: COPD and asthma, driven by tobacco, biomass smoke and ambient PM2.5; India carries a disproportionate share of global COPD deaths.
• Others in the syllabus: chronic kidney disease (including CKD of unknown aetiology clusters), mental disorders (NMHS 2015-16: ~10.6% current prevalence in adults), epilepsy, road traffic injuries, blindness and cataract, deafness, oral disease, osteoarthritis, sickle cell disease (National Sickle Cell Anaemia Elimination Mission, 2023).
• Tobacco and alcohol: GATS-2 (2016-17) put adult tobacco use at 28.6%, smokeless forms predominating; NFHS-5 reported alcohol use in about a fifth of men (22.4%) and roughly 1% of women (1.3%).

Levels and strategies of prevention

• Primordial prevention: stop risk factors from ever emerging - school nutrition and activity policy, trans-fat limits, taxation, clean cooking fuel, urban design. Population-wide and the most cost-effective.
• Primary prevention: remove or reduce an established risk factor in a person free of disease - tobacco cessation, salt reduction, 150 minutes/week of moderate activity, HPV and hepatitis B vaccination (cervical and liver cancer), aspirin/statin in high-risk individuals.
• Secondary prevention: detect and treat early - population-based screening for hypertension, diabetes and three cancers; glycaemic and BP control to prevent complications.
• Tertiary prevention: limit disability and rehabilitate - cardiac rehabilitation, diabetic foot care, retinopathy screening, palliative care, stroke physiotherapy.
• Quaternary prevention: avoid over-medicalisation, overdiagnosis and unnecessary intervention.
• Rose's two strategies: the population (mass) approach shifts the whole risk distribution a little; the high-risk approach targets the tail. The prevention paradox - a measure that brings large benefit to the population offers little to each individual - explains poor individual compliance with mass measures.
• Comparison: population approach = large total gain, low cost per head, small individual benefit, no labelling; high-risk approach = large individual benefit, needs screening, medicalises, misses the majority of cases that arise in the low-risk bulk.
• Multisectoral action: health alone cannot deliver - finance (taxes), food regulation, urban development, education, transport, environment.

Screening for NCDs

• Wilson and Jungner criteria: important health problem; recognisable latent stage; natural history understood; acceptable, valid, cheap and safe test; agreed treatment available and accessible; facilities for diagnosis and treatment; agreed policy on whom to treat; continuous, not one-off; cost acceptable.
• India's population-based screening (PBS) under NP-NCD: everyone 30 years and above is enumerated by ASHAs with the Community-Based Assessment Checklist (CBAC) and screened at Ayushman Arogya Mandirs (the renamed Health and Wellness Centres) for hypertension, diabetes, oral, breast and cervical cancer, with data entered on the NCD portal/application.
• Opportunistic vs organised screening: organised programmes have a defined population, invitation, quality assurance and follow-up; opportunistic screening in OPDs misses the unattached and cannot be evaluated.
• Biases that make screening look better than it is: lead-time bias (earlier diagnosis lengthens measured survival without postponing death), length (duration) bias (slow-growing, better-prognosis cases are preferentially caught), selection/volunteer bias, and overdiagnosis of disease that would never have surfaced.
• Screening tools in Indian practice: BP by validated digital sphygmomanometer; capillary or fasting blood glucose, HbA1c where available; Indian Diabetes Risk Score (MDRF-IDRS) - age, waist, family history, physical activity, score ≥60 high risk; oral visual examination; clinical breast examination; visual inspection with acetic acid (VIA) or HPV-DNA testing for cervical cancer.
• Diagnostic cut-offs to quote: diabetes - FPG ≥126 mg/dL, 2-h OGTT ≥200 mg/dL, HbA1c ≥6.5%, or random ≥200 mg/dL with symptoms; prediabetes - IFG 110-125 mg/dL (WHO) or 100-125 (ADA), HbA1c 5.7-6.4%. Hypertension - ≥140/90 mmHg on repeated measurement in Indian and WHO guidance (the ACC/AHA 2017 threshold of 130/80 is not what Indian programmes use).
• Anthropometric cut-offs (Indian/Asian): BMI 18.5-22.9 normal, 23-24.9 overweight, ≥25 obese; waist circumference ≥90 cm (men) and ≥80 cm (women) for abdominal obesity. (Definitions of obesity are being revised internationally toward clinical and pre-clinical categories; Indian programme cut-offs remain as above.)

Surveillance and burden measurement

• WHO STEPwise approach (STEPS): Step 1 questionnaire on behaviours, Step 2 physical measurement (BP, height, weight, waist), Step 3 biochemical (glucose, lipids) - the standard for comparable risk-factor surveys.
• Indian data sources: NFHS (district-level, measures BP, blood glucose, BMI, tobacco, alcohol), National NCD Monitoring Survey (NNMS 2017-18) against the national indicator framework, GATS for tobacco, NCRP registries for cancer, SRS with Medically Certified Cause of Death and Verbal Autopsy for mortality, NSSO/NSS health consumption rounds for out-of-pocket spending, and Global Burden of Disease estimates.
• Summary measures: DALY = YLL + YLD; QALY for economic evaluation; premature mortality measured as unconditional probability of dying between 30 and 70 years from the four main NCDs - the SDG 3.4 indicator.
• Global and national targets: WHO Global Action Plan (2013-2030) with 9 voluntary targets and 25 indicators, including the "25 by 25" cut in premature NCD mortality; SDG 3.4 - one-third reduction in premature NCD mortality by 2030 plus mental-health promotion. India was the first country to adopt a National Multisectoral Action Plan for NCDs (2017-2022) with 10 national targets (e.g. 30% relative reduction in tobacco use, 30% in salt intake, 25% in raised BP, halting the rise in diabetes and obesity).
• Comparing populations requires age standardisation (direct or indirect/SMR), because crude rates reflect the age structure far more than the risk.

India's programmes, law and policy

• NP-NCD (National Programme for Prevention and Control of Non-Communicable Diseases) - the restructured and renamed NPCDCS (launched 2010), run under the National Health Mission: NCD cells at national, state and district level, district NCD clinics, CHC-level NCD clinics, cardiac care units, day-care chemotherapy centres and Ayushman Arogya Mandir-level screening with free drugs and diagnostics.
• India Hypertension Control Initiative (IHCI) - MoHFW with ICMR and WHO, using simple drug protocols, team-based care and treatment cards; the "75/25" initiative (2023) aims to put 75 million people with hypertension or diabetes on standard care.
• Comprehensive Primary Health Care (CPHC) through Ayushman Arogya Mandirs delivers 12 service packages including NCD screening and continuity of care; Ayushman Bharat PM-JAY covers hospitalisation, including cancer and cardiac care, for eligible families.
• Tobacco: the Cigarettes and Other Tobacco Products Act, 2003 (COTPA) - ban on smoking in public places (s.4), on advertising (s.5), on sale to and by minors and near educational institutions (s.6), and mandatory pictorial warnings on 85% of the principal display area (s.7). The National Tobacco Control Programme, the quitline and mCessation support quitting. The Prohibition of Electronic Cigarettes Act, 2019 bans production, sale, import and advertising of e-cigarettes. Gutkha and smokeless tobacco bans are issued by states under food safety law. A bill to raise the legal age to 21 has been discussed but the statutory minimum age remains 18.
• Health taxes: tobacco products and sugar-sweetened aerated drinks sit in the 40% GST slab introduced on 22 September 2025; the GST compensation cess ended on 1 February 2026, with tobacco duties restructured - exact rates and additional excise should be checked against the latest notification before quoting.
• Food and nutrition: FSSAI has capped industrial trans fat at 2% by mass of oils and fats (effective 1 January 2022), regulates health claims and food advertising, and runs Eat Right India; front-of-pack labelling (Indian Nutrition Rating) has been drafted but is not yet in force as a mandatory requirement. ICMR-NIN Dietary Guidelines for Indians (2024) and WHO advice set salt <5 g/day, free sugars <10% of energy and limits on fat.
• Activity and screening of other NCDs: Fit India Movement, yoga promotion, National Programme for Control of Blindness and Visual Impairment, National Programme for Prevention and Control of Deafness, National Oral Health Programme, National Programme for Health Care of the Elderly, National Tele Mental Health Programme (Tele-MANAS, 2022) under the Mental Healthcare Act, 2017, and the National Sickle Cell Anaemia Elimination Mission.
• Access to medicines: NLEM 2022 with price control by NPPA under DPCO 2013, Jan Aushadhi (PMBJP) generic outlets and AMRIT pharmacies for cancer and cardiac drugs.
• WHO "best buys": the highest-return interventions - tobacco taxation and plain/graphic packaging, smoke-free laws, alcohol tax and marketing restrictions, salt reduction, mass media campaigns, drug therapy plus counselling for high CVD risk, HPV vaccination and cervical screening, aspirin for suspected acute MI.

Key points

• NCD - chronic non-infectious disease of multifactorial cause and long latency; four main ones are CVD, cancer, chronic respiratory disease and diabetes.
• Shared modifiable risk factors - tobacco, harmful alcohol, unhealthy diet, physical inactivity, plus air pollution; they act via raised BP, glucose, lipids and adiposity.
• Premature NCD mortality - probability of dying between 30 and 70 from the four main NCDs; the SDG 3.4 indicator.
• Primordial prevention - preventing the risk factor itself from appearing in a population; cheapest and most durable.
• Rose's population approach - shift the whole distribution; most cases arise in the large low-risk majority, not the small high-risk tail.
• Prevention paradox - big population benefit, trivial individual benefit, hence poor compliance.
• Asian Indian phenotype - high visceral fat and insulin resistance at low BMI; hence BMI ≥25 = obese and waist ≥90/80 cm.
• PAF - the share of disease in a population removable by eliminating an exposure; the key policy number.
• Rule of halves - roughly half detected, half of those treated, half of those controlled; used to diagnose programme leakage.
• Wilson and Jungner criteria - the test of whether a screening programme is justified.
• Lead-time and length bias - make screened cases look as if they survive longer even when death is not postponed.
• CBAC + PBS - ASHA-administered risk checklist and screening of all adults 30+ at Ayushman Arogya Mandirs under NP-NCD.
• ICMR-INDIAB (2023) - diabetes 11.4%, prediabetes 15.3%, hypertension 35.5%, abdominal obesity 39.5% in Indian adults.
• COTPA 2003 - smoke-free public places, advertising ban, 85% pictorial warnings, no sale to minors; e-cigarettes banned by a separate 2019 Act.
• FSSAI trans-fat cap - 2% by mass of oils and fats since 1 January 2022.
• DALY - YLL + YLD; one DALY is one lost year of healthy life.
• Age standardisation - mandatory before comparing NCD rates between states, countries or time periods.
• WHO best buys - tobacco tax, salt reduction, alcohol control, CVD risk-based drug therapy, cervical cancer prevention.

Formulas and examples

1. Population attributable fraction of tobacco for lung cancer. Adult tobacco use p = 0.286 (GATS-2); RR for lung cancer in users = 10.

• p(RR − 1) = 0.286 × 9 = 2.574.
• PAF = 2.574 / (1 + 2.574) = 2.574 / 3.574 = 0.720 → 72%.
• Interpretation: about 72% of lung cancer in this population would be averted if tobacco use were eliminated. Note PAF depends on exposure prevalence, so the same RR gives a different PAF in a low-smoking state.

2. Classifying a screened adult. Male, 78 kg, 1.65 m, waist 96 cm, BP 146/92 mmHg on two readings, FPG 118 mg/dL.

• BMI = 78 / (1.65)² = 78 / 2.7225 = 28.65 kg/m².
• Indian cut-offs: ≥25 = obese (WHO international cut-off would call it only "overweight, <30").
• Waist 96 cm ≥ 90 cm → abdominal obesity.
• BP ≥140/90 on repeat → hypertension, start treatment per IHCI protocol.
• FPG 118 mg/dL → impaired fasting glucose (prediabetes), not diabetes; confirm with HbA1c/OGTT and enrol for lifestyle intervention with annual re-screening.

3. Crude versus age-standardised CVD mortality, two districts. Standard-population weights 0.50 / 0.35 / 0.15.

• ASR(A) = 0.50(20) + 0.35(300) + 0.15(1,500) = 10 + 105 + 225 = 340 per 100,000.
• ASR(B) = 0.50(20) + 0.35(400) + 0.15(2,000) = 10 + 140 + 300 = 450 per 100,000.
• Crude rates differ 3-fold (252 vs 766) but standardised rates differ only 1.3-fold: most of the gap is District B's older population, though its true age-specific risk is still higher.

4. Yield and predictive value of diabetes screening. 1,000 adults screened, true prevalence 10%, test sensitivity 85%, specificity 90%.

• Diseased 100 → TP = 85, FN = 15. Non-diseased 900 → FP = 90, TN = 810.
• PPV = 85/(85+90) = 85/175 = 48.6% - barely half of "positives" are diabetic, so confirmation is compulsory.
• NPV = 810/(810+15) = 810/825 = 98.2%.
• Yield = 85 detected per 1,000 screened = 8.5%. If prevalence fell to 2%, PPV would collapse to 17/(17+98) ≈ 14.8% - the lesson that PPV tracks prevalence.

5. ARR, RRR, NNT for a CVD prevention drug. 5-year event rate 8.0% on placebo, 6.0% on treatment.

• ARR = 0.080 − 0.060 = 0.020 = 2.0 percentage points; RRR = 0.020/0.080 = 25%; NNT = 1/0.020 = 50 people treated for 5 years to prevent one event.
• In a low-risk group with rates 1.0% vs 0.75%, RRR is the same 25% but ARR = 0.0025 and NNT = 400 - why risk-based targeting matters and why RRR alone misleads.

6. DALYs from one premature death. Man dies of myocardial infarction at 45; reference life expectancy at that age 27 years; he had 5 years of angina with disability weight 0.08.

• YLL = 27; YLD = 5 × 0.08 = 0.4; DALY = 27 + 0.4 = 27.4 DALYs.

Where it is used

• NHM state and district NCD cells: planning population-based screening, running the NCD portal, monitoring CBAC coverage, treatment initiation and control rates for hypertension and diabetes.
• Primary care at Ayushman Arogya Mandirs: Community Health Officers and ASHAs doing enumeration, CBAC risk scoring, BP and glucose measurement, oral/breast/cervical screening and follow-up of defaulters.
• Research and surveillance institutions: ICMR-NCDIR (cancer and diabetes registries), ICMR-NIE, NIN, PHFI, AIIMS CCM, state medical colleges - roles in STEPS/NNMS-type surveys, registry data management and GBD-style burden estimation.
• Corporate and insurance health: employer wellness programmes and annual health checks in IT and manufacturing; risk stratification and wellness design in insurers and PM-JAY implementation agencies.
• Hospital and NGO programmes: cardiac rehabilitation, diabetic retinopathy and foot clinics, palliative care (the Kerala community model), tobacco cessation centres, tele-mental-health through Tele-MANAS.
• Policy, regulation and advocacy: FSSAI labelling and trans-fat enforcement, COTPA enforcement squads, health-tax analysis, WHO-India and Vital Strategies/Resolve-type technical support roles.

Interview questions

• Q: What are the four main NCDs and the risk factors they share? A: Cardiovascular disease, cancer, chronic respiratory disease and diabetes; they share tobacco use, harmful alcohol, unhealthy diet, physical inactivity and air pollution, acting through raised blood pressure, glucose, lipids and adiposity.
• Q: Why are Indian BMI and waist cut-offs lower than WHO's? A: Because of the Asian Indian phenotype - more visceral fat and insulin resistance at any given BMI - so obesity is defined at BMI ≥25 kg/m² and abdominal obesity at waist ≥90 cm in men and ≥80 cm in women.
• Q: Explain the prevention paradox. A: A population-wide measure prevents many cases overall but gives each individual only a tiny benefit, so people are reluctant to comply even though the policy is highly effective.
• Q: When would you prefer a high-risk strategy over a population strategy? A: When the exposure distribution has a distinct high-risk tail, a cheap valid screening test exists and the intervention is potent - for example drug therapy for people with a 10-year CVD risk above 20% - though the population approach usually prevents more total cases.
• Q: What does a population attributable fraction tell a policymaker that a relative risk does not? A: It combines strength of association with how common the exposure is, giving the proportion of disease in that population that could actually be removed - so it ranks interventions by real-world impact.
• Q: Why do crude NCD death rates mislead when comparing two states? A: Because NCD mortality rises steeply with age, so differences mainly reflect age structure; direct age standardisation or SMR is needed for a fair comparison.
• Q: A screening test has 85% sensitivity and 90% specificity. Why is its PPV poor in a community? A: Because disease prevalence is low, so the far larger non-diseased group generates many false positives; PPV rises with prevalence while sensitivity and specificity stay fixed.
• Q: What are lead-time and length bias? A: Lead-time bias is apparently longer survival merely because diagnosis happened earlier; length bias is the preferential detection of slow-growing, better-prognosis cases by periodic screening.
• Q: What is screened under India's population-based screening and for whom? A: All adults aged 30 and above, enumerated by ASHAs using the CBAC, are screened for hypertension, diabetes and oral, breast and cervical cancer at Ayushman Arogya Mandirs under NP-NCD.
• Q: Name the key provisions of COTPA 2003. A: Prohibition of smoking in public places, of tobacco advertising and sponsorship, of sale to or by minors and within 100 yards of educational institutions, and mandatory pictorial warnings covering 85% of the principal display area.
• Q: What is the rule of halves and how do you use it? A: Roughly half of hypertensives are undetected, half of those detected are untreated and half of those treated are uncontrolled; the cascade shows where a programme loses patients - detection, initiation or retention.
• Q: What is SDG target 3.4 and how is it measured? A: A one-third reduction by 2030 in premature mortality from NCDs plus promotion of mental health, measured by the unconditional probability of dying between ages 30 and 70 from CVD, cancer, diabetes or chronic respiratory disease.

Common mistakes

• Quoting crude rates when comparing districts, states or decades, instead of age-standardised rates or SMRs.
• Confusing relative risk with attributable risk or PAF, and reporting RRR without ARR or NNT - which hides a trivial absolute benefit.
• Applying WHO BMI cut-offs (overweight ≥25, obese ≥30) to Indian data instead of the Indian/Asian cut-offs of 23 and 25.
• Treating a single high BP reading or one random glucose value as a diagnosis; screening positives need confirmation on a separate occasion by the correct test.
• Claiming a screening programme works because screened patients live longer, without accounting for lead-time, length and volunteer bias or overdiagnosis.
• Saying "NPCDCS" and "Health and Wellness Centre" in current answers - the programme is NP-NCD and the facility is the Ayushman Arogya Mandir; also mixing up the separate 2019 Act that bans e-cigarettes with COTPA 2003.
• Listing only clinical treatment as prevention and omitting primordial and primary measures - taxation, trans-fat limits, salt reduction, smoke-free laws - which carry most of the marks.

Related topics

Epidemiology: principles, measures and study designs · Biostatistics for health sciences · Demography and vital statistics · Nutrition and public health nutrition · Health policy, programmes and health systems in India · Health economics and health promotion.