Screening and Disease Prevention

Disease prevention and early detection are central goals of population health. Prevention operates at multiple levels, and screening programmes represent a major public health strategy for reducing morbidity and mortality by identifying disease before symptoms develop. This lesson covers the principles of prevention, the criteria for screening, and the interpretation of screening test performance.

Levels of Prevention. Primary prevention aims to prevent disease onset by reducing exposure to risk factors in healthy individuals. Examples include vaccination, smoking cessation programmes, folic acid supplementation in pregnancy to prevent neural tube defects, and fluoridation of water to prevent dental caries. Secondary prevention involves early detection and treatment of disease before clinical symptoms appear, typically through screening. Examples include cervical cancer screening (Pap smear, HPV test), mammography, colorectal cancer screening (colonoscopy, faecal occult blood test), and blood pressure measurement to detect hypertension. Tertiary prevention reduces the impact of established disease by minimising disability and preventing complications. Examples include cardiac rehabilitation after myocardial infarction, physiotherapy after stroke, and glycaemic control in diabetes to prevent retinopathy and nephropathy.

Wilson-Jungner Criteria for Screening (WHO, 1968). Not every disease is suitable for a screening programme. The ten classic criteria are: (1) The condition should be an important health problem; (2) The natural history of the disease should be well understood; (3) There should be a recognisable latent or early symptomatic stage; (4) There should be a suitable test or examination for the latent stage; (5) The test should be acceptable to the population; (6) Adequate treatment facilities should be available; (7) Agreed policy on whom to treat as patients; (8) The cost of case-finding should be economically balanced in relation to possible expenditure on medical care; (9) Case-finding should be a continuing process, not a once-and-for-all project; (10) There should be an effective treatment -- early detection is of no benefit unless treatment at an early stage leads to better outcomes than treatment at a symptomatic stage (lead time must translate into genuine benefit, not merely apparent longer survival).

Screening Test Performance. A 2x2 contingency table cross-classifying test results with true disease status is fundamental. True positives (TP): diseased with positive test; False positives (FP): disease-free with positive test; True negatives (TN): disease-free with negative test; False negatives (FN): diseased with negative test.

Sensitivity = TP / (TP + FN) -- the proportion of truly diseased individuals correctly identified as positive. A highly sensitive test is good for ruling out disease (if negative, disease is unlikely; SnNout mnemonic). Specificity = TN / (TN + FP) -- the proportion of truly disease-free individuals correctly identified as negative. A highly specific test is good for ruling in disease (if positive, disease is likely; SpPin). There is a fundamental sensitivity-specificity trade-off: lowering the positivity threshold increases sensitivity but decreases specificity (more FP). ROC curves (Receiver Operating Characteristic) display this trade-off graphically; the area under the curve (AUC) is a global measure of discriminatory ability (AUC = 0.5 is no better than chance; AUC = 1.0 is perfect).

Positive predictive value (PPV) = TP / (TP + FP) -- the probability that a person with a positive test result truly has the disease. Negative predictive value (NPV) = TN / (TN + FN) -- the probability that a person with a negative test result truly does not have the disease. Unlike sensitivity and specificity, PPV and NPV depend on disease prevalence (pre-test probability). A test with 99% sensitivity and 99% specificity applied to a condition with 0.1% prevalence still yields a PPV of only ~9%, meaning most positive tests are false positives -- the fundamental problem with mass screening for rare conditions.

Biases in Screening Evaluation. Lead time bias: screening advances the time of diagnosis but if treatment does not improve prognosis, survival from diagnosis will appear longer even though death occurs at the same time as without screening -- giving a false impression of benefit. Length time bias: screening programmes preferentially detect slowly progressing (less aggressive) disease, which has a longer detectable preclinical phase; these patients have better outcomes regardless of screening, inflating apparent screening benefit. Overdiagnosis: the detection of disease that would never have caused symptoms or death in the patient's lifetime; represents true harm of screening (unnecessary treatment, anxiety, cost). Volunteer bias: people who attend screening programmes tend to be healthier, more health-conscious, and of higher socioeconomic status than non-attenders, biasing programme evaluation.