Key Facts

Screening aims to reduce morbidity and mortality by detecting disease at an early, treatable stage. Diagnostic tests are applied to symptomatic individuals to establish a diagnosis. Understanding test properties and their interpretation is fundamental to clinical practice.

Epidemiology

The UK National Screening Committee oversees population-based screening programmes. Approximately 11 million women are invited for breast screening, 15 million for cervical screening, and 4 million for bowel cancer screening each year. Newborn screening covers all ~700,000 UK births annually.

Aetiology

Wilson-Jungner criteria for screening programmes:

1. Important health problem
2. Accepted treatment available
3. Facilities for diagnosis and treatment available
4. Recognisable latent or early symptomatic stage
5. Suitable test available
6. Test acceptable to the population
7. Natural history understood
8. Agreed policy on who to treat
9. Cost-effective (finding cases)
10. Case-finding should be a continuous process

Pathophysiology

Test properties are determined by the 2×2 table (test result vs true disease status).

Bayes' theorem underlies the relationship between pre-test probability (prevalence), test characteristics, and post-test probability:

• Post-test odds = Pre-test odds × Likelihood ratio
• PPV and NPV are directly affected by disease prevalence in the tested population

ROC curves (Receiver Operating Characteristic) plot sensitivity vs (1−specificity) for different test thresholds — area under the curve (AUC) measures overall diagnostic accuracy (1.0 = perfect, 0.5 = no better than chance)

UK NHS Screening Programmes

• Breast cancer: Mammography every 3 years, women aged 50-70 (trial age extension 47-73)
• Cervical cancer: HPV primary screening — age 25-49 every 3 years, age 50-64 every 5 years
• Bowel cancer: FIT (faecal immunochemical test) every 2 years, age 60-74 (expanding to 50+)
• AAA: Abdominal USS for men aged 65
• Diabetic eye screening: Annual digital retinal photography for all diabetics aged ≥12
• Newborn blood spot: Day 5 — PKU, congenital hypothyroidism, sickle cell, CF, MCADD + 4 others
• Newborn hearing: Automated otoacoustic emissions (AOAE)
• Newborn and infant physical examination (NIPE): Within 72 hours and at 6-8 weeks
• Antenatal: Trisomy screening (combined test 11-14 weeks), infectious disease, anomaly scan (18-21 weeks)

Red Flags in Screening Interpretation

• False-positive results cause anxiety, further invasive investigations, and potential overtreatment
• False-negative results provide false reassurance
• Consider lead-time, length-time, and selection biases when evaluating screening programme effectiveness
• PSA screening: Not recommended as population programme due to high false-positive rate and overdiagnosis

Diagnostic Test Properties

• Sensitivity = TP/(TP+FN) × 100% — proportion of disease-positive correctly identified
• Specificity = TN/(TN+FP) × 100% — proportion of disease-negative correctly identified
• PPV = TP/(TP+FP) — probability that a positive test truly has disease
• NPV = TN/(TN+FN) — probability that a negative test truly is disease-free
• Accuracy = (TP+TN)/(TP+FP+FN+TN) — overall proportion correctly classified

Likelihood Ratios

• LR+ = Sensitivity/(1−Specificity) — how much a positive test increases probability of disease
• LR− = (1−Sensitivity)/Specificity — how much a negative test decreases probability
• LR+ >10 or LR− <0.1 are very useful tests

ROC Curve Analysis

• Plots sensitivity (y-axis) vs 1−specificity (x-axis) for different test cut-offs
• AUC (area under curve): 0.9-1.0 = excellent, 0.8-0.9 = good, 0.7-0.8 = fair, <0.7 = poor
• Optimal cut-off balances sensitivity and specificity (depends on clinical consequences of FP vs FN)

Non-pharmacological

• Informed choice: Patients must understand benefits and harms of screening before participation
• Quality assurance: All NHS screening programmes have rigorous QA standards
• Failsafe systems: Ensure no patients are lost from screening pathways

Pharmacological

• Treatment pathways following positive screening results should follow NICE guidelines
• Risk-stratified approaches emerging — e.g., lung cancer screening with LDCT for high-risk smokers

Referral Criteria

• Positive screening test → appropriate diagnostic pathway
• Breast: Triple assessment (clinical, imaging, biopsy)
• Cervical: Colposcopy for positive HPV + cytological abnormality
• Bowel: Colonoscopy for positive FIT
• AAA: Surveillance schedule based on aneurysm size (small 3-4.4cm: annual USS, medium 4.5-5.4cm: 3-monthly USS, large ≥5.5cm: consider repair)

• Breast screening: Estimated to save ~1,300 lives per year in England, but ~4,000 women may be overdiagnosed
• Cervical screening: Prevents approximately 5,000 cervical cancers per year in the UK; HPV vaccination will further reduce incidence
• Bowel screening: Reduces bowel cancer mortality by approximately 16% in screened population
• AAA screening: Reduces AAA-related mortality by approximately 42% in screened men
• The balance between benefits (earlier detection, mortality reduction) and harms (overdiagnosis, false positives, anxiety) must be continually evaluated

2×2 Contingency Table

• Sensitivity = a/(a+c)
• Specificity = d/(b+d)
• PPV = a/(a+b)
• NPV = d/(c+d)

Biases in Screening Evaluation