Screening

Question 1

What are some conditions required for screening?

Answer 1

1968 prerequisites by Junger

• important public health problem
• accepted treatment
• latent phase of disease
• natural history understood
• cost effective
• cut off point on who to treat

Question 2

What diseases are suitable for screening?

Answer 2

Must be relatively common w severe consequences
Have latent phase
Early treatment has advantage over later
Screening has net benefit

Question 3

What is lead time bias?

Answer 3

Increase in survival in screened population may just be because of earlier diagnosis (death after disease onset may not actually change)

Question 4

What is length bias?

Answer 4

Screening more effective in indolent diseases (not aggressive)
With aggressive diseases, patient would have died before screening
Hence survival following screening may just be due to less aggressive nature

Question 5

The grid for false negatives etc

Answer 5

a= true p
B= false p 
C= false n 
D = true n

Question 6

What is the PPV + NPV?

Answer 6

Positive predictive value: proportion of those who tested positive who actually have condition (a/a+b)
Negative predictive value: proportion of those who tested negative that actually don’t have condition (d/d+c)

Question 7

What does a lower PPV indicate?

Answer 7

Rare condition - not many people actually have disease and there are lots of false positives

Question 8

Sensitivity vs specificity

Answer 8

Sensitivity: proportion of those who have condition that tested positive (a/a+c)
Specificity: proportion of those who don’t have condition that tested negative (d/b+d)

Question 9

What cancer screening programmes are available in England?

Answer 9

Breast, cervical, bowel

Question 10

Describe the NHS breast screening programme

Answer 10

1988
Screening every 3 yrs for 50-70
Being extended to 47-53 in England 
Women over 70 can request it 
V high NNT (400 screened over 10yrs so 1 less woman will die)

Question 11

What is a disadvantage of breast screening?

Answer 11

20% of detected cancers are actually DCIS (benign lesions, normally not palpable)
Hence lots of false positives + unnecessary further investigation + treatment

Question 12

Benefits of breast cancer screening

Answer 12

20% reduction in risk of dying from breast cancer compared to w/o programme
~ 1300 deaths prevented per year

Question 13

What is the uptake of breast screening in UK? How can this be improved?.

Answer 13

93% from cross sectional survey of private households

• simple information provision
• our of hours appointments
• provision of transport (lower uptake in low socio economic groups)
• follow up of non attendees to address concerns + give info

Question 14

Why is there no screening for prostate cancer?

Answer 14

3 potential tests: PSA, DRE, transurethral ultrasound
2/3 of men w high PSA don’t have cancer - lots of false positives + risk of further investigations
15% w normal PSA actually have cancer (false negatives)
Natural history is poorly understood