What is screening?
Secondary prevention method, rolled out to try and identify people at risk of diseases before onset of symptoms.
Early diagnosis= early treatment——> better outcomes
Outline the screening process
- Screening
- Test results/ next steps:
- Positive result (doesn’t give conclusive evidence patient has disease). So undergoes further diagnostic tests. Patient then receives positive or negative results. Treatment starts- post-diagnostic phase
- Screen negative= low risk (will not completely rule out chances of having disease/ developing disease in the future)
What is the purpose of screening?
To give a better outcome compared with finding something when the patient presents with symptoms.
(If treatment can wait until there are symptoms, there is no point screening)
Give some examples of population screening programmes
Abdominal aortic aneurysm programme Bowel cancer screening Breast screening programme Diabetic eye screening programme Newborn heel prick test
What are the 5 areas of criteria for a screening programme to meet?
- Condition
- Test
- Intervention
- Screening programme
- Implementation
Describe the meaning of ‘condition’
- Must be established whether this is an important health problem (frequency and severity) with epidemiology, incidence, prevalence and natural history understood
- all cost effective primary prevention measures should have been implemented
Outline the meaning of ‘test’
- simple, safe, precise and validated screening test.
- distribution of test values in population known and agreed cut-off level defined
- acceptable to target population
- agreed policy on further diagnostic investigations on those who test positive/ choices available to them
Intervention means?
There must be effective intervention for patients identified through screening. And this intervention at a pre-symptomatic phase leads to better outcomes—- there is benefit to identifying positive individuals and providing next steps.
What must happen in relation to the screening programme?
- proven effectiveness in reducing mortality or morbidity (high quality randomised control data)
- evidence that the screening is clinically, socially and ethically acceptable to health professionals and public
- benefit gained should outweigh any harms (over diagnosis, over treatment, false positives, false reassurance, uncertain findings, complications)
- opportunity cost should be economically balanced in relation to medical expenditure as a whole
What is meant by implementation in terms of the screening programme?
How is the programme going to function? How should this be optimised?
Management and monitoring programme- quality assurance
Adequate staffing and facilities
Evidence based information available (enabling informed decision making)
Decisions should be scientifically justifiable to the public
Why should screening programmes be evaluated?
Programmes must be based on good quality evidence (should improve patient outcomes)
Can be great pressure to start screening programmes ie for prostate cancer. However, no robust evidence that earlier detection improves outcomes. Also screening could cause harm- unnecessary treatment/unwanted side effects.
Outline the 3 limitations associated with screening
- Lead time bias
- Selection bias
- Length time bias
What is meant by ‘Lead time bias’?
- early diagnosis falsely appears to prolong survival
- screened patients appear to survive longer, but only because they were diagnosed earlier
- patients live the same length of time with the disease, but longer knowing they have disease (could affect quality of life)
What is meant by ‘Length time bias’?
- screening programmes are better at picking up slow growing, unthreatening cases than aggressive, fast-growing ones (more likely to present with symptoms in between times asked to engage with screening)
- diseases that are detectable through screening are more likely to have a favourable prognosis and may never have caused a problem (problem particularly in the elderly)
- could lead to false conclusion that screening is beneficial in lengthening lives of positive cases- curing people that didn’t need curing?
What is meant by ‘Selection bias’?
Studies of screening often skewed by ‘healthy volunteer’ effect
-people engaging with service have a positive definition of health; likely do other things that protect them from disease (healthy diet, regular exercise, don’t smoke)
Randomised control trial would help deal with this bias
What are the 2 types of error likely produced by screening tests?
False positives: refers well people for further investigation that do not have the disease. (Increases anxiety/may undergo further invasive testing)
False negatives: fails to refer people who actually have an early form of the disease (inappropriate reassurance- when patients do develop symptoms may delay seeking treatment)
What are feature of test validity?
- sensitivity
- specificity
- positive predictive value
- negative predictive value
Outline the different outcomes of screening(4)
Positive test result, disease present = true positive
Positive test result, disease absent= false positive
Negative test result, disease present= false negative (false reassurance)
Negative test result, disease absent= true negative
Which features of test validity distinguish people with/without the disease in the population?
Sensitivity
Specificity
What is meant by sensitivity?
The proportion of people with the disease who test positive.
High sensitivity= screening test is good at detecting the disease
How is sensitivity calculated?
Disease present individuals:
True positives/ total individuals with disease
What is meant by specificity?
The proportion of people without the disease who test negative.
High specificity= good at ruling out who does not have the disease
How is specificity calculated?
Disease absent individuals:
True negatives/ all people who do not have the disease
If applied in the same way in different populations how will test sensitivity and specificity vary?
Values should remain the same, as they are a function of the characteristics of the test- inherent to how the test performs
