Screening Flashcards
What is screening, What should we screen? and describe the potential for harm
What is screening?
- Why screen?
- What is a screening programme?
- NZ examples
When should be screen?
- is the disease appropriate?
- is th test appropriate?
- would a programme be effective?
- benefits of screening
Potential for harm
- missed cases of disease
- cost effectiveness
- over-diagnosis
What is screening?
Screening: the widespread use of a simple test for a disease in an apparently healthy (asymptomatic) population
Screening programme: an organised system using a screening test among asymptomatic people in the population to identify early cases of disease in order to improve outcomes
What is a screening test?
A test, usually relatively cheap and simple, used to test large numbers of apparently healthy people to identify individuals suspected of having early disease who will then go on to have darter diagnostic tests to confirm diagnosis. A screening test differs from a diagnostic test in that there is greater emphasis on cost and safety… and less on definitive diagnosis
- a screening test is NOT a diagnostic test
- a screening test is NOT a screening programme
Why do we try to detect early?
Aim to limit the consequences of disease through early diagnosis and treatment. Screening is an example of secondary prevention.
- preclinical phase is the time between the first biological onset to when symptoms appear
- clinical phase is from when symptoms appear to when the outcome develops
- we want to screen within the lead time, which is between when the disease is detectable and before symptoms appear.
- if they have a positive screening result then they need a diagnostic test to provide a definite diagnosis
what are the key concepts of screening?
- aims to improve outcomes, usually to reduce mortality
- screening programmes vs. case finding (opportunistic screening)
- all screening programmes do harm; some can do good as well
- screening is a pathway, not a test
what is a screening programme? what does it involve?
Distinct from case finding or opportunistic screening
- health promotion initiatives
- invitation
- screening procedure
- negative: go back to recall
- positive: progress - diagnosis
- treatment
- quality control, monitoring and evaluation
- recall
When should we screen?
Is the disease appropriate?
- seriousness of the disease
- ability to alter course of disease (is there a point where prognosis can be improved? is there an effective therapy/treatment available?)
- lead time
- prevalence of pre-clinical disease
- a sound knowledge of the natural history of the disease is important
Is the test appropriate?
- is the test accurate? (in an ideal world you would screen, then do diagnostic tests to confirm. but in the real world once you do diagnostic tests you can get false positives and false negatives)
- is the test acceptable and safe? (sensitivity and specificity)
Would a programme be effective?
- are there resources to implement and cope with positives?
- is the programme actually effective? (evidence from RCTs of benefit prior to initiation. ongoing evaluation of programmes once implimented)
- crucial to determine if screening programme actually leads to benefit
Consider benefits vs. harms of screening (including cost-benefit)
describe the four questions/considerations that come under is the disease appropriate for screening?
Seriousness of disease:
- screening is resource intensive
- so makes sense to screen for diseases with potentially severe consequences
Prevalence of pre-clinical disease:
- more efficient when high prevalence of pre-clinical disease (unless: its a serious disease or there is a cheap and easy to administer test)
Lead time: the time between the disease being detectable and the symptoms appearing. The longer the lead time, the greater the chance of detecting disease early
Ability to alter the course of the disease:
- between first biological onset and disease being detectable you can’t detect the disease
- during the lead time the screening may be of benefit
- once the symptoms appear then it is usually diagnosed anyway by a clinical so no benefit to screen
- to be able to change the course of the disease there has to be some effective therapy of treatment available. The screening has to improve the length and or quality of someones life.
what are two types of bias that can arise from screening and the ability to change the course of the disease
- lead time bias: diagnosing early will increase the perceived survival time, but not actually impact the actual length of life
- over diagnosis (diagnosing in someone who will never actually get the disease)
- which can lead to over treatment: the over diagnosis can lead to overtreatment which can have adverse effects
Describe the intrinsic test properties (specificity and sensitivity)
Can measure accuracy with sensitivity and specificity
- sensitivity: the proportion of people with the disease who test positive (true positives[a]/all the people who actually have the disease[a+c])
- specificity: proportion of people without the disease who test negative (true negatives[d]/all those without the disease [d+b])
Which is best out of sensitivity and specificity?
We want to maximise both, which can be achieved through improving the screening test and choice of disease threshold.
BUT
there is a tradeoff between sensitivity and specificity
- need to determine the cutoff point, most of the time its not obvious where that is
- reducing the cutoff leads to less specificity
How do you choose which to maximise?
- consider consequences of missing cases (false negatives) vs. fall alarms (false positives)
- sensitivity: detecting as many cars as possible important, costs or risks of next step not too hight
- specificity: costs or risks of next step high
- also consider whether we have the capacity to undertake lots of people in the test
What are the other measures of test performance?
Sensitivity and specificity: intrinsic properties of the test
- what proportion of people with or without disease the test correctly classifies
Predictive values: measure test performance in a particular population
- what proportion of people who don’t test positive/negative do/don’t have disease
- positive predictive value: portion of people who test positive and have the disease (true positives[a]/all people with positive test[a+b])
- negate predictive value: proportion of people who test negative and don’t have the disease (true negatives [d]/all those who test negative [c+d])
What are predictive values influenced by?
disease prevalence in the population of interest, unlike sensitivity or specificity
- high risk populations will have high rates of PPV and NPV
- low risk populations will have a different distribution, low PPV and high NPV
describe the resources that may need to be included for a screening test
Facilities and systems:
- manage participation
- cost and acessability
- quality control and monitoring
Treatment:
- capacity to treat true positives
Cost effectiveness:
- many people over long period
- cost vs. benefit
what are the potential benefits and harms of a screening programme?
Benefits:
- potential for early detection and intervention (rescued mortality and or morbidity, possibly less radical treatment required)
- reassurance
- improved health of population
Harms:
- increases in health inequities from unequal participation or treatments
- physical - from complications, invasive tests and/or treatments, especially if false positive, or from delayed presentation if falsely negative
- psychological: - from anxiety from eating, distress from invasive tests o procedures, knowing about serious diagnosis for longer, also negative and false positive results
- false positives - period of stress and uncertainty until diagnostic test
- financial - to individual or health service
- over diagnosis and/or treatment - may increase morbidity without reducing mortality (lead time bias)
- may diagnose a disease that would never have become apparent (length bias)
