Epi - Lecture 6 - Screening - 23/01

screening definition

• tests among asymptomatic people to identify those at an increased risk of a disease or disorder, who are offered a subsequent diagnostic test or procedure

3 types of prevention

• primary → prevent disease or injury before it occurs

  • ban of asbestos, education about healthy habits

• secondary → reduce impact of disease / injury that has already occured

  • screening

• tertiary → soften the impact of an ongoing illness or injury

  • support group, rehabilitation programme

diagnostics versus screening

• diagnostic tests are used to determine the presence or absence of a disease when a subject shows signs/symptoms of a disease

surveillance versus screening

• surveillance tends to be a long-term close, continuous observation of an individual/population → screening is a short-term cross-sectional operation (BUT can be repeated) of a population at risk.

9 screening programs in the Netherlands

• breast cancer

• cervical cancer

• colorectal cancer

• prenatal screening for infectious disease (HIV, hepatitis)

• prenatal screening for down syndrom, edwards, patau

• screening for physical abnormalities

• ultrasound scan for genetic disorders

• newborn hearing test

• newborn heel prick test

informed decision

• people need to get the information to be able to weigh all the pro’s and con’s of participation against each other

natural history of a disease

• progression of the disease over time in the absence of treatment

• need to know this for screening correctly

4 stages of natural history of disease

• stage of susceptibility

  • end of this stage is the exposure

• stage of subclinical disease

  • where screening takes place

  • here the disease is asymptomatic

  • incubation period

  • latency period

  • between stage 2 and 3 onset of symptoms

  • after this the time of diagnosis

  • beginning of subclinical phase is the pathological changes

• stage of clinical disease

• stage of recovery, disability or death

lead time

• amount of time which the diagnosis is early (time between usual diagnosis and diagnosis due to screening) → screening should lead to early diagnosis

• no lead time screening programme will not be effective but not only this is enough

• also need measurable improvements in morbidity and mortality

side effects of screening

• inconvenient

• unpleasant

• expensive (mammagraphy)

• involves some risks

• leads to difficult decisions.

explain sensitivity

• ability of a test to find those with the disease (true positives) → perfect sensitivity; everyone with positive result has the disease.

explain specificity

• ability of a test to find those without the disease (true negatives) → perfect specificity; everyone with a negative result does not have the disease

4 measures to asses validity of a test

• sensitivity → ability of a test to find those with the disease (true positives) → perfect sensitivity; everyone with positive result has the disease.

• specificity → ability of a test to find those without the disease (true negatives) → perfect specificity; everyone with a negative result does not have the disease

• positive predictive value → probability of being affected if there is a positive test result.

  • If the test result is positive, what is the probability that the person truly has the disease

• negative predictive value → probability of being not affected if there is a negative test result.

calculation sensitivity, specificity

• sensitivity = TP/ (TP + FN)

• specificity = TN/ (TN + FP)

• positive predictive value = TP/ (TP + FP)

  • PPV of 50% means that you have a 50% chance of having the disease and 50% chance of not having it

• negative predictive value = TN/ (FN + TN)

calculation PPV, and NPT

• positive predictive value = TP/ (TP + FP)

  • PPV of 50% means that you have a 50% chance of having the disease and 50% chance of not having it

• negative predictive value = TN/ (FN + TN)

the 3 biases in screening

• lead-time bias

• overdiagnosis bias

• length time bias

lead time bias

• comparing survival times gives an incorrect assumption, because with and without screening gives the same outcome (the person that was diagnosed dies). person was only diagnosed earlier

overdiagnosis bias

• error from detecting disease that would never have been found were it not for the screening test → people would normally die without ever knowing they have a certain disease, but due to screening these diseases may be detected, leading to an overdiagnosis of the disease → is negative, because treatments they receive can have side-effects, it can change courses of life → is the extreme of the length-time bias.

• error from detecting the disease that would never have been found were it not for the screening test → in the first group, the survival rate is high (1/3), while in the second group 100% dies, BUT there is the same outcome (one person dies)

length time bias

• error from detecting disease with a long latency of pre-clinical period → for aggressive diseases, there is a small window for detection without symptoms being present → for slow-growing tumors there is a longer time to detect the disease without any symptoms, therefore these are more likely to be detected as well → BUT, then screening seems to be responsible for higher survival rates, which might not be true.

• a higher proportion of indolent/less aggressive tumors (easier to treat) is found in the screened group, causing an apparent improvement in survival → appears to be more positive outcomes. → avoid bias: do an RCT with an arm with and one without screening. Ethical considerations: (a) screening addresses asymptomatic people (so you are intervening in the life of healthy people) → (b) prevalence of diseases is usually low → (c) false-positive results can lead to anxiety → (d) false negative results lead to false reassurance → (e) info about the screening is key → (f) equity to screening (people that participate are often healthier, so need to involve vulnerable groups).