Loading presentation...

Present Remotely

Send the link below via email or IM

Copy

Present to your audience

Start remote presentation

  • Invited audience members will follow you as you navigate and present
  • People invited to a presentation do not need a Prezi account
  • This link expires 10 minutes after you close the presentation
  • A maximum of 30 users can follow your presentation
  • Learn more about this feature in our knowledge base article

Do you really want to delete this prezi?

Neither you, nor the coeditors you shared it with will be able to recover it again.

DeleteCancel

Make your likes visible on Facebook?

Connect your Facebook account to Prezi and let your likes appear on your timeline.
You can change this under Settings & Account at any time.

No, thanks

Screening

No description
by

geraldine mcdarby

on 27 October 2016

Comments (0)

Please log in to add your comment.

Report abuse

Transcript of Screening

Screening
Controversies
Screening Test
Screening Programme
Dr. Geraldine McDarby
Geraldine.McDarby@hse.ie

Explain how tests may be used to screen for cases of disease
Define and calculate validity measures of screening tests
Be aware of the criteria for a screening programme
Apply the principles of screening for disease
Outline the evaluation of a screening programme (bias)

Learning Outcomes (lecture + tutorial):
means of identifying apparently healthy individuals who may be at risk of a particular condition
Prevention strategy
-aims to reduce disease
incidence
by affecting factors that cause disease
Separate the normal from the abnormal
must know distribution of variable in
population
Unimodal Distribution
normal
abnormal
(healthy)
(disease)
"normal"
decreased risk
"abnormal"
increased risk
Validity
ability to distinguish between normal and abnormal
i.e. to reflect reality
Sensitivity
Specificity

Predictive Values
ability to correctly identify those
with the disease
ability to correctly identify those
without the disease
probability of a positive test among those with the disease
probability of a negative test among those who do not have the disease
Improve outcomes
natural history of disease
preclinical phase
clinical phase
outcome
Iceberg phenomenon
Reality(best proxy)
Test
+
-
+
-
all
with
disease

without
disease

population
size
(prevalence)
T+
F+
F-
T-
a
b
c
d
a+b
c+d
a+c
b+d
Diagnostic value of the test within a specific population
all
with disease
T+
(a)
(a+c)
T-
all
without disease
(d)
(b+d)
Exercise ECG (stress test) as screening test for heart disease
10,000 asymptomatic people under the age of 50 are screened for heart disease using an exercise ECG. The screening test (Exercise ECG) identifies 1,200 people as 'positive'. On follow up angiography, 250 of these are deemed to have heart disease. Given a prevalence of heart disease of 5% in asymptomatic patients under 50, what is the sensitivity and specificity of exercise ECG in this population?
Angiography
Exercise
ECG
250
250
950
8,550
1,200
10,000
8,800
500
9,500
a
b
c
d
a+c
b+d
a+b
c+d
+
-
+
-
Sensitivity= T+/all with disease
250/
500
=50%
Specificity=T-/all without disease
8,550/
9,500
=90%
unmeasured burden
of
disease
2x2 table
True positive
a positive test result in an individual with the disease
False positive
a positive result in an individual who, on further investigation is shown NOT to have the disease
False negative
negative result in an individual who has the disease
True negative
negative result in a patient who does not have the disease
depend on prevalence of the condition in your target population
Positive Predictive Value (PPV)
percentage of true cases among those with a positive test result
T+
all positives
all negatives
all positives
(a)
(a+b)
Negative Predictive Value (NPV)
percentage of true disease free individuals among those with a negative test result
T-
(d)
all negatives
(c+d)
Positive Predictive Value=T+/all positives
250/
1,200
8,550/
8,800
=21%
Negative Predictive Value=
=97%
Screening test + Approach + Process = Screening programme
T+
-
Improved due to earlier treatment=benefit
Outcome
Overdetection/Overtreatment
greater number of cancers than would be expected based on numbers seen in unscreened populations
RCT evidence
Observational
studies
places with screening programmes show screened age bands have more cancer than expected which persists over time
Tissue studies
autopsy evidence of disease 'reservoir'
References/Resources:
UK Screening Portal http://www.screening.nhs.uk
http://www.healthknowledge.org.uk/interactive-learning/screening
Screening for breast cancer with mammogrphy. Cochrane Database Sytematic Review 2009. Getzsche P., Nelson M
Overdiagnosis in publicly organised mammography screening. BMJ 2009. Jengensen K, Getzsche P.
Using autopsy services to estimate the disease 'reservoir' for DCIS of breast: How much more breast cancer can we find? Annals of Internal Medicine 1997. Welch H, Black W.
Shapiro S (1977) Evidence on screening for breast cancer from a randomised trial. Cancer, 39 (6) : Suppl, 2772-2782.
(Marmot 2012) Independent UK Panel on Breast Cancer Screening. The Benefits and harms of Breast Cancer. Lancet 2012. Nov 17:380(9855):1778-86
Bleyer and Welch. Effect of three decades of screening mammography on breast cancer incidence. New England Journal of Medicine 2012. Nov 22;367(21):1998-2005
Baum, Michael. (personal view) Harms from Breast cancer screening outweigh benefits if death caused by treatment is included. BMJ 2013;346:f385
Labrie F, Candas B, Dupont A, et al. (1999) Screening decreases prostate cancer death: first analysis of the 1988 Quebec prospective randomised controlled trial. Prostate, 38 (2) : 83-91.
Labrie F, Candas B, Cusan L, et al. (2004) Screening decreases prostate cancer mortality: 11-year follow- up of the 1988 Quebec prospective randomised controlled trial. Prostate, 59 (3) : 311-318.
Boston School of Public Health. Screening for Disease. http://sph.bu.edu/otlt/MPH-Modules/EP/EP713_Screening/EP713_Screening.html
Health Knowledge. Screening. http://www.healthknowledge.org.uk/interactive-learning/screening
weedon-Fekjaer H, Romundstad P. Modern Mammography screening and breast cancer mortality: population study. BMJ 2014;348:g3701
+
+
-
-
Breast Cancer
Mammography
F+
F-
T-
Outcome
-Unchanged=no benefit/ ?harm (further investigation, anger, worry)
Outcome
-Outcome worsened/unchanged=no benefit/?harm
anger vs. resignation depends on pre-test understanding of the
limitations of screening programmes
Marmot (2012):
20% relative reduction in mortality in women invited for screening (UK/50-70yrs/3yr interval)
IN a population of 10,000, 43 death prevented

benefit
harm
affordability
cervical cancer
Outcome
-Correctly reassured=?benefit
Test
Test Interpretation
Administration
Programme
validity, threshold (cut off), applicability
clinical error
processes
accessibility, chosen test threshold, chosen interval
Ethics of screening
Evaluation
Social Issues
Genetic Screening
Improved outcome for 2/10 women invited for screening
prevention paradox
Not Diagnostic!
law of diminishing returns
Benefits
Harms
Demand led screening
Justice
Autonomy
vs.
Non-malfeasance
Beneficience
Screening Criteria
Wilson & Jungner 1968
UK National Screening Committee
Test
Health Service
important health problem
natural history understood
high prevalence of preclinical stage
long period between first signs and overt disease
sensitive, and specific
simple and cheap
safe and acceptable
reliable
adequate facilities for diagnosis and treatment
effective, acceptable, safe treatment available
Condition
Important health problem
Epidemiology and natural history should be adequately understood:
(detectable risk factor/marker & a latent period/early symptomatic stage
All cost-effective primary preventive interventions should have been implemented
Simple, safe, precise, validated
Known distribution in target population
Agreed cut off level
Acceptable to the target population
Agreed policy on further investigation of those with positive test result and on the choices available
Treatment
Effective treatment/ intervention available for those identified through early detection
Evidence that early treatment leads to better outcomes
Agreed evidence based policies re who should be offered treatment
Clinical management of the condition and patient outcomes should be optimised by all healthcare providers prior to programme

Programme
Evidence from high quality RCTs that screening is effective in reducing morbidity / mortality
Complete programme is clinically, socially and ethically acceptable
Benefit should outweigh the physical and psychological harm
Opportunity cost should be economically balanced in relation to expenditure on medical care as a whole
Plan for management, monitoring and quality assurance
Adequate staffing and facilities
All other options should have been considered
Evidence based information should be available to participants
Public pressure should be anticipated. Decisions should be scientifically justifiable

Disease
Test
Process
Outcome
Access target population
Administer screening test
Interpret result of the test
Communicate results of the test
Follow up testing
Treatment
Data management systems
Programme coordination, oversight, and quality assurance
Ongoing monitoring and evaluation
Structure
setting
facilities
equipment
qualification of providers of care
administration
finance
how care is provided
acceptability
completeness
competence
end points in care
Effectiveness
(5 D's-Death, Disease, Discomfort, Disability, Disatisfaction)
Observational
Case-Control
Cohort
RCT
-population level data; cannot examine mortality among screened & cannot adjust for confounding
-groups may not be comparable due to bias
systematic error
Self Selection/Volunteer Bias
NY Breast Screening Trial 1960's (RCT)
300,000 women
not offered
mammography
offered
mammography
Death rates from
other
causes
57.6/10,000
56.9/10,000
Accepted
screening
Refused
screening
42.4/10,000
85.6/10,000
Opportunistic
High Risk
Population
Length-time Bias
Lead-time Bias
Lung Cancer
100 diagnosed with lung cancer
80 of those are dead within 1 year
each year in your local population:
1 year mortality
1 year survival
80%
20%
Introduce screening
purpose is to detect early changes-cancer like tissue changes confined to the bronchial epithelium
10 years later:
150 diagnosed with lung cancer/year
80 of these are dead within 1 year
1 year mortality
1 year survival
53%
47%
prognostic selection
Motor Neurone Disease
early symptomatic phase 8yrs
diagnosis in 40's
survival 2-3yrs
with screening:
no screening
early symptomatic phase shortened
diagnosis earlier
survival appears longer
Privacy/confidentiality
Family members
Consent
Discrimination
Equity
Alcohol
Violence
What do we screen for?
Who decides & how?
Evidence
Politics
Socio-cultural issues
1,000 men aged 55-69 screened every 1-4 years for 10 years with PSA test
US Preventive Services Task Force Recommendation Statement, Annals of Internal Medicine, 2012
(reduction in breast cancer mortality)
Biological onset
of disease
Disease diagnosed
clinically
Death
survival time
Disease diagnosed
by screening
Private Screening
screening without the programme
MORTALITY
improve length +/- quality of life
Harm vs Benefit
cut off
50% chance that a person with the disease
will test positive
90% chance that a person without the disease will test negative
21% of those with a positive test result will actually have the target condition
97% of those with a negative test result will not have the target condition
T-/all negatives
Marmot(2012): 10,000 population (43 deaths prevented)
129 overdiagnosed
1:3 deaths prevented to overdiganosis
Baum (2013): old RCT's
cause specific mortality

(effectiveness of treatments)
overdiagnosis/overtreatment contributes to all cause mortality
deaths due to chemo/RXT
Bleyer and Welch (2012): 31% of all cancer (50% of those detected by screening) are overdiagnosed each year
70,000 annually
Who creates the demand, patients or PR??
who picks up the tab? for genuine as well as false positives??
may be at risk
-aims to reduce disease
prevalence
through early identification and treatment
-aims to minimise the
consequences
of disease
primary
secondary
tertiary
Weedon-Fekjaer & Romundstad (BMJ 2014)
28% reduction in mortality in women invited to attend for screening with NNT 368 to prevent 1 death (50-69y/o biennial)
“A preventive measure which brings much benefit to the population [yet] offers little to each participating individual” Rose
what effect is it likely to have on health inequalities??
Prostate cancer
Lung Cancer
CVD, Obesity, Diabetes
choice relates to consequences of Errors (F+/F-)
50-64yrs
every 2 yrs
these measures relate to the test
When interpreting a result in the real world you want to know how likely it is that a positive test means you have the disease and/or a negative result means you do not-for this you need predictive values
Sensitivity relates to the test & is the ability of the test to correctly identify cases
It is linked with the F+ rate-Improving the Sensitivity increases the F+ rate
Specificity also relates to the test & is the ability of the test to correctly identify non cases
It is linked with the F- rate-improving the specificity increases the F-rate
The predictive values help you to interpret test results. PPV tells you how likely it is that a + result means the individual has the disease while the NPV tells you how likely it is that a - result means that the individual does not have the disease
-Not improved (poor response to treatment)=no benefit/ ? harm
-Unchanged (low grade)=no benefit + exposed to risk of harm
-Unnecessary treatment=harm (latent disease)
Full transcript