1 Lecture 5 of the 6 part mini med school seriesHow to pass the test Lecture 5 of the 6 part mini med school series Presenter: Sergiy Shatenko

2 Disclosures I am a medical studentThis session is not intended to give you a diagnosis or replace you seeing your health professional

3 Overview Statistics (I will try to make this as fun as possible)Screening tests Common blood tests

4 Quiz How is a diagnostic test different from a screening test?What does” Pap” stand for in Pap test? What does hemoglobin A1C measure?

5 Levels of disease preventionPrimary – health protection and prevention of disease onset (“An apple a day keeps the doctor away” lecture) Secondary prevention – early detection of disease to minimize morbidity and mortality (This lecture) Tertiary prevention – treatment and rehabilitation of disease to prevent progression and permanent disability Morbidity – the condition of being diseased

6 definitions Population – a collection of individuals who share a common trait Sample – selection of individuals from that population Sample size – contributes to the precision of the estimate Bias – trend in the collection, analysis, interpretation, publication, or review of data that can lead to conclusions that are systematically different form the truth ( Sampling bias, measurement bias, recall bias) Confounder – a variable that is related to both exposure and outcome that is not measured or is not distributed equally between groups Maybe use our study to give examples Types of samples – random, systematic, stratified, cluster, convenience Sample size reduces type I and II errors

7 Bias in screening Lead-time bias – over-estimation of survival based on earlier detection with screening (Example: Huntington’s disease) Length-time bias – overestimation of survival time due to sampling of prevalent as opposed to incident cases (Example: Lung cancer) Example: huntington disease

8 Definitions Incidence – number of new cases in a population in a period of time Prevalence – total number of cases in a population in a period of time Prevalence depends on incidence rate and duration of of disease (to cure or to death). Point prevalence vs period prevalence

9 Interpreting test resultsMedical test Example: lung cancer Disease Present Absent Test result Positive 24 14 Negative 6 56 38 62

10 Interpreting test resultsSensitivity = true positive/(true positive + false negative) = 24/30 = 80% Specificity = true negative/(true negative + false positive) =56/70 =80% Disease Present Absent Test result Positive 24 14 Negative 6 56 38 62

11 What it actually means SnOut SpINhigh sensitivity – if negative -> rule out SpIN High specificity – positive -> rule in What clinicians want to know when applying the test to their patients Sensitive test – has very few false negatives Specific test – has very few false positives

12 What patients want to knowPositive predictive value (PPV) = True positive/(True positive + False positive) = 24/(24+14) = 63% Negative predictive value (NPV) = True negative/(True negative + False negative) = 56/(56+6) = 90% PPV – if tested positive, where are they in the grey box NPV – if tested negative where are they in the light grey box Disease Present Absent Test result Positive 24 14 Negative 6 56

13 PPV and NPV are dependent on prevalenceSame test if the disease is 10 times less prevalent: PPV = 11% NVP = 99.2% PPV of 11% still higher odds than random (3 in 100) Reminder – test with 80% sensitivity and 80% specificity Disease Present Absent Test result Positive 24 194 Negative 6 776

14 Gold standard test Best available testNot necessarily 100% sensitive and specific Used as a diagnostic test Sometimes not practical Ex. post-mortem histology in parkinson’s

15 Likelihood ratio (LR) Positive likelihood ratio = sensitivity/(1-specificity) (= 4 in our example) Determines whether a test usefully changes the probability that a condition exists Used in the context of pre-test probability Changes the post-test probability If LR >1, there is increased probability that the disease state exists LR = 1, does not change the probability LR <1, decreases the probability LR of the 80% 80% test in the example is 4

16 How a diagnosis is made Pre - test probability LR test 1 Post - test probability Pre - test probability LR test 2 Post - test probability Based on Bayes theorem Pre-test probability – based on incidence in age, gender Post-test probability – Not doing this explicitly

17 Example of likelihood ratio for migrainePOUND criteria Pulsating duration of 4-72 hOurs Unilateral Nausea Disabling If 4 of the 5 criteria are met, the LR for migraine is 24 if 3 are met, the LR is 3.5 if 2 or fewer are met, the LR is 0.41 Prevalence of migraine is 12%

18 Likelihood ratio for acute coronary syndrome (heart attack)Radiation to both arms LR+ 2.6, LR- 0.93 Prior coronary artery disease LR+ 2.0, LR- 0.75 ST elevation on ECG LR , LR – 0.1

19 screening Purpose of screening :Take asymptomatic individuals and divide them into high risk versus low risk group High risk group goes on to more diagnostic investigations Tries to catch the disease at an earlier stage NOT TO DIAGNOSE

20 Types of screening Mass screening – screening all members of the population for a disease (Example: Scoliosis) Selective screening – screening a specific subgroup of the population who are at risk (Example: Breast Cancer) Multiphasic screening – the use of multiple screening tests on the same occasion (Example: annual health check up) Opportunistic screening – screening of persons who come to a health practitioner for some other purpose (Example: screening for high blood pressure when a patient comes in for a flu shot) Mass screening – PKU and hypothyroidism in newborns. Scoliosis exams back in the day Selective screening – breast cancer women in BC Multiphasic – annual heath checkup Opportunistic – doing BP when a patient comes in for a flu shot

21 Criteria for screening testDisease Should be serious Natural history must be understood Must have an asymptomatic stage that can be detected by a test Early detection and intervention must result in improved outcomes Goldilocks incidence (not too high, not too low)

22 Criteria for screening testHigh specificity and sensitivity Safe, rapid, relatively inexpensive Acceptable to providers and to populations Sensitivity more important – don’t want to miss anyone

23 Criteria for screening testDiagnosis and treatment There is an available, effective, acceptable and safe treatment Early treatment should be more effective than later

24 Criteria for screening testHealth Care system Adequate capacity for reporting, follow-up, and treatment of positive screens Cost effective Sustainable program Clear policy guidelines

25 cervical cancer screeningThe Pap test (Papanicolaou Test) 1 in 156 women will develop cervical cancer in their lifetime BC Cancer Agency screening guidelines Women should be screened every 3 years (recently changed) Named after a Greek Doctor

27 Positive Pap tests BC Cervical cancer screening program report2013 – percentage of those tested that year with an abnormal result High-grade squamous intraepithelial lesion” (HSIL) – BAD. Can be used to make a Dx persistent HPV infection and a higher risk of progression to cervical cancer Low-grade squamous intraepithelial lesion (LSIL) - Transient HPV infection Atypical squamous cells, cannot exclude HSIL (ASC-H) Atypical squamous cells of undetermined significance (ASCUS) Atypical glandular cells (AGC) BC Cervical cancer screening program report

29 Post colposcopy PPV for Cervical intraepithelial neoplasia grade 3 or adenocarcinoma in situ is 4.6% Premalignant changes with high rate of progression PPV for invasive cancer 0.5% These numbers will change with the new screening guidelines CISIII or AIS – premalignant changes with a high rate of progression

30 Pelvic exams in well womenAmerican College of obstetricians and gynecologists recommend doing them yearly, however this is expert opinion only American College of physicians recommends against doing them Canadian Task Force on Preventive Health Care recommended adopting the American College of Physicians recommendation Kauffman RP, Griffin SJ, Lund JD, Tullar PE.  Current recommendations for cervical cancer screening: do they render the annual pelvic examination obsolete? Med Princ Pract. 2013;22(4):313–22. - See more at:

31 breast cancer screening1 in 9 women will develop breast cancer in their lifetime BC Cancer agency screening guidelines all women women 50 – 74: every two years 40 – 49 talk to your doctor about the benefits and limitations of screening 40 – 74 with a 1st degree relative with breast cancer: every year Younger then 40 if you have BRCA1 or BRCA2 mutation, or chest wall radiation or strong family history 75+ talk to your doctor about the benefits and limitations of screening 1st degree relative

32 What are they looking forLooking for asymmetry, architectural distortions, calcifications Benign

33 Breast cancer screeningSensitivity 87.9% Specificity 93.1%  With a prevalence of under 1% PPV of about 6.3% (2.5% in and 13.5% in 70-79) Bc breast cancer report

34 Sensitivity and specificity depends on the density, interval of screening

35 What are the next steps? Diagnostic mammogram Ultrasound Needle biopsyOpen biopsy

36 6% of those who have a positive mammogram actually have breast cancer

37 Self Breast Exams? Studies show that self-examinations don't save women's lives and that they can lead to unneeded tests, such as biopsies. The Canadian Cancer Society recommends that all women be familiar with how their breasts look and feel and to talk to their doctors about any changes Healthlink BC: https://www.healthlinkbc.ca/health-topics/hw3791

38 Colorectal cancer screening4.4% of people will get colon cancer in their lifetime BC Cancer agency screening guidelines Men and women ages should get screened using the FIT test Men and women ages with a family history or personal history of adenomas should get colonoscopy screening

40 Colon cancer screening4.8% of those screened had positive result After a colonoscopy 4% PPV for colorectal cancer 60% PPV for adenoma Data from

41 Terminology Epithelium Benign Malignant Glandular AdenomaAdenocarcinoma Transitional Transitional papilloma Transitional cell carcinoma Liver Hepatocellular carcinoma Skin Papilloma Squamous cell carcinoma Nevus Melanoma Basal cell carcinoma Bone Osteoma Osteosarcoma Fat Lipoma Liposarcoma Cartillage Chondroma Chondrosarcoma Smooth muscle Leiomyoma Leiomyosarcoma “Striped” muscle Rhabdomyoma Rhabdomyosarcoma Vessels Angioma Angiosarcoma

42 Adenomas are not malignant by definition, what’s the big deal?Vast majority of colorectal carcinomas are adenocarcinomas These are often preceded by adenomas Adenomas can be identified and removed during colonoscopy

43 Prostate cancer screening1 in 6 males will develop prostate cancer Not a population screening test BC cancer agency recommendations Digital rectal exam (DRE) should be done annually in men 50-70 Digital rectal exam should be done if obstructive urinary symptoms are present PSA (Prostate specific antigen) recommended if suspicious DRE or suspicious urinary symptoms Often grows slowly, and patients die of other causes before the disease becomes clinically significant

46 Prostate cancer screeningDRE Sensitivity 59% Specificity 94% 24% Positive predictive value PSA (4.0 ng/mL) Sensitivity 21% Specificity 91% PPV 25% (40-60% if >10 ng/mL) Next steps? Biopsy MRI Reason why PSA is not implemented as a population screening program. Sensitivity is way too low. Would be missing many cancers and giving people false sense of security

47 Movember foundation Prostate cancer Testicular cancerMen’s health and suicide prevention

48 Common medical tests and how they workHemoglobin A1c CBC Iron Thyroid

49 Hemoglobin A1C Measure of the percentage of hemoglobin that has been altered by glucose (glycosylated) Indirect measurement of 3 month average blood glucose Life span of a red blood cell is days Can now be used to diagnose diabetes (>6.5%) Advantages: provides an average, no need to fast Disadvantages: provides an average

50 Complete blood count (CBC)Most commonly ordered test Analyzed automatically

51 What does it tell you White blood cells Hemoglobin Mean cell volumeIncreased in: infection, immune reaction, drugs, bone marrow disease Decreased in: infection, drugs, bone marrow disease Hemoglobin Increased in: lung disease, heart disease, doping Decreased in: nutrient deficiency, liver/kidney disease, cancer Mean cell volume Helps differentiate the different types of anemia Platelets If low Increased bleeding risk Causes: medications, some infection, pregnancy, cancer Pathological classification. Decreased production vs increased destruction

52 Iron Ferritin – storage form of iron, a measure of iron reservesIron in the blood Serum iron Total Iron Binding Capacity Transferrin saturation Fun fact. Iron is poorly absorbed. Maximize absorption by taking it on an empty stomach, with orange juice

53 Ferritin

54 Serum iron and total iron binding capacity

55 Thyroid Thyroid stimulating hormone Pituitary is the sensorReleases TSH to regulate thyroid function

56 A clinical scenario A patient comes in complaining of fatigueStarted slowly over the past few months Patient looks pale 105 Diagnosis? Anemia

57