Reexam questions in Statistics and Evidence-based medicine, august 2013 2. sem. Medis/Medicin, Modul 2.4. Statistics : ESSAY-TYPE QUESTION 1. Intelligence tests are constructed such that the average score among adults is 100 points. In this example, we take a small sample of undergraduate students at a university (n = 10), and try to determine if the average of intelligence scores for all students at the university is different from 100. The scores obtained from the 10 students are as follows: 110, 110, 130, 120, 115, 118, 116, 100, 100, 117. Questions: a) What test do you choose? Show, that the data satisfy the assumptions for this test. b) Formulate the appropriate null and alternative hypotheses (as a text and in mathematical terms). c) Perform the test in SPSS. Outline the main results (the observed difference, p-value, CI interval of the difference). d) Interpret the p-value and the CI obtained. e) What conclusion can you make based on the results of the test? Max point: 21 ESSAY-TYPE QUESTION 2. What is the difference between cohort studies and experimental studies? Give an example of a situation, where an experimental study is not possible. Max point: 3 1
ESSAY-TYPE QUESTION 3. Suppose that a study of oral contraceptive (OC) use and development of bacteriuria was conducted among 14 women, all of whom were initially free from bacteriuria. At the start of the study, the women were surveyed to determine whether or not they currently used OCs. 4 were using OCs and 10 were not. By the end of the follow-up period, a total of 1 of the OC users and 1 of the nonusers had developed bacteriuria. Questions: a) Is this a cohort study or an experimental study? b) Set up the two by two table for the OC use and development of bacteriuria data. c) Calculate the risk ratio (RR) describing the strength of the relationship between OC use and bacteriuria. Interpret the result and make a conclusion. d) Calculate the odds ratio (OR) describing the strength of the relationship between OC use and bacteriuria. Interpret the result and make a conclusion. e) Calculate the 95% confidence interval for the OR in (d). Interpret the result and make a conclusion. f) Based on the results of this study, are you convinced that OC users are (are not) at higher risk of developing bacteriuria than non-users? Max point: 23 2
ESSAY-TYPE QUESTION 4. Table below shows how TV affects a teen s math grade. Questions Student ID a) Plot TV time against Math grade; Daily TV time [hours] Math grade [%] 1 5 80 2 3.2 84 3 1.2 88 4 0.5 99 5 2 88 6 2 95 7 4 80 8 4.2 82 9 4.4 82 10 2.4 90 b) Calculate the correlation coefficient between TV time and Math grade; c) Comment on the nature and strength of any association between the two variables; d) Is this association statistically significant? Max point: 11 3
ESSAY-TYPE QUESTION 5. A researcher is conducting an experiment to see if a given therapy works to reduce test anxiety in a population of medical students. In the sample the researcher draws the measurements of test anxiety are as follows (n=10): 10, 30, 20, 15, 25, 18, 22, 24, 25, 16. Questions: a) Is the distribution of measurements of test anxiety normal? (How did you check that?) b) If the distribution of test anxiety is assumed to be normal, what formula is appropriate to be used to calculate a 95% confidence interval (CI) for the population mean test anxiety? c) Using this formula, calculate a 95% CI for the population mean test anxiety and explain the result. d) Using a statistical software (e.g., SPSS), calculate a 95% CI for the population mean test anxiety and explain the result. e) Are the results in (c) and (d) similar or different? Why? Max point: 15 ESSAY-TYPE QUESTION 6. List the Bradford Hill criteria for causation and briefly describe meaning of each of them. Max point: 12 4
Evidens : EXTENDED MATCHING 7. Evidens Fejlkilder: A. Referenceprogrammer B. Kliniske retningslinjer C. Den Danske Kvalitetsmodel Opgave: Vælg fra listen af fejlkilder den, der passer bedst til hvert af de nedenstående udsagn. Er et nationalt kvalitetsudviklingssystem for sundhedsydelser Er monofaglige, ofte lokale bør være evidensbaserede Omfatter hele sundhedssektoren (både primær- og sekundær sektor) Er evidensbaserede, tværfaglige og nationale Indeholder udover kliniske retningslinjer også organisatoriske og økonomiske retningslinjer Kortlægger evidensen af ydelser inden for en række kliniske områder Max point: 4 5
ESSAY SPØRGSMÅL 8. Evidens Opgave: Besvar indenfor rammen: a) Hvorfor er det er vigtigt at basere medicinsk praksis på evidens: b) Hvordan gennemføres en evidensvurdering indenfor et fagligt område? Max point: 5 6
MULTIPLE CHOICE OPGAVE 9. Hvilket af nedenstående studiedesigns rangerer højest i evidenshierarkiet? A Metanalyser af kohorde studier B Randomiserede kontrollere forsøg (RCT) C Kohorte studier D Metaanalyser af RCT E Case-control studier Max point 2 MULTIPLE CHOICE OPGAVE 10. Hvilket af nedenstående udsagn er korrekt? A Evidensbaseret viden betyder: viden baseret på ekspertudsagn B Der findes evidensbaseret viden på højeste niveau indenfor alle medicinske fagområder C Evidensbaseret praksis betyder: at medicinsk praksis er baseret på eksperternes viden D De epidemiologisk metoderegler er et internationalt anerkendt sæt af forskningsregler, som forskere skal følge for at sikre den højest mulige grad af evidens som grundlag for deres forskningsresultater E Metodereglerne skal sikre genkendeligheden i videnskabelige forskning Max point 2 MULTIPLE CHOICE OPGAVE 11. Hvilket af nedenstående udsagn er korrekt? A Cochrane databasen indeholder kun litteratur baseret på primære studier B Systematiske reviews af case-control-studier rangerer øverst i evidenshierarkiet C Styrken og evidensniveauet af et studie afhænger af dataindsamlingsmetoden D Meta-analyser er sammenfattende statistisk analyse af resultater af tidligere udførte undersøgelser E Ved primære studier forstås studier, der sammendrager og konkluderer på baggrund af den originale forskning Max point 2 7