FMHC QUESTIONS AND ANSWERS 2024

FMHC QUESTIONS AND ANSWERS 2024 Association Outcome occurs either significantly more often or less often in individuals who are exposed to the cause than in individuals who are not exposed ie) Near-perfect correlations exist between the death rate in Hyderabad, India, from 1911 to 1919, and the membership count of the International Association of Machinists during the same period ie)Taking daily fish oil supplements decreases the likelihood of recurrent MI. Some differences may be due to random variation (check for statistical significance, the difference muct be large enough) Risk proportion of people who are unaffected at the beginning of a study, but undergo the risk event (disease, death, injury) during the study period Causality •def: a factor produces or contributes to the production of a specific outcome ie) exposure to a carcinogen causes cancer Y = outcome/dependent variable X = factor/predictor/independent variable ε = error; unaccounted for variation in outcome Y = X1 + X2 + X3 + X4 + X1*X2 + ε MI= age + sex+ height+ fish oil+ interaction + error outcome/dependent variable The variable that is measured for change. The variable that is affected by the test (independent) variable y Predictor/factors/ independent variable Any variable that is assumed to produce an effect on, or be related to, a phenomenon of interest is a (an) _______________. x error unaccounted for variation in outcome ε types of casual relationships Sufficient Cause Necessary Cause Risk Factor Sufficient Cause if the factor is present the effect will always occur e.g. taking fish oil supplements will always reduce risk for recurrent MI Necessary Cause the factor must be present for the effect to occur, but the effect may not occur if the factor is present .e.g. fish oil may not reduce risk for recurrent MI in all subjects Risk Factor If the factor is present, the probability that effect will occur is increased e.g. taking fish oil supplements will decrease likelihood of future MI, it is not a sufficient or necessary cause Steps In determinator of Casue and Effect 1. •Is there a statistical association? 2. •Is there a temporal relationship? 3. •Are there alternative explanations? statistical association 1.Strength – difference in rates if disease between those with and without risk factor is large 2.Consistency – difference always observed if factor is present 3.Specificity – difference does not appear if factor is absent 4.Biological plausibility – association makes sense based on what is known 5.Dose-response relationship – risk is greater with stronger exposure to factor Temporal Relationships •Determine order of exposure to risk factor and outcome –Effect-cause instead of cause-effect... our thinking could be backwards –Outcome has caused the predictor X >>> Y Y >>> X Example: Type II diabetes is associated with increased risk of pancreatic cancer... could be the reverse chance •Spurious association •def: random error, no underlying explanation •Example: Drinking coffee increases the likelihood of having a MI. Alternative Explanations to Causality Chance A.No underlying explanation, random error B.Spurious association .Example 1.Statistically significant difference found in small samples (n < 20) 2.Coffee drinking and MI are not related Bias •Spurious association •def: systematic error, sources of error that distort findings in one direction Detection Bias caused by particular diagnostic technique or type of equipment changes in ability to detect cancer may be the result or failure to detect a case of disease. ie) screening in the community health clinic vs pulmonary wing of hospital Measurement Bias Observer - training, tiredness, environment, interobserver--> may be difference between the way one observer measures and the way another measures Instrument - calibration ie) measuring the height of a person with their shoes on= in which heights would be too tall Subject Bias distortion of measurement by study subject A sub set of subject bias is recall bias recall bias may occur if people who have experienced an adverse event, such as disease, are more likely to recall previous risks than people who have never experienced that event people who experience adverse event more likely to report risk factors Selection Bias not random subjects can choose which group to go into •Example: To examine the effects of coffee drinking on MI, more subjects who drink coffee than those who do not drink coffee are recruited. Allocation Bias not random subjects are assigned nonrandomly to control and experimental groups ie) interesting patents get to be seen by the residents and students in the teaching hospital where they were just supposed to be seen by attendings Synergism •def: interaction of two causal factors so that the combined effect is greater than individual effects •Example: Taking fish oil supplements decreases the risk of recurrent MI. Low-dose aspirin also decreases the risk for recurrent MI. Fish oil and aspirin taken together creates an even lower risk of recurrent MI. Confounding Effect •def: a third factor is the real cause of the outcome and the predictor is associated, but not the cause of the outcome -Confounding variable is associated with both the predictor and outcome -Mediators are on the causal path between predictor and outcome -Common confounders: age, sex, ancestry, education, genetics, social status •Example: Childhood obesity is caused by poor diet.-->could be due to socioeconomic status or genetic predisposition Effect Modification •def: the magnitude of the effect of a factor on an outcome differs depending on the level of a third variable Example: A new drug to increase HDL is developed. Tested placebo and experimental groups. No difference in HDL levels are found..... this could be due to starting HDL levels or sex. no effect in FM but we do see change in M. sex modifies the effect of the drug given A pharmaceutical company develops a new drug to reduce the size of brain tumors. Tumor size is reduced in patients given the drug, in all cases. Tumor size is not reduced in patients who do not receive the drug. Which type of causal relationship exists between the dependent and independent variables? a. Risk factor b. Necessary cause c. Sufficient cause c. Sufficient cause Sufficient cause is correct because the outcome followed the cause in all cases. While evaluating a statistical association, it is found that the association between two variables is always observed if a factor is present. This is an example of a. Biological plausibility. b. Specificity. c. Consistency. d. Strength. e. Dose-response. Consistency Consistency is the correct answer because if a difference or association is always observed if the risk factor is present. The state of Ohio changed the criteria for defining manner of death by suicide. A population health study shows that suicides rates increased four-fold over a two-year period coinciding with the changed definition. The observed association is due to which type of bias? a. Allocation b. Detection c. Selection d. Subject e. Recall detection- caused by particular diagnostic technique or type of equipment Allocation bias is incorrect because no subjects have been systematically, differently assigned to specific treatment groups. Recall bias is incorrect because the subjects have not been asked to report exposure to f actors in the past with no systematic error in the accuracy of the recollections. Selection bias is incorrect because there is no evidence that the sample is unrepresentative of the population. Occurs with non-random selection of subjects. Subject bias is incorrect because the subjects are not responding according to how they think the researchers want them to respond. A study hypothesizes that coffee drinking increases the risk of MI. The study finds that certain behavioral traits may cause people to drink more coffee; thus, increasing the risk for MI. Which alternative explains the effect of behavior on the factor and the outcome? a. Confounding effect b. Chance c. Effect modification d. Bias e. Synergism Confounding effect Confounding effect is the correct answer because certain behavioral traits may be associated with both coffee drinking (risk factor) and myocardial infarction (outcome), but it is not on the causal pathway. That is, certain behaviors do not cause MI, but they may influence if one drinks coffee. A prospective cohort study is conducted to determine the cause of falls in elderly individuals. Males and females aged 70 to 90 years old are followed over 10 years and the number of falls is recorded. Subjects are interviewed to collect additional data on house type (apartment, home, single floor, multiple floors), household composition (alone, living with partner, living with other family), mobility limitations, and history of falls. Which is the dependent variable in this study? a. Number of falls b. House type c. History of falls d. Age e. Sex Number of falls 1. A 20-year cohort study examines the effects of prophylactic statin use on the cardiovascular health of 5,387 patients aged 57+ years with a medical history remarkable for MI. Subjects are randomly allocated to non-statin, statin, and placebo groups. The diet of the patient was not considered during randomization. Results indicate the statin decreases the risk of MI occurrence. Which is the dependent variable in this study? A.Age B.MI occurrence C.Sex D.Statin use B.MI occurrence 2. A 20-year cohort study examines the effects of prophylactic statin use on the cardiovascular health of 5,387 patients aged 57+ years with a medical history remarkable for MI. Subjects are randomly allocated to non-statin, statin, and placebo groups. The diet of the patient was not considered during selection or randomization. Results indicate the statin decreases the risk of MI occurrence. Which is a possible alternative explanation for the observed relationship between statin use and MI occurrence? A.Bias B.Chance C.Confounding Effect D.Effect modification E.Synergism C.Confounding Effect 3. An open-label, randomized clinical trial examined the effect of behavioral intervention to increase physical activity in individuals with type 2 diabetes. Males and females aged 30-years or older were randomly sampled from a diabetes database. One group is given practical counseling and the other group is given general physician recommendations. Which is the dependent variable? A.Intervention strategy B.Patient sex C.Physical activity D.Physician E.Type 2 diabetes Physical Study designs Experimental Studies Can control one or multiple variables to establish cause-effect relationships Manipulate factor and observe outcome Type: Randomized controlled trial E.g., Phase I clinical trial dose-ranging for new SSRI (selective serotonin reuptake inhibitor) in treating clinical depression. Allocate subjects to groups with different concentrations of drug to determine effects. Observational Studies Hierarchy of Evidence in Clinical Research •case report •Detailed description of a single individual •Includes symptoms, signs, diagnosis, treatment, follow-up • •Considered anecdotal evidence case report advantages and disadvantages Advantages •Reflects the complexity of a whole medical condition, not a single variable--> looks at the whole individual as opposed to a randomized control trial where we look at only 1-2 variables •Identifying rare conditions and adverse effects •Educational in sharing lessons learned Limitations •Lack of sampling •Cases not generalizable •Causes or associations may have other explanations •cross-sectional study •Describes the distribution of variables at one point in time like taking a still picture frequency of diseases and frequency of riskrelated factors are accessed in the present asks, "what is happening" helps determine prevalence