In a cohort of mine workers in Jharkhand, the relative risk of lung cancer from dust exposure is 2 among non-smokers and 12 among smokers. The relative risks in the two strata are significantly different from each other. What does this finding indicate?

Effect modification of dust by smoking

Selection bias among the miners

Misclassification of smoking status

Epidemiological study designs and measures

38 NEET-PG practice questions on Epidemiological study designs and measures, part of Community Medicine. The first 25 are below; the first 5 with answers and explanations (tap Show answer after you have chosen), the rest in Learn with a free account.

Question 1 · Hard
A cohort study in a Kerala district follows 2,000 adults who chew tobacco and 4,000 who do not for 10 years. Oral cancer develops in 40 chewers and 20 non-chewers. Among the chewers, what proportion of oral cancer is attributable to chewing tobacco?
1. A25%
2. B50%
3. C67%
4. D75%
Show answerHide answer
Answer: D · 75%
Incidence in chewers is 40/2,000 = 20 per 1,000 and in non-chewers 20/4,000 = 5 per 1,000, so the relative risk is 4. Attributable risk is 20 - 5 = 15 per 1,000, and attributable risk per cent is 15/20 x 100 = 75%, which equals (RR - 1)/RR. It means three-quarters of oral cancer among chewers would be prevented if they stopped chewing. Comparing raw case counts (40 against 20) ignores the different group sizes and is the commonest error.
- A: 25% is the incidence in non-chewers divided by that in chewers.
- B: 50% comes from comparing raw case counts, which ignores the group sizes.
- C: 67% is the share of all cases who chewed, not the attributable fraction.
Question 2 · Medium
In a cohort of mine workers in Jharkhand, the relative risk of lung cancer from dust exposure is 2 among non-smokers and 12 among smokers. The relative risks in the two strata are significantly different from each other. What does this finding indicate?
1. ASelection bias among the miners
2. BConfounding of dust by smoking
3. CEffect modification of dust by smoking
4. DMisclassification of smoking status
Show answerHide answer
Answer: C · Effect modification of dust by smoking
Effect modification, or interaction, is present when the size of the association between an exposure and a disease differs across levels of a third factor. Here dust raises the risk far more in smokers than in non-smokers, so the stratum-specific relative risks should be reported separately rather than pooled. Confounding, in contrast, distorts the crude estimate while the stratum-specific estimates stay similar. Interaction has public health value: it identifies groups, here smoking miners, in whom removing one exposure prevents the most disease.
- A: Selection bias concerns who enters the cohort, not a difference across strata.
- B: Confounding gives similar stratum estimates that differ from the crude one.
- D: Misclassification blurs estimates; it does not create a consistent difference between strata.
Question 3 · Easy
A district surveillance officer completes a case-control study of typhoid in an urban ward, comparing 80 cases with 160 neighbourhood controls for the habit of eating cut fruit from street vendors. Which measure of association can be estimated directly from this study?
1. AAttributable risk
2. BIncidence rate
3. CRelative risk
4. DOdds ratio
Show answerHide answer
Answer: D · Odds ratio
A case-control study selects people by disease status, so the proportion of cases in the study is fixed by the investigator and the incidence of disease in the exposed and unexposed cannot be measured. The odds ratio, the odds of exposure among cases divided by the odds of exposure among controls, is therefore the measure of association. When the disease is rare and cases and controls are representative, the odds ratio is a good estimate of the relative risk. Relative risk and attributable risk both need incidence, which comes from a cohort study.
- A: Attributable risk is a difference of incidences, measured in cohort studies.
- B: Incidence needs a population followed over time, not cases chosen by disease.
- C: Relative risk needs incidence, which a case-control study cannot give.
Question 4 · Hard
In a stone-crushing area, a cohort of quarry workers contributes 4,000 person-years of follow-up with 20 new cases of pulmonary tuberculosis, while unexposed farm workers from nearby villages contribute 6,000 person-years with 15 new cases. What is the incidence rate ratio of tuberculosis in quarry workers?
1. A1.33
2. B2.0
3. C2.5 per 1,000 person-years
4. D0.5
Show answerHide answer
Answer: B · 2.0
Incidence density in quarry workers is 20/4,000 = 5 per 1,000 person-years, and in farm workers 15/6,000 = 2.5 per 1,000 person-years. The rate ratio is 5/2.5 = 2.0, so quarry workers develop TB at twice the rate. Person-time is used because workers join and leave at different times, and comparing raw case counts (20 against 15) ignores the different follow-up. Silica dust damages alveolar macrophages, which is why silicosis greatly raises the risk of TB and why such workers are a key population for active TB case finding.
- A: 1.33 compares raw case counts and ignores person-time.
- C: 2.5 per 1,000 person-years is the rate difference, not a ratio.
- D: 0.5 is the rate ratio inverted.
Question 5 · Medium
A cohort study of workers in a cement factory finds that their death rate is lower than that of the general population of the district. The factory employs only people who pass a pre-employment medical examination, and sick workers tend to leave. Which bias explains the lower death rate?
1. AHawthorne effect
2. BLead-time bias
3. CRecall bias
4. DHealthy worker effect
Show answerHide answer
Answer: D · Healthy worker effect
Employed people are healthier than the general population, which includes the sick, disabled and elderly who cannot work. Selection at hiring and the departure of sick workers both lower the observed mortality of a workforce, hiding a real hazard of the job. This healthy worker effect is a form of selection bias. It is reduced by comparing exposed workers with unexposed workers from the same or a similar industry rather than with the general population.
- A: The Hawthorne effect is a change in behaviour from being observed.
- B: Lead-time bias belongs to evaluating screening programmes.
- C: Recall bias concerns remembered exposures, not death rates.
Question 6 · Easy
In a hospital-based case-control study of congenital heart disease, mothers of affected babies recall far more medicines taken in early pregnancy than mothers of healthy babies, although prescription records show similar use in both groups. Which bias explains this?
1. AInterviewer bias
2. BRecall bias
3. CSelection bias
4. DBerkson's bias
Answer and explanation in LearnFree account
Question 7 · Medium
In a case-control study of risk factors for cataract at a district hospital, each case is matched with a control of the same sex and of age within 2 years. After the study, the investigator wants to know whether female sex is a risk factor for cataract. What can she conclude from her own data?
1. ASex cannot be assessed because it was matched
2. BSex is a confounder that must be adjusted
3. CSex is not a risk factor for cataract
4. DSex is an effect modifier of every exposure
Answer and explanation in LearnFree account
Question 8 · Medium
A single household survey in an urban ward measures depression scores and body mass index in 3,000 adults at the same visit and finds that obesity and depression are associated. What is the main limitation of this design for judging cause?
1. AIt cannot study more than one variable
2. BIt cannot show which condition came first
3. CIt cannot estimate the prevalence of either
4. DIt needs many years of follow-up
Answer and explanation in LearnFree account
Question 9 · Easy
An expert committee reviews several studies linking a pesticide to Parkinson disease in farm workers, using the Bradford Hill criteria. Which criterion must be satisfied for the association to be called causal?
1. AConsistency of association
2. BSpecificity of association
3. CDose-response relationship
4. DTemporal association
Answer and explanation in LearnFree account
Question 10 · Medium
In a case-control study of fluorosis in Rajasthan, interviewers who know which participants are cases ask cases about drinking water sources more searchingly and more often than they ask controls. Which bias results?
1. ARecall bias
2. BSelection bias
3. CBerkson's bias
4. DInterviewer bias
Answer and explanation in LearnFree account
Question 11 · Medium
In a trial of a new antimalarial at a district hospital, the investigator holds the printed random allocation list. Knowing that the next patient will get the new drug, he enrols the less sick patients and asks sicker ones to come back later. Which flaw does this describe?
1. ALack of double blinding
2. BFailure of intention-to-treat analysis
3. CInadequate sample size
4. DInadequate allocation concealment
Answer and explanation in LearnFree account
Question 12 · Easy
A paediatric oncologist at a medical college in Lucknow wants to find out whether pesticide exposure of the mother during pregnancy is linked to acute leukaemia in her child. The disease is rare, and she has one year and a small grant. Which study design suits her question best?
1. ARandomised controlled trial
2. BProspective cohort study
3. CCase-control study
4. DCross-sectional survey
Answer and explanation in LearnFree account
Question 13 · Medium
A case-control study in a tertiary hospital takes both cases and controls from inpatients. It finds a strong association between two chronic diseases, because patients who have both conditions are more likely to be admitted than those with only one. What is this bias called?
1. ARecall bias
2. BBerkson's bias
3. CNeyman bias
4. DHealthy worker effect
Answer and explanation in LearnFree account
Question 14 · Medium
In a randomised trial of weekly iron-folic acid for adolescent girls, 40 girls in the drug group stopped the tablets because of nausea, and 25 girls in the placebo group took iron from a private clinic. Under intention-to-treat analysis, how should these 65 girls be analysed?
1. AIn the group to which each was randomised
2. BExcluded from the final analysis
3. CIn the group whose treatment each actually took
4. DReplaced by newly randomised girls
Answer and explanation in LearnFree account
Question 15 · Medium
In a cohort study in an Anganwadi area, 400 exclusively breastfed infants and 600 infants who were not exclusively breastfed are followed for 6 months. Hospital admission for diarrhoea occurs in 40 and 150 infants respectively. What is the relative risk of admission among exclusively breastfed infants, and what does it mean?
1. A2.5; breastfeeding raises the risk
2. B0.4; breastfeeding is protective
3. C0.4; breastfeeding raises the risk
4. D0.15; breastfeeding is protective
Answer and explanation in LearnFree account
Question 16 · Easy
A team studying hand hygiene in a district hospital finds that nurses wash their hands far more often during the weeks when they know observers are watching them than at other times. What is this effect called?
1. APygmalion effect
2. BPlacebo effect
3. CHawthorne effect
4. DHealthy worker effect
Answer and explanation in LearnFree account
Question 17 · Medium
A case-control study at a cancer centre in Mizoram compares 150 patients with oesophageal cancer and 300 controls. Habitual drinking of very hot tea is reported by 90 cases and 75 controls. What is the odds ratio for this habit?
1. A0.22
2. B1.2
3. C4.5
4. D2.4
Answer and explanation in LearnFree account
Question 18 · Medium
A postgraduate resident plans her thesis as a prospective cohort study of smokeless tobacco and hypertension. Her guide lists four statements about this design: 1. Incidence can be calculated directly. 2. It is the best design for a rare disease. 3. Several outcomes of the one exposure can be studied. 4. Loss to follow-up can bias the results. Which statements are correct?
1. A3 and 4 only
2. B1, 3 and 4 only
3. C2 and 3 only
4. D1, 2 and 4 only
Answer and explanation in LearnFree account
Question 19 · Medium
At a district TB unit, a 2-year cohort study follows 1,000 adults with diabetes and 2,000 adults without diabetes, all free of tuberculosis at the start. Pulmonary tuberculosis develops in 30 people with diabetes and 20 without. What is the relative risk of tuberculosis in people with diabetes?
1. A3.0
2. B1.5
3. C2.0
4. D0.33
Answer and explanation in LearnFree account
Question 20 · Easy
In a trial at a tertiary hospital comparing two antihypertensive drugs, neither the patients nor the doctors who measure their blood pressure know which drug each patient receives. The statistician who analyses the data does know the allocation. What is this design called?
1. AOpen-label trial
2. BDouble-blind trial
3. CTriple-blind trial
4. DSingle-blind trial
Answer and explanation in LearnFree account
Question 21 · Medium
A hospital-based case-control study in Chennai finds that regular alcohol use is associated with lung cancer, with an odds ratio of 2.5. When the data are analysed separately for smokers and non-smokers, the odds ratio is about 1.0 in each group. What is the best explanation for the original association?
1. AEffect modification by smoking
2. BConfounding by smoking
3. CRecall bias about alcohol use
4. DBerkson's bias in hospital controls
Answer and explanation in LearnFree account
Question 22 · Hard
A cohort study of a new typhoid conjugate vaccine in a Navi Mumbai slum reports a relative risk of typhoid of 0.70 in vaccinated compared with unvaccinated children (95% confidence interval 0.52 to 0.94). Consider four statements: 1. Vaccinated children had a 30% lower risk of typhoid. 2. The association is statistically significant at the 5% level. 3. The vaccine efficacy in this study is 70%. 4. A larger study with the same result would give a narrower interval. Which statements are correct?
1. A1, 2 and 4 only
2. B1, 2 and 3 only
3. C3 and 4 only
4. D1 and 3 only
Answer and explanation in LearnFree account
Question 23 · Hard
A hospital records study finds that ICU patients given a new anti-inflammatory drug had higher mortality than those not given it. On review, consultants had given the drug mainly to the sickest patients, and severity was recorded inconsistently. Which design would best settle whether the drug itself raises mortality?
1. ARepeated cross-sectional audit
2. BLarger prospective cohort
3. CCase-control study matched for age
4. DRandomised controlled trial
Answer and explanation in LearnFree account
Question 24 · Hard
A cohort study in a small town follows 10,000 adults for 10 years: 3,000 smokers and 7,000 non-smokers. Myocardial infarction occurs in 90 smokers and 70 non-smokers. What proportion of all myocardial infarctions in the town is attributable to smoking (population attributable risk per cent)?
1. A20 per 1,000
2. B37.5%
3. C62.5%
4. D66.7%
Answer and explanation in LearnFree account
Question 25 · Medium
In an oral cancer screening programme in a coastal district, cancers found by screening have a longer survival after diagnosis than cancers found when symptoms appear. On review, the screen-detected patients were diagnosed about 2 years earlier but died at the same age as the others. What explains the longer survival?
1. AHawthorne effect
2. BLead-time bias
3. CLength-time bias
4. DVolunteer (selection) bias
Answer and explanation in LearnFree account

13 more Epidemiological study designs and measures questions

Answer them one by one, keep your score and see your weak topics in Learn. Signed-out visitors try 25 per exam; sign in with Google to see them all.

Study Community Medicine

Epidemiological study designs and measures

A cohort study in a Kerala district follows 2,000 adults who chew tobacco and 4,000 who do not for 10 years. Oral cancer develops in 40 chewers and 20 non-chewers. Among the chewers, what proportion of oral cancer is attributable to chewing tobacco?

In a cohort of mine workers in Jharkhand, the relative risk of lung cancer from dust exposure is 2 among non-smokers and 12 among smokers. The relative risks in the two strata are significantly different from each other. What does this finding indicate?

A district surveillance officer completes a case-control study of typhoid in an urban ward, comparing 80 cases with 160 neighbourhood controls for the habit of eating cut fruit from street vendors. Which measure of association can be estimated directly from this study?

A cohort study of workers in a cement factory finds that their death rate is lower than that of the general population of the district. The factory employs only people who pass a pre-employment medical examination, and sick workers tend to leave. Which bias explains the lower death rate?

In a hospital-based case-control study of congenital heart disease, mothers of affected babies recall far more medicines taken in early pregnancy than mothers of healthy babies, although prescription records show similar use in both groups. Which bias explains this?

In a case-control study of risk factors for cataract at a district hospital, each case is matched with a control of the same sex and of age within 2 years. After the study, the investigator wants to know whether female sex is a risk factor for cataract. What can she conclude from her own data?

A single household survey in an urban ward measures depression scores and body mass index in 3,000 adults at the same visit and finds that obesity and depression are associated. What is the main limitation of this design for judging cause?

An expert committee reviews several studies linking a pesticide to Parkinson disease in farm workers, using the Bradford Hill criteria. Which criterion must be satisfied for the association to be called causal?

In a case-control study of fluorosis in Rajasthan, interviewers who know which participants are cases ask cases about drinking water sources more searchingly and more often than they ask controls. Which bias results?

In a trial of a new antimalarial at a district hospital, the investigator holds the printed random allocation list. Knowing that the next patient will get the new drug, he enrols the less sick patients and asks sicker ones to come back later. Which flaw does this describe?

A paediatric oncologist at a medical college in Lucknow wants to find out whether pesticide exposure of the mother during pregnancy is linked to acute leukaemia in her child. The disease is rare, and she has one year and a small grant. Which study design suits her question best?

A case-control study in a tertiary hospital takes both cases and controls from inpatients. It finds a strong association between two chronic diseases, because patients who have both conditions are more likely to be admitted than those with only one. What is this bias called?

In a randomised trial of weekly iron-folic acid for adolescent girls, 40 girls in the drug group stopped the tablets because of nausea, and 25 girls in the placebo group took iron from a private clinic. Under intention-to-treat analysis, how should these 65 girls be analysed?

A team studying hand hygiene in a district hospital finds that nurses wash their hands far more often during the weeks when they know observers are watching them than at other times. What is this effect called?

A case-control study at a cancer centre in Mizoram compares 150 patients with oesophageal cancer and 300 controls. Habitual drinking of very hot tea is reported by 90 cases and 75 controls. What is the odds ratio for this habit?

At a district TB unit, a 2-year cohort study follows 1,000 adults with diabetes and 2,000 adults without diabetes, all free of tuberculosis at the start. Pulmonary tuberculosis develops in 30 people with diabetes and 20 without. What is the relative risk of tuberculosis in people with diabetes?

In a trial at a tertiary hospital comparing two antihypertensive drugs, neither the patients nor the doctors who measure their blood pressure know which drug each patient receives. The statistician who analyses the data does know the allocation. What is this design called?

A cohort study in a small town follows 10,000 adults for 10 years: 3,000 smokers and 7,000 non-smokers. Myocardial infarction occurs in 90 smokers and 70 non-smokers. What proportion of all myocardial infarctions in the town is attributable to smoking (population attributable risk per cent)?