Select all of the following statements which you believe to be true. Longitudinal studies:
A. Are either prospective or retrospective.
B. Are either experimental or observational.
C. Are particularly suitable for estimating the point prevalence of a condition.
D. Cannot be used to estimate the incidence of a disease.
E. Can be used for assessing causality.
Answers: A, B, E
Let’s evaluate each option:
C: Point prevalence is concerned with the number of cases of a disease existing at a point in time, and longitudinal studies are concerned with looking at features of the study over a period of time. It is a cross-sectional study which is particularly suitable for estimating the point prevalence.
D: The incidence of a disease is concerned with the number of new cases of the disease which occur in a period of time. A longitudinal study takes place over a period of time, and therefore can be used to estimate the incidence of a disease.
Select all of the following studies that are repeated cross-sectional studies.
A. The UK national census, which takes place every 10 years.
B. A natural history study of individuals infected with hepatitis C virus followed from the time of diagnosis for 5 years.
C. A study of dietary patterns of first year medical students in the first week of October carried out for five consecutive years.
D. A study to consider the incidence of AIDS events in patients infected with HIV.
Answer: A, C
Explanation:
A: This study is repeated cross-sectional as each census may study a different group of people (i.e. some people may die or leave the UK whereas others may be born or move to the UK between each census) and information on the same individuals is not linked between the censuses.
B: This study is not repeated cross-sectional as the same group of individuals is followed over time.
C: This study is repeated cross-sectional as a different random sample of medical students is chosen each year and the same students are not followed over time.
D: This study is not repeated cross-sectional as, in order to study the incidence of AIDS, the same individuals must be followed over time.
A cohort study has the following advantages:
Select all of the following statements which you believe to be true.
A. It can be used to study the exposure to factors that are rare.
B. It is relatively cheap to perform because it follows a defined group of individuals.
C. It is unusual to have losses to follow-up because it follows a defined group of individuals.
D. It requires a reasonably small sample size if the outcome is rare.
E. The time sequence of events can be assessed.
Answer: A and E
Explanation:
A: True.
B: A cohort study tends to be expensive because it has to follow the cohort of individuals over a long period of time.
C: A cohort study does have losses to follow-up as individuals may leave the study during the time period of the investigation.
D: The reverse is true in that, if the outcome is rare, a large cohort must be studied.
E: Because individuals in the study are followed longitudinally over time, it is possible to determine whether the exposure occurred before or after the event of interest.
The relative risk of a disease:
Select all of the following statements which you believe to be true.
A. Always lies between zero and one.
B. Is always positive.
C. Measures the increased (or decreased) risk of the factor when the individual has the disease.
D. Measures the risk of the disease in the population.
E. Takes the value zero when the risk is equally likely in those exposed and unexposed to the factor of interest.
Answer: B
Explanation:
A: The relative risk is always positive but can take a value which is greater than one. If the relative risk is greater than one, it is interpreted to mean that there is an increased risk of the disease if the factor is present compared to when it is absent.
B: Ture.
C: The relative risk measures the increased (or decreased) risk of the disease if the factor is present compared to when it is absent.
D: The relative risk is the ratio of two risks in the population
E: The relative risk takes the value one (unity) when the risks in those with and without the factor are equal.
A case-control study may suffer from the following disadvantages:
Select all of the following statements which you believe to be true.
A. It is not suitable for rare disease outcomes.
B. It is not suitable when the exposures to the risk factor are rare.
C. It is relatively expensive to perform.
D. It is limited to investigating only one risk factor.
E. It does not allow the direct evaluation of the relative risk.
Answer: A, B and E
Explanation:
A: A case-control study is particularly suitable for studying rare disease outcomes because the investigator starts off with diseased individuals.
B: True.
C: Case-control studies are relatively inexpensive as they rely on collecting retrospective information and do not require the investigators to follow groups of individuals forward in time.
D: Any number of risk factors (within reason) can be investigated in a case-control study.
E: We evaluate the odds ratio which is an estimate of the relative risk.
The odds ratio:
Select all of the following statements which you believe to be true.
A. Is an estimate of the relative risk when the incidence of the disease is rare.
B. Is calculated in a case-control study because the relative risk cannot be estimated directly.
C. Is equal to zero when the odds of being a case in the exposed and unexposed groups are equal.
D. Is the ratio of the probability of being a case in the exposed group to the probability of not being a case in exposed group.
E. Cannot be negative.
Answer: A and C
Explanation:
A: True.
B: It is not possible to estimate the risk of disease (and therefore the relative risk) in a case-control study as the number of subjects with and without the disease of interest is selected (often in equal numbers) as part of the design of the study.
C: The odds ratio is equal to one (unity) when the odds of being a case in the exposed and unexposed groups are equal.
D: The odds ratio is the odds of being diseased in the group exposed to the factor divided by the odds of being diseased in those unexposed to the factor. The odds of being diseased in the exposed group is the probability of being diseased in those exposed to the factor divided by the probability of not being diseased in those exposed to the factor. The odds of being diseased in those not exposed to the factor is calculated in a similar way.
(Odds ratio could be expressed as either: OR = the odds of exposure among cases divided by the odds of exposure among controls, or OR = odds of the outcome among the exposed divided by the odds of the outcome among the unexposed)
Source: Screening Tests lecture materials; Medical Statistics at a Glance Workbook, Petrie et al.