Question

Cross- sectional studies are sometimes done with one respondent answering for the whole household. What problems do you anticipate in such a survey? Use terms presented in the course materials to name, explain, and provide an example of at least 2 specific kinds of bias that may be present.

Answer 1

Cross Sectional studies are the simplest form of observational studies in which a single examination of a cross section of population at one point in time is conducted and the results of which can projected on the whole population provided the sample has been done correctly

What problems do you anticipate in such a survey?

1. Not useful for short lived diseases.

Cross sectional studies are more useful for chronic than short-lived diseases. For example in a study of hypertension , we can also collects data during the survey about age, sex, physical exercise, body weight, salt intake, and other variables of interest. A point must be stressed that time sequence which is essential to the concept of causativity cannot be deducted from cross sectional data. People who dies quickly or recover quickly are less ilikely to identified in cross sectional studies.

2. Less information about incidence of a disease or less information about cause and effect

Although cross sectional studies provides information about disease prevalence , it provides very little information about the natural history of disease or about the rate of occurrence of new cases (incidence). In cross sectional studies the temporal relationship between exposure and outcome can be unclear that is which came first.

3. Problems in inferring changes over time

Difficult to make casual inference is a problem of cross sectional studies.

In cross sectional studies designed to study change , there are frequently several alternative explanations for the findings and that is precisely what good research design tries to avoid.

4.Unable to investigate the temporal relationship between outcome and cause. Associations identified might be difficult to interpret.

4. Findings can be flawed or skewed

5. The cross section are not always guaranteed to be representative sample.

6. Not good for the study of rare diseases

7. Susceptible for the biases such as non responsive and recall biases.

Problems of bias relies on memory or past records , the accuracy which may be uncertain, validation of information, obtained is difficult or sometimes impossible.

Use terms presented in the course materials to name, explain, and provide an example of at least 2 specific kinds of bias that may be present.

Bias

Bias is any systematic error in the determination of the association between the exposure and disease.

The biases may be selection biases or information biases

1. Selection biases

Selection biases occurs when sample selected for the study no longer reoresentive of overall population.

It occurs in various circumstances

A. Sampling bias

Some individuals within the the target population are more likely to be selected for inclusion than others.

B.Allocation bias

There is a systematic difference between participants in exposed and non exposed.

C. Loss to follow up bias

Some individuals lost to follow up differ from those who were not list to follow up with respect to the exposure and outcome.

D. Non responsive bias

There is a systematic difference between responders and nonresponders . That is significant difference between people who complet complete the survey and people who do not complete a survey.

E. Neyman Bias

Also known as Prevalence incidence bias.

A selection bias in which individuals with severe or mild disease or both are excluded.

F. Berksonian bias

Due to different rates of hospital admissions.

2. Information bias.

Information bias occurred when the key variables of the study are collected, measured , and interpreted inaccurately

They are

A. Memory or Recall bias.

Recall of information of disease decent on the exposure . Participants recall information on exposure differently depending on their outcome status.It may be more likely for the cases to recall the existence of certain events or factors than the controls who are healthy persons

B. Observer bias.

The investigators prior knowledge of the disease status or treatment of the subject leads the researcher to ask questions or assess the subject differently.

C. Interviewer bias

Bias may occur when the interviewer knows the hypothesis and also knows the cases are. Prior information may lead him to question cases more or there may be a tendency of the interviewer to obtain answers that support preconceived notions of the disease and healthy person.

D. Detection bias

It occurs due to the systematic differences between groups in how outcomes are determined.

F. Confounding bias

A confounding factor is defined as one which is associated with exposure and disease and is distributed unequally in study and healthy group.

Examples

1. Memory or recall bias

Those who have had a myocardial infarction might be more likely to remember and recall certain habits or events than those who have not . Thus cases have a different recall of past than controls.

2. Bias due to Confounding

In a study of the role of alcohol in the etiology of esophageal cancer , smoking is a confounding factor it is associated with the consumption if alcohol and is an independent risk factor for esophageal cancer.