Home Assignment 3 for Epidemiology Course VHM 811 at AVC - Fall Semester 2007

The assignment is worth 10% of the final course mark.

The home assignment has six questions which should all be answered. Recall that home assignment are to be answered by each student individually; collaboration is not allowed (limited discussion is probably inevitable, and therefore acceptable). For other practical details about the assignment, consult the guidelines for home assignments for VHM 801. Note also that your solution should include text explaining the procedures used even if the calculations were done using computer software. A datafile is available (Stata format) for the data given in the table below.

The data for the assignment originate from a study on the impact of swine production factors on the occurrence of a particular respiratory disease condition in pigs. The data given here comprise 249 farms that were selected on the basis of information collected over a three-year period prior to the study. Farms with an average disease rate of 5% or more were considered as "cases", and farms with a disease rate less than 5% were considered as "controls". From the selected farms, information about management factors and history of other diseases were obtained. For this assignment we will focus on one particular management factor: the presence or absence of a mechanical air circulation system in the barns. The table below gives the breakdown of farms according to air circulation system and farm size, categorized into 5 categories.

Farm status	Farm size 0-199		Farm size 200-299		Farm size 300-399		Farm size 400-499		Farm size >=500
	Air circulation		Air circulation		Air circulation		Air circulation		Air circulation
	yes	no	yes	no	yes	no	yes	no	yes	no
case	2	4	15	8	13	7	7	2	54	4
control	7	27	30	18	19	10	5	4	12	1

Question 1.
Identify the study type from the description given above; if you think some further clarification is needed to ensure that the study is actually of the type under consideration, explain and make appropriate assumptions. Draw a diagram for the causal structure you would hypothesize for how the two factors affect the presence of a high level of respiratory disease on a farm. (Hint: For the relation between farm size and air circulation, choose the causal direction you find most sensible.)

Question 2.
Use the data to determine relevant measure(s) of association between air circulation and disease. Our focus here is on "crude" measures of association so the farm size should be ignored for this calculation. Determine also crude measure(s) of association between farm size and disease. Supplement with (crude) assessments of the statistical significance of these associations. (Hint: For this and the following questions, you will have to deal in a sensible way with the fact that farm size is categorized into five categories. In some cases it may be appropriate to reduce the number of categories, in other cases it may not be necessary or it may lead to a substantial loss of information. Whenever you decide to change the farm size categories, make sure to describe and motivate your decision.)

Question 3.
Carry out an epidemiological analysis to determine whether farm size is a confounder for the relation between air circulation and disease. Conversely, determine also whether air circulation is a confounder for the relation between farm size and disease.

Question 4.
Continue your epidemiological analysis from Q3 to investigate whether there is an interaction effect between farm size and air circulation on the disease. Give appropriate measures of association to describe the relations of the two factors with disease, and assess their statistical significance. As a summary of your analyses in Q3 and Q4, try to describe the causal structure in terms of the classification system introduced in Section 13.11 of VER.

Question 5.
Irrespective of your findings in Q3 and Q4, assume that confounding exists for one of the two relations studied above. If possible, choose a relation for which you described confounding, otherwise choose freely one the two relations as your focus of this question. Describe two methods to control for confounding in the design of a future study (involving the same factors and of the same type). Give sufficient detail to allow someone with no specific knowledge about study design to implement the selection of farms for the new study. You may also comment on which of the two methods you think would be most useful (or efficient) for this particular context.

Question 6.
For the last question we will consider an extra variable that was either recorded in the present study or is under consideration for inclusion in a future study. We will call this variable farm type. Different types of swine production farms exist, and farm type could also refer to some farms being enrolled in programs with increased level of preventive measures against disease, such as restrictions on the movement of animals. Assume that farm type has a causal relation with disease, and that some association also exists between farm type and farm size. Draw causal diagrams involving all three factors (and disease) to represent each of the following three scenarios:

1. farm type is not a potential confounder for disease,
2. farm type is a potential confounder for disease but does not need to be controlled if the farm size has been controlled for (analytically),
3. farm type is a potential confounder for disease and needs to be controlled even after the farm size has been controlled for (analytically).