Advanced Statistical Methods in Epidemiology Assessment Exercise 2025 | UOS

Advanced Statistical Methods in Epidemiology 2025 Assessment Exercise 

For this assessment exercise, you are asked to analyse data from a study of childhood diarrhoea in a birth cohort in Uganda.

Pregnant women attending for antenatal care at a health facility serving a town and the surrounding peri-urban and rural areas were invited to join a study investigating the effect of maternal infections on the subsequent health of their infants. A total of 2,507 women were enrolled in the second or third trimester of pregnancy, and data were collected on their socio-demographic and household characteristics. Specimens were also collected at the time of enrolment and analysed for infections, including hookworm, Mansonella and malaria. The women were then followed up until they gave birth.

Excluding multiple births, there were 2,315 live births, and these formed a birth cohort of infants who were followed up to 5 years of age. Data were collected whenever they presented to the study clinic when they were sick. Episodes of childhood diarrhoea diagnosed at the study clinic were the outcomes of interest for this analysis. Data were also available on the sex, birthweight, immunisation and HIV status of the infants.

You are asked to use the data from this birth cohort to address the following questions:

• What was the incidence rate of diarrhoea in this cohort?
• How did this rate vary over time?
• Did each of the maternal infections have any effect on the incidence of diarrhoea in their infants?
• Was there any evidence that the incidence of diarrhoea either increased or decreased if children had experienced a previous episode?

In addressing these questions, you may use either or both of the datasets provided.

The Data

There were 2,315 children in the cohort and a total of 6,117 episodes of diarrhoea:

• 483 children had no episodes
• 444 children had 1 episode
• 409 children had 2 episodes
• 328 children had 3 episodes
• 214 children had 4 episodes
• 150 children had 5 episodes
• 287 children had more than 5 episodes

Two datasets are provided. The first is restricted to the first event of diarrhoea experienced by each child during the follow-up period. The second gives data on all events.

The Stata dataset Uganda.diarr.first.dta contains 2,315 observations and 25 variables as shown below. In this dataset, observation ends at the time (timeout) of the first episode of diarrhoea (diarr = 1) or exit from the study (diarr = 0).

The Stata dataset Uganda.diarr.multiple.dta contains 8,382 observations and 25 variables as shown below. In this dataset, there are multiple observations for each child with more than one episode of diarrhoea recorded. Each additional observation begins at the time of the previous episode (timein) and ends at the time (timeout) of the subsequent episode (diarr = 1) or exit from the study (diarr = 0).

Coding of Variables in Datasets

Background Notes

Maternal infections during pregnancy may affect the susceptibility of their infants to infectious diseases, for example, through effects on the child’s immunity. In this analysis, the main question is whether maternal infections (hookworm, Mansonella or malaria) during pregnancy are related to the subsequent incidence of diarrhoea in the children. Diarrhoea is a major cause of childhood morbidity and mortality, especially in low- and middle-income countries, leading worldwide to over half a million deaths each year. Episodes of childhood diarrhoea may be caused by a range of infectious agents. Not all such infections give rise to severe symptoms, and not all symptomatic cases are seen or treated at health facilities. Poor water and sanitation facilities may contribute to a high incidence of diarrhoea.

The Project Report

Your report on this analysis should be single-spaced and must be no more than 5 pages (A4), including any tables or figures. Fonts Times New Roman 12pt or Arial 11pt are both acceptable. Small margins (less than 2 cm) must not be used. The report should include:

• A brief discussion of the strategy you used in analysing the data (maximum 1 page). This should be more detailed than a Methods section of a scientific paper, since you should make explicit the structure of your analyses.
• A concise presentation of your results, including tables and figures as appropriate. Because of the strict space limits, you will need to be selective in the analyses you present.
• A brief discussion, summarising your main conclusions and discussing any potential sources of error or bias.

You do not need to explain the background to the study or the methods of data collection, nor is it necessary to cite or discuss relevant literature. The project report should be uploaded to Moodle by 14:00 BST (British Summer Time) on Friday, 30 May. Please see the ASME Moodle page for instructions.

There will be an opportunity on the afternoon of Friday, 16 May, to discuss your progress with the assessment project in small groups. A member of the course team will be present to discuss problems and suggest possible approaches.

Criteria for Grading

5   Excellent: An outstanding report which clearly answers the question, shows an in-depth understanding of the analysis and is well-explained with a comprehensive discussion.

4   Very good:  A thorough analysis, well explained, with all major points addressed; less thorough than “5”: less clearly explained or with more limited discussion of the findings.

3   Good:  Sound analysis, but some relevant points are omitted, and/or the presentation lacks clarity.

2   Satisfactory: Basic understanding of major points is shown, but some errors in the analysis or interpretation, or muddled presentation.

1   Unsatisfactory: Inadequate analysis and a lack of understanding shown.

0   Very poor: Serious lack of understanding shown: inappropriate analysis used or serious misinterpretation of results.