PUBH3017 Epidemiology Case Analysis Nut Consumption & Diverticulitis

SECTION A (35 Marks)

Study A

Study A was conducted in the city of Aym. Advertisements were placed in newspapers and popular on-line social media sites calling for volunteers for the study. Participants needed to be 50 years of age or older and not to have previously been diagnosed with diverticulitis to be eligible to take part in the study. Eligible participants were asked questions about their diet, including if they normally ate nuts at least once per week. They were followed for two years and any diagnosis of diverticulitis during those two years was noted. Persons who reported normally eating nuts at least once per week were younger (mean age=58 years) and were less likely to be female (55%) than persons who reported they didnt normally eat nuts at least once per week (mean age=66 years; 68?male).

Due to the excellent work of the study research officer, there were no dropouts during the study. One hundred and fifty of the 945 participants who normally ate nuts at least once per week were diagnosed with diverticulitis during the study. Fifty-five of the 560 participants who normally did not eat nuts at least once per week were diagnosed with diverticulitis during the study.

1) Study A

(i) Ignoring any error that might have arisen from measurement issues, do you think the study is likely to provide an unbiased estimate of the effect of normally eating nuts once per week on the risk of being diagnosed with diverticulitis? If you think there might have been bias, explain why you think this would have lead to an underestimate or an overestimate of the effect or why are you unable to tell? If you think there would not have been bias, explain why you think this? (10 marks)

2) Study A
Calculate the appropriate measure of effect that shows the relationship between normally eating nuts at least once per week and being diagnosed with diverticulitis. To do this you will need to draw a 2*2 table.

1. (i) Show this table. (5 marks)


2. (ii) Show your working. (5 marks).


3. (iii) Interpret the result you obtained (in Question 2 (ii)) in words that someone who has not studied

epidemiology would understand. (5 marks)

1. (iv) Based on these results, what proportion of the diverticulitis in people who normally eat nuts at least

once per week is due to normally eating nuts at least once per week. Show your working. (5 marks)

1. (v) How many extra people would need to normally eat nuts at least once per week for 10 additional

people to be diagnosed with diverticulitis? (5 marks)

SECTION B (65 Marks)

Study B

Study B was conducted in the city of Atsoca. All people aged 50 years or more who attended the Atsoca Central Surgical Clinic and were diagnosed with diverticulitis between 1 January 2012 and 31 December 2016 were included. The Atsoca Central Surgical Clinic was the largest of the three surgical clinics in Atsoca and covered about 50% of the Atsoca adult population.

Comparison people aged 50 years or more without diverticulitis were selected from the Atsoca electoral role. The electoral role was known to include 98% of all adult residents in Atsoca (and not to include people who were not resident in Atsoca). The comparison people were selected randomly from the electoral roll.

Of the 500 people diagnosed with diverticulitis, 400 agreed to take part in the study. The 100 who did not agree to take part did not give any reason for this. The 100 who did not take part in the study were considerably older than the other 400 people who did agree to take part in the study.

Of the 600 people selected from the electoral roll, 30 were excluded because they had recently been diagnosed with diverticulitis and 210 did not agree to take part in the study (they didnt give any reason for this). The 210 who did not take part in the study were considerably younger than people who did agree to take part in the study. Ninety of the 400 people with diverticulitis reported normally eating nuts at least once per week. Forty-five of

the 360 people without diverticulitis reported normally eating nuts at least once per week.

1. 3) Draw a flow diagram that summarises what happened in Study B. (15 marks)


2. 4) Study B




1. (i) Do you think the study base in Study B is well defined or not well defined? Justify your answer. (5 marks)


2. (ii) Do you think selection issues in Study B are likely to have biased the estimate of exposure (normally eating nuts at least once per week) in persons with diverticulitis? If so, do you think this would have lead to an underestimate or an overestimate or are you unable to tell? Justify your answer. (10 marks)


3. (iii) Do you think selection issues in Study B are likely to have biased the estimate of exposure (normally eating nuts at least once per week) in persons without diverticulitis? If so, do you think this would have lead to an underestimate or an overestimate or are you unable to tell? Justify your answer. (10 marks)


4. (iv) Do you think (randomly) choosing people from the electoral roll was an appropriate way to choose people as a comparison group for this study? (5 marks).




3. 5) Study B
   Calculate the appropriate measure of effect that shows the relationship between normally eating nuts at least once per week and the risk of being diagnosed with diverticulitis. To do this you will need to draw a 2*2 table.




1. (i) Show this table. (5 marks)


2. (ii) Show your working. (5 marks).


3. (iii) Interpret the result you obtained (in Question 5(ii)) in words that someone who has not studied

epidemiology would understand. (5 marks)

1. (iv) Based on the study results, what proportion of the cases of diverticulitis that occur in the community

are due to normally eating nuts at least once per week? (5 marks)