HLTH7025 Healthcare Data for Decision-Making, Spring semester 2022 Assignment 1 (Due Sunday 4 September 2022)
Please answer all 3 questions. Record your answers in the template document provided and submit via Turnitin before 11:59 pm on the due date. The marks allocated to each question are shown in the assignment. A total of 30 marks are available, and this assignment is worth 30% of your overall grade.
Question 1 requires you to analyse the unique assignment data set which was created for you.
This is labelled ‘dataforxxxxxxxx.xlsx’ where xxxxxxxx represents your Student ID number. The description of this data set is provided below. Your Assessment 1 data set was emailed to your student email address.
Copying answers from other students, sharing answers with other students, and writing answers in a group are all forms of cheating. Any student doing any of these will be reported for academic misconduct.
Note:
- Each student will get different answers as the data sets differ.
- Please use Excel to analyse data but format your answer by following the guidelines for preparing tables and Graphs.
Question 1 [9 marks]
A randomised controlled trial (RCT) study was carried out to determine the effect of smoking on coronary heart disease (CHD) in rural India. The study was conducted on 100 farmers. The baseline data of 100 farmers were presented. The data included demographic information (identification number, age in years, family income in Rupees, cholesterol levels (mg/L), smoking status, alcohol use and CHD levels. The data contain 7 variables. The 7 variables are id (identification number), age in years, family income in Taka, cholesterol levels(mg/L), smoking status (1=smokers and 2=non-smokers), alcohol use (1=No use, 2= mild use, 3=moderate use and 4 = heavy use) and CHD levels (1= Low risk, 2= Intermediate risk, 3= High risk).
a. Use row percentage to describe the relationship between smoking status and level of CHD and write 3-4 sentences summarising your findings, including public health implications of the findings with one recommendation. (4 marks).
b. Use appropriate statistics to describe measures and variability of the distributions of age in years, family income in Rupees, cholesterol levels (mg/L), – write the answer in your own words) (2 marks)
c. Use a scatterplot to describe the relationship between age in years cholesterol levels (mg/L), and report your correlation coefficient (r ) (2 marks)
d. Use the pie chart to describe the proportion of alcohol use among farmers in rural India. (1 mark).
Question 2a [9 marks]
Considering the following data from two countries (Tanzania and Bangladesh).
Country Household wealth # of pregnancy of
7+ months duration # of stillbirth # of early neonatal death # of perinatal death
Poorest 1753 25 29 54
Poor 1959 30 25 55
Tanzania Middle 1874 38 38 76
Rich 1549 23 29 52
Richest 1184 27 35 62
Total 8319 143 156 299
Poorest 1920 51 49 100
Poor 1819 51 39 90
Bangladesh Middle 1681 46 38 84
Rich 1798 44 44 88
Richest 1712 29 37 66
Total 8930 221 207 428
a. Calculate the stillbirth, early neonatal and perinatal mortality death rate for each household wealth in the two countries. What do you notice? (4 marks).
b. What are the total mortality rates for stillbirth, early neonatal and perinatal in the whole data? (2 marks).
c. Which of these two countries needed an urgent intervention to reduce the rate of perinatal mortality? and why? (2 marks)
d. What is the perinatal mortality rate for the overall total population? (1 mark)
Question 2b [5 marks]
In 2010, the Bangladesh government created a new division called the “Rangpur” division. (Abir et al., Global health action, 2017;10(1):14100482017) were interested in examining trends of stillbirth in Bangladesh using a population-based survey from 2004-2014. Their analysis produced two graphs for stillbirth, one graph with Rangpur division and the other graph without Rangpur division (as shown below).
Figure 1a was the rate of stillbirth and their 95% Confidence Intervals (CI) with Rangpur division, while Figure 1b was the rate of stillbirth and their 95%CI without Rangpur division.
a. Looking at the two graphs, do you think Abir et al. (2017) should produce two different graphs for stillbirth and provide reasons for your answer? (2 marks)
b. If you were employed as a public health officer in Bangladesh and were asked by the Minister of Health to provide some policy statement to him on ways to reduce stillbirth in Bangladesh using two graphs produced by Abir et al. (2017), which of the graphs would you choose for your policy statement and why? (3 marks)
Question 3 [7 marks]
The following graphs below show the prevalence of stunting by severity with 95% CIs of severely stunted, moderately stunted, mildly stunted and not stunted among children under 5 years in the Gicumbi (Rwanda) Kitgum (Uganda) and Kilindi (Tanzania) districts. [Agho et al., BMC paediatrics. 2019;19(1):1-1].
0 10 20 30 40 50 60 70 80 90 100
Kilindi (Tanzania) Kitgum (Uganda) Gicumbi (Rwanda)

Figure 1: Prevalence and 95% confidence intervals (CIs) of stunting by severity
a. What explanation can you offer for these differences in stunting prevalence in three disadvantaged East African communities? (4 marks)
b. What public health measures would you recommend to these three disadvantaged East African communities to reduce their stunting prevalence by a third by 2030? (3 marks).
Reference:
Agho KE, Akombi BJ, Ferdous AJ, Mbugua I, Kamara JK. Childhood undernutrition in three disadvantaged East African Districts: a multinomial analysis. BMC paediatrics. 2019 Dec;19(1):1-1.
Abir T, Agho KE, Ogbo FA, Stevens GJ, Page A, Hasnat MA, Dibley MJ, Raynes-Greenow C. Predictors of stillbirths in Bangladesh: evidence from the 2004–2014 nation-wide household surveys. Global health action. 2017 Jan 1;10(1):1410048.