Q3. A study on participants from the Prediction and Prevention of Preeclampsia and Intrauterine  Growth  Restriction  (PREDO)  cohort  investigated  the  link  between maternal  early  pregnancy  obesity  and  depressive  symptoms  during  and  after pregnancy. The following table represents the association between early pregnancy BMI and depressive symptoms and provided the statistics in terms of mean difference, confidence interval and p-value.

Using the given information in the table:

a.  Is the mean difference an absolute or relative measure of the effect size? (1.5 points)

b.  What  statistical  test  is  appropriate  to  define  the  association  between  BMI categories   and   depression   (assume   outcome   variable   is   not   normally distributed)? Justify your answer. (1.5 points)

c.  Is  the  mean  difference  for  obese  women  statistically  significant  during pregnancy and after pregnancy? Explain based on the CI. (2 points)

Q4. A cross-sectional study assessed the prevalence and association of obstructive sleep apnoea (OSA) prevalence in South Asians and White Europeans with type 2 diabetes mellitus (T2DM).

As part of investigation, the following table provides the prevalence of OSA among South Asian and White European with T2D

Using the given information in the table:

a.  What is the prevalence (%, n) of having moderate to severe OSA among South Asian participants? (1.5 points)

b.  What is the difference in prevalence of OSA (%, n) between White Europeans and South Asians with T2DM (show the calculation). Is the severity of OSA between   White   Europeans    and   South   Asians    participants   statistically significant? (2 points)

c.  What is the Inter-Quartile Range (IQR) of Apnea-hypopnea index for South Asian and White European participants? (1.5 points)

Q5. The table below contains odds ratios and their 95% confidence intervals for a number of outcomes for women in the treatment group (cervical pessary) and in the expectant management group (ignore the p value column). The data is source from a randomised controlled trial on the use of a cervical pessary in women with a short cervix to prevent pre-term births.

Using the given information in the table:

a.  Do you think the information above supports the use of a cervical pessary to reduce pre-term births (any birth prior to 37 weeks)? What is the basis of your answer? (3 points)

b.  Based on the information above, would you recommend the use of cervical pessary  to  reduce  chorioamnionitis  and  pregnancy  bleeding?  Justify  your answer. (2 points)

PART 2

Q1. Table 2 below is taken from a study that examined changes in traumatic stress and  co-occurring  emotional  and  behavioural  symptoms  following  treatment  in  a sample of children treated for complex trauma at a usual care clinical agency. The table shows the cut-off for statistical significance was set at p< .008.

Source: Data from Dauber et al., (2015)

a.  From Table 2, what inferential test statistic is applicable to test the hypothesis that children would show significant declines in symptoms from pre- to post- treatment? Why was this test appropriate for the analysis? What is the null and alternative hypothesis? (3 points)

b.  Briefly explain the change in pre- and post-treatment trauma symptoms for the analytic sample in reference to the alpha value of <.008. (6 points)

c.  How can you justify the difference between pre-and post-treatment trauma symptoms based on Cohen’s d (the measure of effect size)? According to Cohen (1988), effect size estimates can be interpreted as small (d = 0.0–0.20), medium (d = 0.30–0.50), and large (d = 0.60–0.80). (6 points)

PART 3

Q1. The figure below is from a longitudinal ecological study to investigate whether the uneven rise in prosperity between 1999 and 2008 accounted for differential increases in life expectancy (the outcome variable) in English local authorities (LA). The authors measured trends in prosperity as being represented by corresponding changes in unemployment, in household income, and in educational achievement, these were the covariates. A deprivation variable was also added to the model.

Source: Data from Barr et al. (2012)

a.  Do you think the model has a better fit (greater explanatory power) for men or for women? (3 points)

b.  Whose  life expectancy changes the most when unemployment decreases – men or women? Does it increase or decrease? How can you tell? (4 points)

Q2. The figure below from a study to assess the risk of adverse perinatal events of vaccination of pregnant women with an MF59 adjuvanted vaccine (a flu vaccine).

Source: Data from Rubinstein et al (2013)

a.  What does the figure above tell you about the odds (adjusted) for vaccinated   women, compared to non-vaccinated women, of delivering pre-term or having babies with low birthweight? (3 points)

Q3. The table is from a study was to determine the accuracy of parentally reported weights and lengths (of infants) at 12 months of age compared with measured weights and lengths obtained from a paediatrician’s visit, and to examine factors associated with parentally reported inaccuracies.

Source: Data from Hazrati et al. (2015)

a.  Interpret the Pearson’s correlation coefficients given in the table. (4 points)

b.  Explain  the  circumstances  which  would  persuade  you  to  use  Spearman’s correlation coefficient rather than Pearson’s. (2 points)

Q4. The figure below is from a time trend analysis comparing the actual number of suicides in 2009 with the number that would be expected based on trends before the financial crisis (2000–2007). The objective was to investigate the impact of the 2008 global economic crisis on international trends in suicide and to identify sex/age groups and countries most affected. The table shows Spearman’s correlation coefficients, 95% confidence intervals and p-values, between changes in unemployment rate and suicide rate between 2007 and 2009 in 50 countries, stratified by unemployment rate before the financial crisis in 2007.

Source: Data from Chang et al. (2013)

a.  What  do  these  results  suggest  about  the  association  between  changes  in unemployment and suicide rates? (3 points)

Q5. The  table  below  is  from  a  study  to  investigate  the  temporal  trends  in socioeconomic inequalities in adult obesity in Scottish population between 1995 and 2010/2011 using multiple socioeconomic indicators. The authors hypothesised that socioeconomic inequalities in adult obesity have widened throughout the study period in Scotland. The table contains the results of a chi-squared test of temporal trends in obesity prevalence within each socioeconomic stratum. Also shown is a chi-squared test for equality of proportions of body mass index (BMI) across years.

Data source: Jin et al. (2015)

a.  Interpret the results presented in the table. (4 points)

b.  Why was the chi-square the appropriate test to use in this analysis? (2 points)

PART 4

Read the Pasdar et al. (2021) paper and answer the following questions:

Here is the link to the paper:

https://bmcpublichealth.biomedcentral.com/articles/10.1186/s12889-022-13166- 0#Sec11

Q1. In Table 3, check how the data for socioeconomic status (SES), body mass index (BMI) and gender were reported. Classify whether they are nominal, ordinal, interval or ratio. Justify your answers. (3 points)

Q2. Table 3 presents the baseline characteristics in cases and sub-cohort groups. Which inferential tests were used to compare the following variables between the groups: a) gender,  b)  SES, c)  BMI,  and  d) total vegetables  intake?  Justify your answers. (5 points)

Q3. In  table  5,  why  do  you  think  the  Pearson  correlation  was  used  instead  of Spearman’s correlation? Explain the correlation between healthy eating index and dietary  pattern  based  on  the  given  r  and  p  values  obtained  from  the  Pearson correlation. You must provide your justification along with relevant data. Please do not simply copy and paste the answer from the paper. (5 points)