In the introduction section of this paper, the authors state that previous studies of folic acid supplementation and preterm birth have reported conflicting results or imprecise estimates.
A-To which one of the criteria for causality proposed by Sir Bradford-Hill does the issue of “conflicting results” relate? Please explain in your own words what this means
B-What has likely been the main cause of “imprecise estimates” in previous studies?
C-What statistical indicator would you generally use to assess the precision of risk estimates?
In the introduction, the authors state that many researchers believe that folic acid is the critical factor affecting preterm birth, despite the fact that the majority of women participating in clinical studies (of preterm birth) have taken folic acid as part of a multivitamin supplement, rather than as a sole supplement.
What type of bias does this statement relate to? Please explain in your own words
The authors of this paper have carried out a longitudinal cohort study in which women were assessed for use of a folic acid supplement. They were followed-up over time, and it was recorded which of the women had a preterm birth. An alternative research design could have been that of a case-control study, comparing reported use of folic acid supplements in women who had a preterm birth (cases) with that of women who delivered their baby after a full term pregnancy (controls).
What main type of bias were the researchers avoiding by choosing a longitudinal cohort study instead of a case-control study? Please explain in your own words.
In the Methods section, the authors first make clear that a public health campaign to promote folic acid supplement taking in women was conducted in two southern provinces and one northern province of China. But then in the section “Selection of study subjects” they state that only women from southern China were included in the analyzes.
Please describe in your own words whether this may have affected the internal validity or the external validity of the study, and why (hint: the authors briefly discuss this in the Discussion section, but please expand on this using your own words).
In the Results, the authors write that the adjusted RR (relative risk) suggested a 14% reduction in preterm birth risk among folic acid users compared to non-users.
Which relative risk estimate in Table 2 does this finding refer to? Please provide the RR estimate and the 95% confidence interval.
Also, please indicate whether or not this relative risk indicates a “statistically significant” relationship and describe in your own words how you judged this.
The authors then describe their findings for subgroups of preterm birth and write that there was an association between folic acid use and spontaneous preterm birth, but not for the iatrogenic and PPROM subtypes.
What can you say about the incidence of these three subtypes in relation to the confidence intervals and precision of the RR estimates?
The analyses were repeated in three different groups of gestational age (20-27w, 28-31w, 32-36w).
Please describe why you think the authors were doing this. Please use the appropriate epidemiological term(s) that apply.
The third footnote note of Table 2 indicates that the relative risks presented in the final column were adjusted for 6 different variables.
Why was this done, and what criteria would have been used for selection of these variables?
In Table 4 the authors stratified their analyzed by subgroups of education level and party.
Please describe and briefly discuss in your own words the difference in findings for primiparous women (who gave birth to their first child) compared to multiparous women (who already had given birth to other children previously).
The folic acid supplements of this study were not randomly allocated and compared with a placebo, as in a randomized trial. Instead, in this observational study, the authors assessed the use of a folic acid supplement after a public health promotion campaign and they supplied supplements to women who wanted to take these.
Please describe the main advantage that a randomized trial would have offered compared to this observational study.
(Fictional) colleagues of Dr. Li have carried out a case-control study to investigate whether low folate intake is associated with the risk of developing colorectal cancer. They enrolled persons recently diagnosed with colorectal cancer into their study as cases. Controls were selected from the general population in the same area-of-residence as the cases. All study participants were asked to report their usual dietary habits from which folate intake was estimated and classified as either ‘low’ or ‘high’.
There were 88 colorectal cancer cases and 63 control persons who had low levels of folate intake, and there were 215 colorectal cancer cases and 398 controls with high levels of folate intake.
Construct a 2×2 table and calculate an appropriate measure of the strength of association between low intake of folate and risk of colorectal cancer.
How would you interpret the finding in Question 11?
Using the data provided in Q11, what proportion of colorectal cancer in the general population is potentially preventable (by eliminating low folate intake levels), assuming a causal association between low folate intake and colorectal cancer risk?
There are several factors that may cause confounding of the relationship between folate intake and colorectal cancer risk.
Please list one possible confounding factor that you think should be considered in the study described in Q11, and explain your choice