Medical Statistics Made Ridiculously Simple » Week 3 » Understanding Medical Research: Your Facebook Friend Is Wrong

Q: The researchers state that the objective of this study was to “test the effect of the order of beer and wine consumption on the next day’s hangover severity.” What is the null hypothesis for this study?

The null hypothesis is a statement assuming no association in the tested measures, i.e. no association between the order of consumption of beer and wine and the severity of the next day’s hangover. Therefore, the null hypothesis would be that the order of consumption has no effect on the severity of the hangover.

Q: Table 1 in the paper states the baseline demographics of the study population. In the “Study Group 1” column, the following is written in the “Age, years” row: 23.9 ± 5.8. What do these numbers signify?

Medical journals often use the format of Mean ± Standard Deviation to display these two values. This can also be seen as Mean(SD). This means that in this particular study, the mean age of the study participants who belonged to group 1 was 23.9 years with a standard deviation of 5.8 years.

Q: The study results state that “There was no significant difference in blood or urine test results between the 3 groups, either before or on the day following the intervention (P > 0.05).” Which of the following is true about the significance of this outcome based on this statement?

A study is said to be statistically significant if the final results have a p-value less than the chosen p-value (in this case, greater than p=0.05 means that it is not statistically significant). This means that the null hypothesis cannot be rejected. Clinical significance is determined by assessing the effectiveness of the intervention (in this case, if the order of drinks had an effect on the hangover intensity). The results are not clinically significant either since the effect of the order of drinks was not statistically significant between the three groups.

Q: The authors state that a t-test was used to evaluate the primary hypothesis of the study. Which of the following is true of the study data? Check all that may apply.

The t-test is only used under the assumption that the data is normally distributed. Since the authors state that they used a t-test in their analysis, we can assume that the data is normally distributed. Normally distributed data is bell-shaped!

Q: Some of the study variables accounted for in this study are gender, body weight, height, and BMI. Which of these variables can be categorized as continuous? Check all that may apply.

Continuous data can be ordered and a meaningful relationship exists within each category. For example, a person who is 6’ 3” tall is exactly 1’ taller than someone who is 5’ 3” tall. Gender cannot be ordered.

Q: The study states that “more women than men vomited on both study days.” Which of the following statements is true for this study?

This study probably does not represent all the women and men — it is critical to check who was included in this study. The study mentions that there was a high loss to follow-up and it is critical to understand what confounders may be present in the study and the other limitations that might be associated with the analysis.