Question

3. Consider the data shown in Table 12–19. Identify three modifications, either design or statistical, that would result in a more effective table.

TABLE 12–19 Baseline Characteristics of Participants in Randomized Trial*

Characteristic	Treatment Group (n = 361)	Placebo Group (n = 344)	p-Value
Age, years	45.8	47.6	0.5467
Male sex	201	198	0.7561
Systolic blood pressure	128.25 ± 14.2	121.7 ± 13.91	0.645
Race/ethnicity
Asian	184	180
White	107	100
Hispanic	70	64

Answer 1

Sol:

Since sample sizes are different, it is advisable to represent categorical variables in terms of proportions (percentages) to afford a more meaningful comparison. The following table shows this modifications:

Characteristic	Treatment Group	Placebo Group
(n = 361)	(n = 344)
Age, years	45.8	47.6
Male sex (%)	55.68	57.56
Systolic blood pressure	128.25 ± 14.2	121.7 ± 13.91
Race/ethnicity (%)
Asian	50.97	52.33
White	29.64	29.07
Hispanic	19.39	18.60

With the above modifications, it is clear that the representation of all the categorical variables have been fairly uniform between the two groups under study.

On the response variable, Systolic blood pressure, the Treatment Group has a higher mean value than the Control Group. However, in terms of variability the two groups do not differ.

Since the response variable is known to be age-related, the observed difference might be more pronounced if the age variable is also factored into the analysis.