Protocol adherence for continuously titrated interventions in randomized trials: an overview of the current methodology and case study

Table 1 Protocol adherence according to different definitions

	Higher MAP	Lower MAP	p value
	(75–80 mmHg)	(60–65 mmHg)
	N = 58	N = 59^a
a. Proportion of total hours on vasopressors in and out of range^b
Each patient weighted equally
Time within range	33% (30–37%)	28% (24–32%)	0.053
Time above-target	39% (34–45%)	66% (61–71%)	<0.001
Time below-target	28% (22–33%)	6% (4–9%)	<0.001
Patients weighted proportionally to total hours on VP
Time within range	32% (29–35%)	31% (28–34%)	0.81
Time above-target	34% (30–38%)	61% (56–65%)	<0.001
Time below-target	34% (30–39%)	8% (3–13%)	<0.001
b. MAP out of range for four consecutive vasopressor hours
Number of occurrences of four consecutive hours/vasopressor days (mean events per day)	821/378 (2.2)	681/314 (2.2)	No important differences
Number of days with MAP out of range for at least four consecutive hours (%)	299/378 (79%)	259/314 (82%)	0.29
Number of patients with at least one occurrence of MAP out of range for four consecutive hours (%)	56/58 (97%)	57/59 (97%)	1.00
c. MAP values outside the targeted range for four consecutive hours without synchronous and consistent adjustments of vasopressor doses
Number of deviation events (mean events per day)	32 (0.08)	43 (0.14)	0.03^c
Number of days with at least one deviation event (%)	31 (8%)	40 (13%)	0.03^d
Number of patients with at least one deviation event (%)	26 (45%)	20 (34%)	0.26^e

MAP Mean arterial pressure
^aThis excludes 1 patient who did not have a vasopressor infusion
^bThe proportion of total hours on vasopressor and within, above or below target were calculated by first determining the proportion of hours on vasopressors that were within, above or below target for each patient. The patient level data was then used to calculate the target group means, confidence intervals and p-values comparing the higher and lower MAP target groups. The group level summary statistics were first calculated weighting each patient equally regardless of hours on vasopressors and were then re-calculated after weighting each patient proportionally to the number of hours they were on vasopressors
^c P-value calculated by generalized estimating equations (GEE) with exchangeable within subject working correlation, Poison dependent distribution and log-link
^d P-value calculated by GEE with exchangeable within subject working correlation, Poison distribution, log-link and the log of the number of vasopressors days as offset
^e P-value calculated Fisher’s exact test

ISSN: 1471-2288