Table 2.
Primary and Secondary Outcomes
| Outcome | Ondansetron (N=45) | Placebo (N=41) | Difference (95% CI) a | P Value |
|---|---|---|---|---|
| Primary outcome | ||||
| Need for morphine to treat NOWS – #. (%) | 22 (49%) | 26 (63%) | −4 | 0.2c |
| Secondary outcomes | ||||
| Mean hospital stay (range) | 15.5 (2 to 64) | 17.4 (3 to 63) | −1.9 (−8.0 to 4.3) | 0.56d |
| Mean hospital stay (range) with a 15-day maximume | 9.3 (2 to 15) | 11.2 (3 to 15) | −1.9 (−4.0 to 0.2) | 0.07 |
| Median dose of morphine in 15 days (range) – mg | 9.3 (0.8 to 37.4) | 11.1 (0.8 to 31.1) | −1.8 (−8.8 to 2.7) | 0.31d |
| Mean modified Finnegan scoreb (standard deviation) | 4.6 (1.6) | 5.6 (2.1) | −1.0 (−1.8 to −0.2) | 0.02f |
| Use of adjunctive medicationg to treat NOWS – # (%) | 5 (11%) | 5 (12%) | 0 | 1c |
a
The 95% confidence interval (CI) between the ondansetron group and the placebo group was calculated with the use of Hodges-Lehmann estimator.
b
The modified Finnegan score of a neonate is represented by the mean of all modified Finnegan scores recorded for that neonate. The mean modified Finnegan score by ondansetron/placebo allocation reported in the table is the mean of all neonates’ scores in the corresponding allocation.
c
Fishers exact test.
d
Wilcoxon Rank-Sum test
e
By capping the maximum stay at 15 days, the statistical significance of the treatment effect is improved.
f
Two-sample t-test.
g
phenobarbital or clonidine