Outcomes and Treatment Characteristics Comparing Infants Receiving Protocol-Driven Weans and Those Not Receiving Protocol-Driven Weans, Accounting for Regional Variation

No Stringent Weaning Protocol, N = 130Stringent Weaning Protocol, N = 417P
 Opioid treatment daysa (mean, CI)32.1 (28.7–35.5)17.7 (14.5–20.8)<.0001
 Day of life of dischargeb (mean, CI)23.6 (11.5–35.8)30.8 (18.6–43.0).0002
 Adjusted day of life of dischargec (mean, CI)32.1 (22.5–41.6)22.7 (13.1–32.2).004
Treatment characteristics
 Highest morphine dose givend (mg/kg; mean, CI)0.15 (0.04–0.27)0.09 (0.03–0.14).33
 Highest methadone dose givene (mg/kg; mean, CI)0.15 (0.07–0.24)0.14 (0.05–0.23).60
 Treated with phenobarbital (%, CI)20.7% (9.8–38.3)27.5% (13.8–47.1).22
 Phenobarbital treatment days (mean, CI)15.2 (8.6–21.9)19.4 (14.1–24.7).19
 Received multiple opioids (%, CI)6.3% (2.3–16.2)2.9% (1.0–8.0).08
  • Mean and 95% CI for each outcome are reported as least squares means derived from mixed models for each outcome. Estimated probabilities and 95% CIs reported as percentages. All statistical comparisons reported from mixed models with treatment site as the random effect. No confounders were statistically significant (P < .05) in the models. Confounders tested included birth wt, race, any prenatal care, insurance status, marital status, and maternal methadone use.

  • a Model includes gestational age as statistically significant covariate (P = .025). Level of care not statistically significant in model (P = .5).

  • b Significant interaction with level 2 versus level 3 care.

  • c Excluding 1 community outlier hospital driving interaction between protocol and level of care.

  • d Among 305 infants receiving morphine treatment, 57 had no stringent weaning protocol and 248 had a stringent weaning protocol.

  • e Among 254 infants receiving methadone treatment, 84 had no stringent weaning protocol and 170 had a stringent weaning protocol.