Table 1.
California birth hospitals and patient characteristics, 2018.
| Category | % distribution | 95% CI | # of hospitals | |
|---|---|---|---|---|
| Annual births (n = 72), American Hospital Association Annual Survey | ||||
| <1000 | 17% | 10% | 27% | 12 |
| 1000–1999 | 32% | 22% | 43% | 23 |
| 2000–2999 | 24% | 15% | 35% | 17 |
| 3000–3999 | 17% | 10% | 27% | 12 |
| 4000+ | 11% | 5% | 21% | 8 |
| Hospital ownership (n = 72), American Hospital Association Annual Survey | ||||
| Government, nonfederal | 19% | 12% | 30% | 14 |
| Nongovernment, not-for-profit | 67% | 55% | 76% | 48 |
| Investor-owned (for-profit) | 14% | 8% | 24% | 10 |
| Respondent-reported NICU level (n = 75) | ||||
| No NICU in hospital | 8% | 3% | 17% | 6 |
| Level I or II | 19% | 11% | 29% | 14 |
| Level III | 60% | 49% | 70% | 45 |
| Level IV | 13% | 7% | 23% | 10 |
| Respondent-reported frequency of maternal-fetal opioid-related exposures (n = 73) | ||||
| Often (1 in 10 patients, or more often) | 11% | 5% | 20% | 8 |
| Sometimes (between 1 in 10 patients and 1 in 100 patients) | 47% | 36% | 58% | 34 |
| Seldom (1 in 100 patients, or less often) | 30% | 21% | 41% | 22 |
| Never | 3% | 0% | 10% | 2 |
| Unsure | 10% | 4% | 19% | 7 |
| Respondent-reported hospital management of NAS (n = 75) | ||||
| Manages all levels of infants observed for or treated for NAS | 81% | 71% | 89% | 61 |
| Manages milder cases and transfers severe cases | 13% | 7% | 23% | 10 |
| Transfers all NAS cases | 1% | 0% | 8% | 1 |
Sources: “Hospital Care and Emerging Practices for Treatment of Maternal Opioid Addiction, the Mother–Infant Dyad and Neonatal Abstinence Care: A Survey of California Hospitals” fielded June 2018 to August 2018 by the Urban Institute in collaboration with the California Perinatal Quality-Improvement Collaborative and the California Maternal Quality Care Collaborative; 2015 American Hospital Association Annual Survey Database
n = sample size. Denominator includes hospitals that have American Hospital Association data, or, for survey responses, all respondents who selected a response in any part of a survey question (e.g., in a multi-item response table). A respondent who selected an answer in one line of the table but left another line blank are treated as “no” (instead of “missing”) for the line or lines for which they did not respond. 75 out of 145 birth hospitals in the sample responded to the survey
NAS neonatal abstinence syndrome, CI confidence interval