Panel comes to consensus on hospital care for opioid-exposed infants
Click Here to Manage Email Alerts
Key takeaways:
- Researchers recruited a panel of experts to determine quality indicators for care of opioid-exposed infants.
- Experts agreed that screening pregnant people for substance use disorder was among the most important.
Noting that practice varies widely across hospitals, an expert panel came to a consensus on the first quality indicators for the care of opioid-exposed infants.
Opioid-exposed infants have poorer outcomes, including longer and more complicated birth hospital stays and higher rates of hospital readmission and ED visits compared with uncomplicated newborns born at full term, one of the authors of a new article describing the project told Healio.
“Complications of opioid use disorder in pregnancy and opioid exposure in newborns are a growing public health challenge.” Jordan M. Harrison, PhD, a health policy researcher at the RAND Corporation, told Healio. “Quality monitoring may help to improve care and outcomes for this population, yet there is a dearth of quality indicators broadly applicable to hospital-based care of opioid-exposed infants.”
Much of the existing literature, Harrison said, has focused on the subset of opioid-exposed infants who develop drug withdrawal, also known as NOWS, and “there has been relatively little focus on opioid-exposed infants who do not develop the syndrome.”
“As we work toward improving care for this vulnerable population, it is important for us to focus on the whole population, and that was one of the motivators for this study,” Harrison said.
Harrison and colleagues recruited 32 expert panelists to use their individual and collective expertise to rate potential quality indicators on their importance and feasibility. The panel included health care providers with expertise in substance use in pregnancy and substance-exposed infants, parents in recovery, and quality and public health experts with expertise in maternal and child health, child welfare and substance use.
“To identify candidate quality indicators for the panel, we reviewed measures used in quality improvement programs, observational studies, and randomized controlled trials,” Harrison said. “We identified 49 candidate quality indicators that assessed structures, processes, and outcomes in multiple domains of clinical care.”
The researchers then solicited input from the panel using ExpertLens, an online modified-Delphi platform. In the first of three rounds, the researchers asked panelists to rate the importance and feasibility of each indicator on a nine-point Likert scale, where numeric scores corresponded with descriptive ratings of “low” (1 to 3), “uncertain” (4 to 6), or “high” (7 to 9). For the second round, panelists viewed and discussed the first-round results in online discussion threads, and in the third round, they had the opportunity to revise their original ratings.
The team then used the RAND/UCLA Appropriateness Method to determine whether the panelists reached consensus on each indicator, and calculated median scores for importance and feasibility, as well as a rapid analysis of comments and discussion threads to understand the rationale for participant ratings.
Ultimately, the experts had strong consensus on several indicators.
“Notably, we found that experts had strong consensus on the importance of quality indicators to assess other aspects of care, such as universal screening of pregnant people for substance use disorder, hospital staff training, standardized assessment for NOWS, nonpharmacologic interventions, and transitions of care,” Harrison said. “Not surprisingly, for indicators focused on processes and outcomes of care, experts saw feasibility as dependent on the information routinely documented in electronic medical records or hospital billing records.”
Harrison added that although they identified numerous quality indicators assessing structures of care in hospitals, such as hospital policies and staff training, there were fewer indicators assessing patient-level outcomes and processes of care.
“The lack of process and outcome measures limits aspects of care that can be used for benchmarking hospital performance or evaluating the effectiveness of interventions,” Harrison said. “More work is needed to validate the candidate process and outcome indicators we identified using electronic medical record and claims data.”
Collapse
Read more about
Click Here to Manage Email Alerts