Quality Measures to Assess Quality of Neonatal ICUs

Results of a multicenter, cross-sectional data analysis [JAMA Pediatr. 2013;167(1):47-54] show that individual measures of quality provide only a modest association with overall quality of neonatal intensive care units (NICUs), suggesting that assessment of overall NICU quality should not be based on a single or a few quality measures.

Although increasing attention is being paid to the quality of care provided by healthcare providers in an effort to increase efficiency and improve patients’ quality of care, little attention has been paid to whether the quality of institutional performance can be adequately assessed by a single or limited set of quality care measurements.

To address this, investigators assessed a number of outcome measures from 5445 very low-birth-weight infants cared for between January 2004 and December 2007 at 22 regional NICUs in California. Infants were excluded from the study if they were transferred out of the NICU for reasons other than convalescent or chronic care, had a gestational age <25 weeks, a body weight >1500 g, or major congenital anomalies, or died within 12 hours of birth, were transferred in the NICU after age 3, or were readmitted or died after transfer out of the NICU.

Eight outcome measures, chosen by an expert panel, were assessed to determine their correlation to quality of care: (1) antenatal corticosteroid use, (2) hypothermia (<36 degrees C) during the first hour of life, (3) nonsurgically induced pneumothorax, (4) healthcare-associated bacterial or fungal infection, (5) survival to discharge or to 36 weeks’ gestational age with chronic lung disease, (6) discharge on any human breast milk, (7) mortality in the NICU, and (8) high growth velocity.

The study found significant variations in clinical processes and outcomes between NICUs within and across each measure of quality, with only modest correlations between most measures of quality. Only 6 of 28 unit-level correlations were significant (P<.05), and correlations between pairs of quality care measures were negligible for 14 pairs (P<.01), weak for 5 pairs (range of P>.01 to P<.03), moderate for 8 pairs (range, P>.03 to P<.05), and strong for only one pair (P>.05). The only pair with a reasonable correlation was high growth velocity and the absence of healthcare-associated infection.

Among the NICUs, the study found little consistency of high performance. Based on the 8 measures of quality of care, the number of times that NICUs were among the top 4 ranks (ie, a high performer) ranged from 0 (never among the top 4 ranks) to 4 (in the top 4 ranks for 4 or 8 measures).

According to the investigators, these findings highlight a low degree of systems integration within the NICU setting and the possibility that a tightly integrated and standardized care delivery system does not exist in neonatal intensive care.

The study also used an exploratory factor analysis, with factor loadings in excess of 0.5 used to classify variables into factors, to determine whether any underlying factors were driving the correlations. This analysis found 4 underlying factors of quality: (1) pneumothorax, mortality in the NICU, and antenatal corticosteroid use loaded on factor 1; (2) high growth velocity and healthcare-associated infection loaded on factor 2; (3) chronic lung disease loaded on factor 3; and (4) discharge on any human breast milk loaded on factor 4.

These findings, according to investigators, suggest the inability to infer overall NICU quality based on a single or a few measures of quality and “call into question the assumption that this measurement approach will lead to widespread improvements in quality.”

Rather, they emphasize the possible need for using composite indicators to measure multiple factors to promote multidimensional improvements using system-based interventions.