Rationing of nursing care interventions and its association with nurse-reported outcomes in the neonatal intensive care unit: a cross-sectional survey

Table 3 Fully adjusted regression models^a of the effects of care rationing on readiness for discharge and pain control (n = 125)

Care rationing (NEWRI)	Overall readiness for NICU discharge Estimate (95 % CI)	Pain control Estimate (95 % CI)
Parental support and teaching and infant comfort care	−0 · 46 (−0 · 73; −0 · 20)^**	0 · 96 (0 · 41; 1 · 50)^**
Discharge preparation	−0 · 53 (−0 · 71; −0 · 35)^**	−0 · 01 (−0 · 38; 0 · 36)

Abbreviations: NEWRI Neonatal Extent of Work Rationing Instrument, NICU neonatal intensive care unit
**p < 0 · 01
^aRegression coefficients are from models using the generalized estimating equation (GEE) framework to adjust for the effect on nurse clustering within a given NICU. To control for the inflation of Type 1 error associated with performing GEE on a small number of clusters (n = 7 NICUs), the modified GEE approach proposed by Morel et al. [36] was used. The adjusted models used the four subscales scores on the NEWRI as independent variables while controlling for nurses’ professional, demographic and employment characteristics, including: number of years worked as a nurse, highest degree actually held (college and hospital diploma vs. baccalaureate degree and above), full- vs. part-time employment, race (Caucasian vs. other)

ISSN: 1472-6955