Estimating nursing intensity and direct cost using the nurse-patient assignment

John M. Welton, Laurie Zone-Smith, Dipankar Bandyopadhyay

Research output: Contribution to journalArticlepeer-review

22 Scopus citations


Background: This study examines the feasibility of using the nurse-patient assignment (NPA) to calculate direct nursing hours and costs for each inpatient-day. The NPA data are collected at every hospital and therefore represent a readily available information source that can establish the intensity and economic value of nursing care at US hospitals. Method: Direct nursing care hours for each patient were collected twice a day using an existing nursing intensity database at a single university hospital between January 2004 and June 2005 for a total of 11,582 patient-days. Nursing intensity was also calculated for each shift using the NPA. Mean unit and hospital nursing hours were calculated and compared with the direct nursing care hours using ordinary least squares regression. Results: For the day shift, the NPA estimate explained 77.2% (r 2 = 0.772) of the variance of patient-level nursing intensity. Unit and hospital mean estimates of nursing intensity had lower r 2 of 0.574 and 0.456, respectively. The night-shift NPA, unit, and hospital r 2estimates were 0.824, 0.633, and 0.579, respectively. Conclusion: The use of the NPA can provide a robust and easy method to calculate nursing intensity for individual patients using assignment data available in nearly all care settings. The NPA estimate can be used to allocate direct nursing time and costs for each patient within the hospital billing system and can also be used in pay-for-performance or for benchmarking nursing intensity within and across hospitals.

Original languageEnglish (US)
Pages (from-to)276-284
Number of pages9
JournalJournal of Nursing Administration
Issue number6
StatePublished - Jun 1 2009


Dive into the research topics of 'Estimating nursing intensity and direct cost using the nurse-patient assignment'. Together they form a unique fingerprint.

Cite this