Premium
How Many Nurses per Patient? Measurements of Nurse Staffing in Health Services Research
Author(s) -
Spetz Joanne,
Donaldson Nancy,
Aydin Carolyn,
Brown Diane S.
Publication year - 2008
Publication title -
health services research
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.706
H-Index - 121
eISSN - 1475-6773
pISSN - 0017-9124
DOI - 10.1111/j.1475-6773.2008.00850.x
Subject(s) - staffing , workforce , unit (ring theory) , minimum data set , data collection , nursing , medicine , health services research , health care , skill mix , workforce planning , medline , public health , psychology , statistics , nursing homes , mathematics education , political science , law , mathematics , economics , economic growth
Objective. To compare alternative measures of nurse staffing and assess the relative strengths and limitations of each measure. Data Sources/Study Setting. Primary and secondary data from 2000 and 2002 on hospital nurse staffing from the American Hospital Association, California Office of Statewide Health Planning and Development, California Nursing Outcomes Coalition, and the California Workforce Initiative Survey. Study Design. Hospital‐level and unit‐level data were compared using summary statistics, t ‐tests, and correlations. Data Collection/Extraction Methods. Data sources were matched for each hospital. When possible, hospital units or types of units were matched within each hospital. Productive nursing hours and direct patient care hours were converted to full‐time equivalent employment and to nurse‐to‐patient ratios to compare nurse staffing as measured by different surveys. Principal Findings. The greatest differences in staffing measurement arise when unit‐level data are compared with hospital‐level aggregated data reported in large administrative databases. There is greater dispersion in the data obtained from publicly available, administrative data sources than in unit‐level data; however, the unit‐level data sources are limited to a select set of hospitals and are not available to many researchers. Conclusions. Unit‐level data collection may be more precise. Differences between databases may account for differences in research findings.