Development and validation of nursing resource weights for the Belgian Nursing Minimum Dataset in general hospitals: A Delphi questionnaire survey approach

Abstract

Background Internationally, nursing is not well represented in hospital financing systems. In Belgium a nursing weight system exists to adjust budget allocation for differences in nurse staffing requirements, but there is a need for revision. Arguments include the availability of a nursing minimum dataset and the adverse consequences of the current historically based nursing weight system. Objectives The development and validation of nursing resource weights for the revised Belgium nursing minimum dataset (NMDS). Design Two independent cross sectional Delphi—surveys. Setting and participants A convenience sample of 222 head nurses from 69 Belgian hospitals participated in the cross sectional survey methods. To assess validity 112 patient case records from 61 nursing wards of 35 Belgian general hospitals representing general, surgical, pediatric, geriatric and intensive care were selected. Methods Nursing resource weights were constructed based on Delphi survey results by NMDSII intervention. The patient case Delphi survey results were used as the primary source for validation. A series of additional validation measures were calculated, based on the different patient classification systems. Finally, three validated nursing resource weighting systems were compared to the constructed NMDSII weighting system: the use of ‘Closon’, ‘Ghent’ and WIN weights. Results A coherent set of nursing resource weights was developed. The comparison of nurse resource weights, based on the survey per NMDS intervention versus the survey on patient cases, yielded high correlations: r = 0.74 to r = 0.97 ( p < 0.01) between three case rating questions, as an indication of reliability in terms of internal consistency, and r = 0.90 ( p < 0.01) between summed intervention weights and patient case weights, as an indication of criterion validity in terms of concurrent validity. Other concurrent validity measures based on summed intervention weights versus patient classification dependency weights showed a correlation ranging from r = 0.14 to r = 0.74. The correlation of summed intervention weights with the Closon, Ghent and WIN weights ranged from r = 0.93 to r = 0.96 ( p < 0.01), as a third indication of concurrent validity. Conclusions A system of valid nursing resource weights has been developed. The system should be further validated within an international context.