|
Substitution of a nursing-led inpatient unit for acute services: randomized controlled trial of outcomes and cost of nursing-led intermediate care |
Griffiths P, Harris R, Richardson G, Hallett N, Heard S, Wilson-Barnett J |
|
|
Record Status This is a critical abstract of an economic evaluation that meets the criteria for inclusion on NHS EED. Each abstract contains a brief summary of the methods, the results and conclusions followed by a detailed critical assessment on the reliability of the study and the conclusions drawn. Health technology The health intervention examined in the study was a service of intermediate care in a nursing-led inpatient unit (NLIU) for the promotion of patient recovery before discharge to the community. The NLIU was a 19-bed ward admitting patients referred from acute wards in the same hospital. Care of patients was delivered by a team of nurses and managed by one of three senior nurses with a nurse-patient ratio of 0.84 nurses per patient over 24 hours.
Economic study type Cost-effectiveness analysis.
Study population The study population comprised patients eligible for the NLIU. The main requirement for eligibility in the NLIU was patient stability, defined as "having no major patho-physiological changes for at least 24 hours, having no changes in medical management anticipated and having no major tests or investigations planned or awaiting results". Only patients whose needs were potentially improvable by nurses were eligible for the NLIU.
Setting The setting was hospital. The economic study was carried out in a medium-sized district general hospital in London, UK.
Dates to which data relate No dates for effectiveness or resource use were reported. The price year was not given (Note: correspondence with the authors, after this abstract was written, has indicated that the price year was 1997/98).
Source of effectiveness data A single study was used as the source of the effectiveness evidence.
Link between effectiveness and cost data The costing was performed prospectively on the same patient sample as that used in the effectiveness analysis.
Study sample Power calculations were performed after enrolment and indicated that the study sample was able to identify medium effect sizes: a difference of 1.5 points on the Barthel index and 10 days of hospital stay (alpha=0.05; power=0.8). Of an initial group of 585 patients referred, 350 were eligible, but only 176 actually participated in the study: 89 patients (mean age: 77.6 +/- 10.7 years; 36% men) were included in the NLIU group and 87 patients (mean age: 79 +/- 12.4 years; 29% men) in the usual care group. The authors stated that no statistically significant difference was found between those who refused to participate and those who were included in the analysis.
Study design This was a randomised controlled trial, carried out in a single centre. Randomisation was performed using sequentially numbered sealed envelopes containing random allocation. Patient allocation was concealed until after recruitment. The length of follow-up was 28 days after discharge, in order to assess potential readmissions. Loss to follow-up was 9 patients (10%) in the intervention group and 7 patients (8%) in the control group. The main reason for loss to follow-up was patient death. No blind method of assessment was performed, due to the nature of the intervention.
Analysis of effectiveness The basis for the analysis of the clinical study was intention to treat. The health outcomes assessed in the analysis were length of inpatient stay, functional dependence at discharge (assessed using the 20-point Barthel index, where 0 indicates the maximum dependence), place of discharge, and readmission. Study groups were shown to be comparable at baseline in terms of age, gender, diagnostic group, medical specialty, and Barthel index.
Effectiveness results The effectiveness results were as follows:
The mean Barthel index score improvement was 3.6 (median = 3; SD = 3.29; 95% CI: 2.86 - 4.34) in the intervention group and 2.6 (median = 2; SD = 3.38; 95% CI: 1.84 - 3.36) in the control group.
Statistical analysis showed that care on the NLIU had no significant effect on the Barthel index at discharge.
Length of hospital stay was 36.9 days (median = 27; SD = 36.2; 95% CI: 29.3 - 44.5) in the intervention group and 26 days (median = 15.5; SD = 29; 95% CI: 26.5 - 36.5) in the control group and the difference was statistically significant, (p=0.036).
The place of discharge was independent (non-institutional) for 74 patients (83%) and institutional (residential or nursing home) for 6 patients (7%) in the intervention group, and independent (non-institutional) for 73 patients (85%) and institutional (residential or nursing home) for 6 patients (7%) in the control group.
The number of readmissions following discharge was 2 (3%) within 7 days and 6 (8%) within 28 days in the intervention group and 5 (6%) within 7 days and 6 (8%) within 28 days in the control group. There was no statistically significant difference in terms of discharge destination or in readmissions.
Clinical conclusions The effectiveness analysis showed that intermediate care in the NLIU led to an increase in total hospital stay, without any benefit in terms of patient functional dependence or number of readmissions and change in discharge destination.
Measure of benefits used in the economic analysis Health outcomes were left disaggregated and no summary benefit measure was used, thus a cost-consequences analysis was performed.
Direct costs Discounting was not relevant due to the short time horizon of the study. Unit costs were not reported separately from quantities of resources. Detailed cost data can be found in the full report of the study which is available on the web (see "Other Publications of Related Interest" below). The analysis of costs included tests and investigations, personnel, and hospital stay. The cost/resource boundary appears to have been that of the hospital. Costs were derived from service providers and published PSSRU. Costs for other professional inputs were based on a 30-minute consultation. Daily costs in the hospital ward were calculated using several approaches to identify the most appropriate method of assessment. The estimation of quantities of resources used was based on patient charts as estimated in the effectiveness analysis. The period of collection of quantities was not reported. No price year was given.
Statistical analysis of costs Standard statistical analyses of costs were performed to test for statistical significance of the results. Bootstrapping was also performed and 95% confidence intervals were calculated around the observed and bootstrapped means.
Indirect Costs Indirect costs were not included in the analysis.
Sensitivity analysis Sensitivity analyses were performed to assess the robustness of the estimated costs to variations in length of hospital stay, costs of tests and therapy, method used to estimate costs, discounting of unproductive nursing time, and use of specialty costs.
Estimated benefits used in the economic analysis Please refer to the effectiveness results reported earlier.
Cost results The cost results were as follows:
The mean hospital cost per stay was 5,144 in the intervention group and 4,100 in the control group, but the difference (1,044) did not reach the statistical significance, (p>0.05).
The 95% confidence interval around the bootstrapped difference was -306 to 2,518.
The mean daily cost was 139 in the intervention group and 158 in the usual care group.
The sensitivity analyses indicated that total inpatient costs became similar across the study groups only when mean treatment group stay was 29.4 days (a 20.3% reduction).
Variations in other factors did not change the main estimated costs.
Synthesis of costs and benefits Not relevant as a cost-consequences analysis was conducted.
Authors' conclusions The authors concluded that the NLIU service was not effective in improving patient functional dependence or number of readmissions and change in discharge destination. Costs were similar in the standard management procedure and in the NLIU intervention, but due to statistically longer length of hospital stay, overall costs tended to be higher in the NLIU group.
CRD COMMENTARY - Selection of comparators The rationale for the choice of the comparator was clear. Inpatient care in general hospital wards was selected as it represented the standard method of management for patients before discharge. You, as a user of this database, should decide whether it represents a routine intervention in your own setting.
Validity of estimate of measure of effectiveness The analysis of effectiveness was based on a randomised controlled trial, which was appropriate for the study question. The method of randomisation was reported. Study groups were shown to be comparable at baseline. The study sample appears to have been representative of the whole study population. The basis for the analysis of the clinical study was intention to treat and both length of follow-up and loss to follow-up were clearly reported. Power calculations were performed, although this was done after enrolment. These issues enhanced the internal validity of the analysis.
Validity of estimate of measure of benefit No summary benefit measure was used in the economic analysis. The analysis was therefore categorised as a cost-consequences study (see validity of effectiveness comments above).
Validity of estimate of costs The analysis of costs was conducted from the perspective of the UK NHS and it appears that all relevant categories of costs were included in the analysis. Three different methods for the estimation of costs were used better to assess inpatient hospital costs. Statistical analyses were performed on total costs. Several sensitivity analyses were also conducted. However, unit costs were not reported separately from quantities of resources and the price year was not given (but see comment under "Dates to which data relate" above), thus making generalising to other settings difficult.
Other issues The authors made some comparisons of their findings with those from other studies. The issue of the generalisability of the study results to other settings was not addressed and sensitivity analyses were performed only on costs. Furthermore, the external validity of the analysis was limited by the fact that the price year was not given and unit costs were not reported, although effectiveness results were fully reported. The study enrolled patients eligible for the NLIU and this was reflected in the conclusions of the study. The authors noted that management arrangements may have limited the development of innovative nursing services.
Implications of the study The authors highlighted the limited likelihood that intermediate care in an inpatient environment would result in cost reductions. Future research should be based on studies with a longer time horizon and should analyse the factors affecting length of stay, which represented the most important variable in the economic analysis.
Source of funding Funded by the Organisation and Management Group, NHS Executive, North Thames Research and Development programme (Grant No OM438).
Bibliographic details Griffiths P, Harris R, Richardson G, Hallett N, Heard S, Wilson-Barnett J. Substitution of a nursing-led inpatient unit for acute services: randomized controlled trial of outcomes and cost of nursing-led intermediate care. Age and Ageing 2001; 30(6): 483-488 Indexing Status Subject indexing assigned by NLM MeSH Aged; Aged, 80 and over; Female; Health Services for the Aged /economics /standards; Hospital Costs /statistics & Humans; Inpatients; Male; Nurse Practitioners; Nurse's Role; Outcome Assessment (Health Care) /statistics & Patient Care Team /economics; Patients' Rooms; Sensitivity and Specificity; numerical data; numerical data AccessionNumber 22002000266 Date bibliographic record published 31/03/2003 Date abstract record published 31/03/2003 |
|
|
|