|
Outcomes-based trial of an inpatient nurse practitioner service for general medical patients |
Pioro M H, Landefeld C S, Brennan P F, Daly B, Fortinsky R H, Kim U, Rosenthal G E |
|
|
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 use of nurse practitioner-based care (NPBC) for patients admitted to general medical wards. Nurse practitioners provided initial and ongoing care for both subacute patients with chronic well-circumscribed illnesses and general medical patients with acute medical conditions under the direction of a senior physician. This included admission assessments, assembly of patient data, co-ordination of care with patients' attending doctors and the implementation of diagnostic and therapeutic plans. Off-hours emergency coverage was provided by medical house-staff.
Economic study type Cost-effectiveness analysis.
Study population The study population comprised general medical patients aged 18 to 69 years, who were admitted to general wards either from outpatient departments or the emergency room. Patients were initially excluded if they were transferred from intensive care units, or they were admitted during "off-hours" (when there were no nurse practitioners available). From July 1994, patients admitted during "off-hours" from Monday to Friday were considered eligible for inclusion in the study.
Setting The setting was secondary care. The economic study was performed in Cleveland, USA.
Dates to which data relate The effectiveness and cost data related to March 1994 to September 1995. The price year was not stated.
Source of effectiveness data The effectiveness data were derived from a single study.
Link between effectiveness and cost data The costing was performed retrospectively on the same sample population as that used for the effectiveness analysis.
Study sample It was not reported that power calculations were performed during the planning phase of the study in order to assure a certain power. Consecutive patients who met the inclusion criteria were enlisted. The final sample consisted of 381 eligible individuals, of which 193 (50.7%) were randomised to NPBC and 188 (49.3%) to HSC. After crossovers, 104 patients (27.3%) were admitted to the NPBC ward and 277 (72.7%) to the house-staff ward. The authors did not provide evidence that the study sample was representative of the study population.
Study design This was a randomised controlled trial that was performed at a single centre. The patients were randomly allocated using envelopes containing computer-generated random ward assignments. Randomisation seems to have depended heavily on the availability of beds: patients could be allocated to a group only if there were beds available in the ward (NPBC or house-staff). Ninety crossovers occurred between the groups, as the attending physicians of the patients could request that the patients change ward after randomisation. The patients were followed up for 6 weeks or until death, if death occurred earlier.
Analysis of effectiveness The basis of the effectiveness analysis was both intention to treat (ITT) and treatment completers only (TCO). The primary health outcomes assessed were:
the length of stay;
transfer to intensive care units;
in-hospital mortality;
30-day post-discharge mortality;
the percentage of patients experiencing more than one hospital-acquired complication;
the overall adverse event rate;
the percentage of patients discharged either home, to a skilled nursing facility or left against medical advice; and
the percentage of patients discharged home who received home care services.
Post-discharge vital status was assessed until 31 December 1995 using the National Death Index registry. Patient interviews were carried out to assess most of the effectiveness results. Among other questions, these included one validated survey (as stated by the authors) to assess patient satisfaction and the SF-36 questionnaire.
Patients randomised to NPBC and house-staff wards, following both ITT and TCO analyses, were shown to be comparable in sociodemographic characteristics such as age, gender, race and insurance. They were also shown to be comparable in terms of co-morbidity (mean Charlson co-morbidity score), severity of illness (mean APACHE III acute physiology score), functional status (ability to perform activities of daily living, ADL, and instrumental ADL), health status (SF-36 questionnaire) and mean symptom severity at admission, (p>0.10).
Effectiveness results The NPBC and house-staff patients had similar outcomes (either by ITT or TCO analyses) for length of stay, rates of in-hospital adverse events, transfer to intensive care units, mortality rates and discharge destination. They also had similar outcomes in improvements in functional status (ADL and instrumental ADL scores), health status (SF-36 scores) and symptom severity, (p>0.10).
The patients' assessment of care (based on overall care, patient-perceived problems, and doctor and nurse care) was also similar in patients receiving NPBC or HSC, independent of the analysis used (ITT or TCO), (p>0.10).
Among patients returning home after discharge, more NPBC patients received home care services than house-staff patients. The respective proportions were 9.2% (NPBC) and 3.9% (HSC) for the ITT analysis, (p=0.046), and 12.5% (NPBC) and 4.3% (HSC) for the TCO analysis, (p=0.006).
Clinical conclusions The clinical and functional outcomes were similar for both groups of patients, and were independent of the kind of care received (NPBC or HSC) and the analysis undertaken (ITT or TCO). The only statistically significant difference was that a higher percentage of NPBC patients discharged home received home care after discharge, compared with HSC patients discharged home.
Measure of benefits used in the economic analysis No summary measure of benefit was used in the economic analysis. The study was therefore categorised as a cost-consequences analysis.
Direct costs Some, but not all, of the resource quantities were reported separately from the costs. The direct costs considered in the study appear to have been those of the hospital. The authors reported the total hospital charges and the total ancillary charges. They also identified specific cost areas such as pharmacy, radiology, laboratory and respiratory therapy. However, they did not report a final figure summarising the total cost results. The direct costs were obtained from a review of the patients' medical records, the hospital administrative databases and the cost management information system of the hospital. Therefore, the costs and charges were estimated from actual data. Discounting was not performed, but was irrelevant since the follow-up was of a short duration. The study reported some average costs and some average charges. The price year was not stated.
Statistical analysis of costs The authors reported mean charges or costs for NPBC and HSC. They also reported the mean charge or cost difference (and the associated 95% confidence interval, CI) between NPBC and HSC, considering both ITT and TCO analyses.
Indirect Costs No indirect costs were considered in the economic analysis.
Sensitivity analysis No sensitivity analyses were performed.
Estimated benefits used in the economic analysis See the 'Effectiveness Results' section.
Cost results Following the ITT analysis, the mean hospital charges per patient were $8,854 for NPBC and $9,426 for HSC. The mean difference was -$572 (95% CI: -2,704, - 1,560).
The mean ancillary charges were $4,960 for NPBC, $5,358 for HSC, and -$399 for the difference between NPBC and HSC ancillary charges (95% CI: -1,820 - 1,023).
Following the TCO analysis, the mean hospital charges per patient were $8,866 for NPBC and $9,240 for HSC. The mean difference was -$374 (95% CI: -2,761 - 2,013).
The mean ancillary charges were $5,074 for NPBC, $5,189 for HSC, and -$115 for the difference between NPBC and HSC ancillary charges (95% CI: -1,707 - 1,477).
Synthesis of costs and benefits The costs and benefits were not combined due to the cost-consequences approach undertaken.
Authors' conclusions The results of this study suggested that nurse practitioner-based care (NPBC), supported by attending doctors, was associated with similar resource utilisation and similar clinical and functional outcomes in comparison with house-staff care (HSC) and attending doctors. These results were consistent and were independent of whether the analysis was conducted on an intention to treat (ITT) or treatment completers only (TCO) basis.
CRD COMMENTARY - Selection of comparators A justification was given for the comparator chosen. HSC was the type of care currently used before the introduction of the NPBC in the training hospital where the study was performed. You should consider if this is a widely used type of care for general medical patients admitted to general wards in your own setting.
Validity of estimate of measure of effectiveness The analysis used a randomised controlled trial, which was appropriate for the study question. The patients were randomised to one of the types of care, although the availability of beds in both the NPBC and HSC groups had a considerable influence on the final allocation of the patients (after crossovers). The authors stated that the high crossover rate, mainly from NPBC to HSC, might have introduced selection bias. However, the patient groups were shown to be comparable in terms of their sociodemographic characteristics (age, gender, race, insurance), co-morbidity, severity of illness, functional status, health status and mean symptom severity at admission, independent of whether the basis for the effectiveness analysis was ITT or TCO. The authors did not show any evidence that the study sample was representative of the study population. The eligibility criteria were not uniform for the whole study period, one criterion varied with time. The authors also highlighted two further limitations of the effectiveness analysis. First, the study had low power to detect clinically meaningful differences between the two groups of patients. Second, the patients and interviewers were not blinded to the treatment assignments.
Validity of estimate of measure of benefit The authors did not derive a measure of health benefit. The analysis was therefore categorised as a cost-consequences study.
Validity of estimate of costs Some, but not all, of the resource quantities were reported separately from the costs. The reporting of the costs was not sufficiently explained and it was not entirely clear what the final costs included. The authors used charges as the cost estimates of the interventions under analysis, which was appropriate given the hospital perspective. The authors highlighted the fact that they did not consider the differences in salaries between nurse practitioners and house-staff, the costs of the medical director, nor the costs of providing off-hours coverage by residents. Therefore, it seems that not all of the relevant costs have been included. Discounting was not performed, but was irrelevant since the costs were incurred in less than two years. The price year was not stated. These issues introduce uncertainty into the reliability of the cost results and hinder reflation exercises to other settings.
Other issues The authors reported that there were no medical records available for seven patients (2%), but they did not state how this lack of information was handled within the analysis. The authors compared their findings with those from other studies with different study populations, because there were no randomised controlled trials that compared NPBC and HSC in general medical patients admitted to general wards. The results of these comparisons led them to conclude that the results of this study may be generalisable to general medical patients, but that the generalisability to other teaching hospitals should be established. The results do not appear to have been selectively presented. The conclusions reflected the scope of the analysis.
Implications of the study The authors commented that it is unlikely that nurse practitioners can replace house-staff, although NPBC can complement HSC and reduce the number of house-staff required. They also recommended that organisational issues be considered before the implementation of NPBC. For example, the doctors' perceptions of nurse practitioners' capabilities (especially among patients perceived as being very ill), and the decreased flexibility of nurse practitioners to accommodate off-hours admissions and wide fluctuations in the number of patients.
Source of funding Supported by a grant from the Robert Wood Johnson Foundation's Health Care Financing and Organization Initiative.
Bibliographic details Pioro M H, Landefeld C S, Brennan P F, Daly B, Fortinsky R H, Kim U, Rosenthal G E. Outcomes-based trial of an inpatient nurse practitioner service for general medical patients. Journal of Evaluation in Clinical Practice 2001; 7(1): 21-33 Other publications of related interest Knickman JR, Lipkin M, Finkler SA, Thompson WG, Kiel J. The potential for using non-physicians to compensate for the reduced availability of residents. Academic Medicine 1992;67:429-38.
Mundinger MO. Advanced-practice nursing: good medicine for physicians? New England Journal of Medicine 1994;330:211-4.
Schaffner JW, Ludwig-Beymer P, Wiggins J. Utilisation of advanced practice nurses in health care systems and multispecialty practice. Journal of Nursing Administration 1995;25:37-43.
Sekscenski ES, Sansom S, Bazell C, Salmon ME, Mullan F. State practice environments and the supply of physician assistants, nurse practitioners, and certified nurse-midwives. New England Journal of Medicine 1994;331:1266-71.
Silver HK, McAtee PA. On the use of nonphysician 'associate residents' in overcrowded specialty training programs. New England Journal of Medicine 1984;311:326-8.
Indexing Status Subject indexing assigned by NLM MeSH Activities of Daily Living; Adolescent; Adult; Aged; Female; Hospitals, Teaching /manpower /standards; Hospitals, Veterans; Humans; Internal Medicine /manpower /standards; Male; Middle Aged; Models, Nursing; Nurse Practitioners /standards /utilization; Nursing Staff, Hospital /standards; Ohio; Outcome Assessment (Health Care) AccessionNumber 22001000560 Date bibliographic record published 29/02/2004 Date abstract record published 29/02/2004 |
|
|
|