|
Information sharing can reduce laboratory use by emergency physicians |
Ramoska E A |
|
|
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 study considered information sharing in the form of presenting graphs and charts of attending physicians' laboratory use at monthly physicians' meetings, and posting this information on the quality improvement bulletin board. This was compared with no information sharing.
Economic study type Cost-effectiveness analysis.
Study population The study population comprised physicians in an emergency department (ED) who ultimately decided which laboratory tests to order. No exclusion criteria were stated.
Setting The setting was secondary care. The economic study was carried out in the USA.
Dates to which data relate The effectiveness data and cost data were collected between July 1993 and July 1994. A price year was not explicitly stated, but it was likely to have been 1994.
Source of effectiveness data The effectiveness data were derived from a single study.
Link between effectiveness and cost data Resource use was derived using the same study as that used to analyse the effectiveness measures, but the costing was performed retrospectively.
Study sample Power calculations to rule out the influence of chance on the results were not reported. The sample was selected by including attending physicians in the ED of the study facility and noting which laboratory tests they ordered over the study period. This sample was appropriate for the study question, as it comprised physicians who were ordering tests. The numbers of physicians, overall and in each group, were not reported. However, the same sample of physicians was used in each group. There was no report of physicians refusing to participate or being excluded.
Study design The analysis used a comparative study with a historical control. Physician laboratory use was monitored without the physicians' knowledge in the first part of the study (July 1993 to March 1994). In the second part of the study (April 1994 to June 1995), laboratory use was presented at monthly meetings and reported on the bulletin board in a blinded manner. Physician laboratory use continued to be measured. The study took place in a single centre, a 200-bed urban community teaching facility. The author did not report any loss to follow-up over the study period.
Analysis of effectiveness The analysis was based on actual laboratory use by attending physicians. The primary health outcomes were the mean number of laboratory orders per patient per month and the change in laboratory use over the two study periods. Since the same group of physicians was used in each period they were perfectly comparable. However, it was unclear whether the same patients were used in each part of the study. Therefore, the morbidity of the patients may have acted as an uncontrolled confounding factor.
Effectiveness results The mean number of laboratory orders per patient per month was 2.36 (+/- 0.21) during the control period. The use of the laboratory increased during this period (r=0.87, p<0.005).
By the end of the intervention period there were 1.94 laboratory orders per patient per month. The use of the laboratory decreased by 17.8% during this period, (p<0.01).
Linear regression showed a decrease in laboratory use (r=0.71, p<0.005).
There was one possible adverse outcome from the 34,320 patients seen during the 15-month study period.
Clinical conclusions The author concluded "information sharing and physician profiling can reduce the number...of laboratory studies ordered by attending emergency physicians without an adverse change in routine quality improvement indicators".
Measure of benefits used in the economic analysis There was no summary measure of health benefit. In effect, a cost-consequences analysis was performed.
Direct costs A perspective for the costing analysis was not reported. However, the perspective of the hospital appears to have been used. The costs were measured over the course of the clinical study. Discounting could have been conducted for the intervention group, as the follow-up for this group extended beyond one year. However, no discounting was performed. For the costing of the two groups to be comparable, the same unit costs would need to be used for both groups. It was not stated whether this was the case, or what the price year was. The analysis focused on the cost of the tests performed, the costs being reported to the ED by the pathology department and based on actual data. The laboratory costs represented the actual cost of labour, reagents, equipment and overheads. The unit costs and the quantities were not reported separately.
Statistical analysis of costs No statistical analysis of the costs was reported. However, the author estimated a regression line for the outcomes and costs over the control and intervention periods.
Indirect Costs The indirect costs were not estimated and were not relevant to the clinical question.
Sensitivity analysis No sensitivity analysis was reported.
Estimated benefits used in the economic analysis See the 'Effectiveness Results' section.
Cost results The mean hospital cost of ED laboratory studies was $32,415 (+/- 1,804) per month during the control period, and $26,687 per month during the intervention period.
There was a 17% decrease in the intervention period when compared with the control period, (p<0.01).
The slope of the regression line for the intervention period was twice that for the control period ($444 per month versus $221 per month).
Synthesis of costs and benefits The costs and benefits were not combined as, in effect, a cost-consequences analysis was performed.
Authors' conclusions The author concluded "information sharing and physician profiling can reduce the number and cost of laboratory studies ordered by attending emergency physicians without an adverse change in routine quality improvement indicators".
CRD COMMENTARY - Selection of comparators The author assessed the effects of an ongoing audit of laboratory use on physicians' use of laboratory tests and costs. The comparator was no audit. The comparator represented current practice in the author's setting.
Validity of estimate of measure of effectiveness The analysis used a comparative study with a historical control. This was appropriate for the clinical question, as it compared current practice with a potential future change in practice and included the same physicians in the two groups, thus reducing the possibility of confounding factors. The study sample was representative of the population since it included physicians who were ordering laboratory tests. The author acknowledged the existence of a "guest relations" programme in the second half of the study, which may have positively influenced patient outcomes such as the number of patient complaints. However, the author compared the average control quantity of tests per patient with the quantity of intervention tests per patient in the final week. These quantities were not strictly comparable. Therefore, the benefits of the intervention may have been slightly over stated.
Validity of estimate of measure of benefit The author did not derive a summary measure of health benefit. In effect, a cost-consequences analysis was performed.
Validity of estimate of costs The perspective adopted for the cost analysis was not reported. Therefore, it is not possible to comment on whether all the relevant costs were considered. As the excluded costs may have affected both the control and intervention groups equally, omissions in cost may not have influenced the key results and conclusions presented. The unit costs were not reported separately. The author did not state whether the same unit costs were used in the costing of both groups, given that there was a time lag between the groups and the price year was not stated. This hinders an assessment of the validity of the costing.
Other issues The author made brief comparisons of the findings with those from other studies, highlighting differences in the methodology underlying the studies. The issue of generalisability to other settings was not discussed. The results were not presented selectively, although a complete regression for the outcomes and costs covering both periods would have been useful for the reader. The author's conclusions accurately reflected both the scope of the analysis and the results presented. Limitations due to the quality indicators being too "coarse" were discussed.
Implications of the study The author did not make any recommendations for policy or practice as a result of the study, but noted that better quality improvement indicators would be required to improve the analysis in future.
Bibliographic details Ramoska E A. Information sharing can reduce laboratory use by emergency physicians. American Journal of Emergency Medicine 1998; 16(1): 34-36 Indexing Status Subject indexing assigned by NLM MeSH Clinical Laboratory Techniques /economics /statistics & Emergency Service, Hospital; Humans; Linear Models; Medical Audit; Patient Satisfaction; Practice Patterns, Physicians' /statistics & numerical data; numerical data /utilization AccessionNumber 21998000244 Date bibliographic record published 31/05/2004 Date abstract record published 31/05/2004 |
|
|
|