Evaluation of costs and effects of epidural analgesia and patient-controlled intravenous analgesia after major abdominal surgery. Bartha E, Carlsson P, Kalman S 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 Two treatments for postoperative pain after major abdominal surgery were examined. One was epidural analgesia (EDA) and the other was patient-controlled intravenous analgesia (PCIA). For EDA, a mixture of ropivacaine (2 mg/ml) and morphine (0.03 mg/ml) was delivered at a constant infusion rate of 5.5 mg/hour and rescue pain treatment was given according to a pre-specified algorithm. For PCIA, morphine (5 mg/ml) was used. The pump was programmed individually to meet requirements and delivery was on-demand. Both treatments lasted 3 days. Type of intervention Treatment. Economic study type Cost-effectiveness analysis. Study population The study population comprised patients who had undergone major abdominal surgery in routine care. Setting The setting was a hospital. The economic study was carried out in Sweden. Dates to which data relate The effectiveness and resource use data were gathered from 1997 to 1999. The price year was 2005. Source of effectiveness data The effectiveness evidence was derived from a single study. Link between effectiveness and cost data The costing was carried out retrospectively on the same sample of patients as that used in the effectiveness analysis. Study sample Power calculations were not performed. A sample of 644 consecutive eligible patients included in a database covering Page: 1 / 6
the period 1997 to 1999 was included in the analysis of effectiveness. Of these, 602 patients had received EDA and 42 had received PCIA. However, data were incomplete for 33 patients (all in the EDA group) and they were excluded from the analysis. Thus, the final sample comprised 569 patients (50% women) in the EDA group and 42 patients (57% women) in the PCIA group. The mean age of the patients was 57 (+/- 17) years in the EDA group and 48 (+/- 16) years in the PCIA group. Study design This was a retrospective cohort study that was carried out in Sweden. The number of centres involved was not stated. The patients were followed for 3 days, which corresponded to the duration of treatment. No patient in the final sample was lost to follow-up. Blinding was not carried out. Analysis of effectiveness All patients included in the final study sample were considered in the analysis of effectiveness. The primary outcome measures were the probabilities of the following: an unsuccessful attempt with EDA and thus treatment with PCIA; a poor effect with EDA and thus re-insertion of epidural catheter; a successful attempt with EDA; a poor effect or side effect with EDA and thus treatment with PCIA; a poor effect and side effect with either treatment, and thus change in treatment to bolus doses of morphine; minor problems with the epidural catheter, motor blockade and numbness, but continued treatment; accidental dislocation of epidural catheter; fulfilled treatment; pain during treatment, thus further intervention; no pain during treatment; nausea and vomiting (with and without pharmacological treatment); and respiratory depression (no pharmacological treatment needed). The proportion of patients who were pain-free was also assessed using a visual analogue scale (VAS) to monitor pain intensity every 3 hours. This was the main outcome measure used in the economic analysis. Patient characteristics were reported for both groups, but the baseline comparability of the study groups was not discussed. Effectiveness results The probability values were as follows: 0.014 for an unsuccessful attempt with EDA and thus treatment with PCIA; 0.012 for a poor effect with EDA and thus re-insertion of epidural catheter; 0.974 for a successful attempt with EDA; 0.09 for a poor effect or side effect with EDA and thus treatment with PCIA; Page: 2 / 6
0.05 with both strategies for poor effect and side effect, and thus change in treatment to bolus doses of morphine; 0.18 with EDA for minor problems with the epidural catheter, motor blockade and numbness, but continued treatment; 0.06 with EDA for accidental dislocation of epidural catheter; 0.78 with EDA and 0.95 with PCIA for fulfilled treatment; 0.3 with EDA and 1.0 with PCIA for pain during treatment, thus further intervention; 0.7 with EDA for no pain during treatment; 0.28 with EDA and 0.42 with PCIA for nausea and vomiting (with and without pharmacological treatment); and 0.01 with EDA and 0.04 with PCIA for respiratory depression (no pharmacological treatment needed). The proportion of patients who were pain-free was only reported in the graphical representation of the decision tree, and this depended on the different pathways followed by the patients. In particular, patients who received PCIA and fulfilled the treatment had a probability of 60% for 3 pain-free days, 17.5% for 2 pain-free days, 0% for 1 pain-free day and 22.5% for 0 pain-free days. Patients who received EDA without side effects and with no need for intervention had a probability of 93% for 3 pain-free days, 5% for 2 pain-free days, 1% for 1 pain-free day and 1% for 0 pain-free days. The other pathways were, in general, characterised by a lower probability of having pain-free days. Clinical conclusions The probability values derived from the clinical database were used as clinical inputs in the decision model. Modelling A decision tree model was constructed to assess the costs and benefits of the two pain treatment strategies. The two main branches of the tree represented EDA versus PCIA. The most relevant pathways and events included completed treatment, change of treatment, unsuccessful attempt to introduce epidural catheter, early dislocation of catheter, reinsertion of epidural catheter, and the need for additional pain treatment. A graphical representation of the tree was provided in the paper. The key assumption of the model was that the costs and the main clinical benefit of the treatments had a linear relationship. Measure of benefits used in the economic analysis The summary benefit measure used was the expected number of pain-free days for each treatment strategy. This was obtained using a modelling approach. Pain-free days were estimated both at rest and during activity. Direct costs The perspective of the cost analysis was not explicitly stated. The cost categories taken into consideration were personnel, material, drugs and postoperative care. The unit costs were not presented separately from the quantities of resources used since 3-day costs for each item were reported. The costs were derived from the hospital administration and from the hospital pharmacy. The costs of postoperative care came from official tariffs. Resource consumption was estimated from interviews with the staff. The costs were weighted on the basis of event probabilities derived from the clinical database. Discounting was not relevant as the costs were incurred during a very short timeframe. The price year was 2005. Statistical analysis of costs The costs were treated deterministically. Page: 3 / 6
Indirect Costs The indirect costs were not considered in the economic analysis. Currency The costs were estimated in Swedish kroner (SEK) and then converted into euros (EUR). The conversion rate was EUR 1 = SEK 9. Sensitivity analysis A sensitivity analysis was performed to assess the robustness of the cost-effectiveness results to uncertain model inputs, such as the cost of postoperative care and the incremental effect of EDA over PCIA during activity. A scenario analysis was also carried out by reducing the probabilities of the technical difficulties related to EDA. The influence of potential biases related to data derived from a non-randomised population was investigated: each patient treated with PCIA in the database was matched with one patient treated with EDA on the basis of similar age, gender and ASA-group. Estimated benefits used in the economic analysis The expected pain-free days at rest were 2.36 with EDA and 2.17 with PCIA (difference 0.19). The expected pain-free days during activity were 1.86 with EDA and 1.27 with PCIA (difference 0.59). Cost results The total costs per patient were EUR 1,701 with EDA and EUR 627 with PCIA (cost-difference EUR 1,074). Synthesis of costs and benefits Average and incremental cost-effectiveness ratios were calculated to combine the costs and benefits of the two strategies. The average cost per pain-free day per patient was EUR 721 with EDA and EUR 289 with PCIA. The incremental cost per pain-free day gained with EDA over PCIA was EUR 5,653. From the sensitivity analysis, the incremental cost per pain-free day gained with EDA over PCIA was EUR 1,488 when the lowest tariff for postoperative care was used, EUR 1,896 with better pain values with EDA, EUR 2,664 in the scenario analysis, and EUR 4,308 in the matched population. Authors' conclusions Epidural analgesia (EDA) was three times more expensive than patient-controlled intravenous analgesia (PCIA) in the treatment of pain after major abdominal surgery in Sweden. However, EDA was also more effective in terms of the pain-free days. The question of whether the additional cost of EUR 5,625 per pain-free day is reasonable for society requires a judgement of value, but it is likely that the results of the study indicate the poor cost-effectiveness of EDA. CRD COMMENTARY - Selection of comparators The rationale for the selection of the interventions was clear as the authors stated that EDA and PCIA were the most common pain relief strategies after major abdominal surgery. Medication dosages were clearly reported. You should decide whether they are valid comparators in your own setting. Validity of estimate of measure of effectiveness Page: 4 / 6
The effectiveness evidence came from a cohort study. The retrospective nature of the study limits the validity of the primary estimates, and the impact of selection bias or confounding factors cannot be ruled out. However, the authors considered smaller matched groups of patients in the sensitivity analysis in order to improve the internal validity of the study. The method of sample selection was not clearly described, although the inclusion of consecutive patients increased the robustness of the study. Further, there was no evidence (e.g. power calculations) of whether the sample size was appropriate. The baseline comparability of the study groups in terms of patient demographics and other clinical aspects was not discussed, and this might further limit the robustness of the clinical evidence. The number of medical centres involved in the analysis was not explicitly stated. These issues should be considered when assessing the validity of the clinical study. Validity of estimate of measure of benefit The summary benefit measure was specific to the disease considered in the study. It would not comparable with the benefits of other health care interventions. The authors stated that the impact of the pain treatments on quality of life was not investigated on account of the very short timeframe of the analysis. However, it was acknowledged that painfree days were not an appropriate measure with which to capture the impact of the interventions on patient health. Validity of estimate of costs The cost analysis included the hospital costs associated with post-surgery treatment. A breakdown of the cost items was not reported, and information on the unit costs and quantities of resources used was not given. Thus, it would be difficult to replicate the analysis in other settings. The cost estimates were treated deterministically in that no statistical test was undertaken. However, the impact of changes in uncertain cost estimates was investigated in the sensitivity analysis. The sources of the data were reported. The price year was given, which will facilitate reflation exercises in other time periods. Other issues The authors stated that their cost estimates were comparable with those observed in other studies, although the ranges of differences may vary depending on the setting. The sensitivity analysis showed that the base-case results were sensitive to changes in the values of the VAS measurement. However, the use of alternative values did not alter the conclusions of the analysis. The study referred to the general population of patients who had just undergone major abdominal surgery and this was reflected in the authors' conclusions. Implications of the study The study results suggest that, before a robust conclusion can be drawn on the cost-effectiveness of EDA over PCIA, the difference in costs should be related to other outcome measures such as length of hospital stay, morbidity and mortality. The well-being of patients should also be taken into consideration. Source of funding None stated. Bibliographic details Bartha E, Carlsson P, Kalman S. Evaluation of costs and effects of epidural analgesia and patient-controlled intravenous analgesia after major abdominal surgery. British Journal of Anaesthesia 2006; 96(1): 111-117 PubMedID 16257994 DOI 10.1093/bja/aei270 Page: 5 / 6
Powered by TCPDF (www.tcpdf.org) Other publications of related interest Brodner G, Mertes N, Buerkle H, et al. Acute pain management: analysis, implications and consequences after prospective experience with 6349 surgical patients. Eur J Anaesthesiol 2000;17:566-75. Paulsen EK, Porter MG, Helmer SD, et al. Thoracic epidural versus patient-controlled analgesia in elective bowel resections. Am J Surg 2001;182:570-7. Norris EJ, Beattle C, Perler BA, et al. Double-masked randomized trial comparing alternate combinations of intraoperative anesthesia and postoperative analgesia in abdominal aortic surgery. Anethesiology 2001;95:1054-67. Indexing Status Subject indexing assigned by NLM MeSH Abdomen /surgery; Adult; Aged; Analgesia, Epidural /economics; Analgesia, Patient-Controlled /economics; Cost- Benefit Analysis; Decision Trees; Drug Costs; Female; Health Care Costs; Humans; Male; Middle Aged; Models, Econometric; Pain, Postoperative /drug therapy /economics; Pilot Projects; Sweden; Treatment Outcome AccessionNumber 22006006194 Date bibliographic record published 30/09/2006 Date abstract record published 30/09/2006 Page: 6 / 6