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.

The administration of postoperative analgesics was studied.

Treatment.

Cost-effectiveness analysis.

The study population comprised patients undergoing orthopaedic surgery for hernia or tight lumbar disk. The patients had not received oral morphine before surgery or morphine administered by patient-controlled analgesia (PCA) systems.

The setting was secondary care (a hospital department of orthopaedic surgery). The economic study was carried out at the Hospital of Besancon, France.

The effectiveness and resource use data related to patients who underwent orthopaedic surgery between May 2000 and July 2000. The resources used were estimated from the same effectiveness data source. The price year was 2000.

The effectiveness data were gathered from a single retrospective study.

The costing was undertaken retrospectively on the same patient groups as those used in the effectiveness study.

Power calculations were not reported as having been used to determine the sample size. The study sample consisted of 28 patients who underwent orthopaedic surgery and randomly received analgesic protocol 1 or 2. Four patients were excluded due to surgical difficulties (n=2), severe sepsis (n=1) or unavailable outcomes (n=1). Hence, 11 patients received protocol 1 and 13 patients received protocol 2. There were no statistically significant differences between groups 1 and 2 in terms of the mean age, gender, or surgery indications. The mean age of patients receiving protocol 1 was 51 years (standard deviation, SD=20.3) and 7 were men. The mean age of those receiving protocol 2 was 52.7 years (SD=16.6) and 5 were men.

This was a retrospective cohort study, which was conducted as a randomised trial and carried out in a single centre. The duration of the follow-up was 48 hours after the analgesic treatment was initiated.

The basis for the analysis of the clinical study (intention to treat or treatment completers only) was not specifically stated, but it is likely to have been to complete 48 hours of analgesic treatment. The main health outcomes used in the analysis were patient analgesic response and the side-effects of analgesic treatment (nausea, vomiting, constipation), assessed during the follow-up period. The patients' analgesic response was measured using a visual analogue scale pain score, ranging from 0 (none) to 10 (most severe), at 6-hour intervals (H0, H6, H12 and so on, up to H48). The authors considered that there was a significant difference in pain scores when the difference was above 2.

The authors reported that the pain scores were similar for both groups, except at interval H36. At H36, the difference in pain score between group 1 and group 2 was above 2 (in comparison with group 2).

The authors reported that there was no significant difference in the side effects. However, statistical estimates (chi-squared) were not reported.

There were no significant differences between the analgesic protocols in terms of clinical results.

Since the effectiveness analysis showed no difference in clinical benefit between the groups, the economic analysis was based on the difference in costs only (i.e. cost-minimisation).

The costs and the quantities were reported separately for all categories of costs. The resource quantities for nursing time were expressed in time units (minutes). Detailed quantities for side-effect treatments were not reported. The direct costs were for drugs, medical equipment (e.g. syringes and bandages) and nursing time. The costs of drugs and nursing time were obtained from the study centre (Besancon hospital). The cost of medical equipment was obtained from the French Central buying Service. The direct costs were not discounted since the follow-up period was less than 2 years. The price year was 2000.

The treatment costs and time data were analysed using the Mann Whitney U test (Statview).

No indirect costs were included in the analysis.

Euros.

A sensitivity analysis was not carried out.

Not applicable. See the 'Effectiveness Results' section.

The demand on nursing time for analgesic treatment was lower with protocol 2 (Euro 40.8) than with protocol 1 (Euro 71.1), (p=0.016).

The cost of analgesic treatment was lower with protocol 2 (Euro 14.4) compared with protocol 1 (Euro 40.5), (p=0.002).

The difference in the total direct costs between the two groups was significant, Euro 55.2 (protocol 2) versus 111.5 (protocol 1), (p=0.01).

The treatment costs of side effects were evaluated, but were not statistically analysed.

The costs and the benefits were not combined, as it is likely that the analgesic protocols were therapeutically equivalent. Therefore, the economic analysis included only the costs.

In classic orthopaedic surgery, postoperative pain control with oral morphine may be a good alternative to intravenous protocols, with similar effectiveness, acceptable side effects and simplification of nursing care, thus resulting in a significant cost reduction.

The reason for the choice of the comparator (intravenous analgesics) was clear. This analgesic protocol is widely used in the authors' setting. You should consider if this applies to your own setting.

No justification for the study design was reported. The authors described a retrospective study design for the cost analysis, but it is likely that a prospective study design was used for the effectiveness analysis (to administer prospectively a self-rating scale). A prospective and randomised controlled trial would have been more appropriate. The study sample was representative of the study population and the patients were shown to be comparable at analysis. However, biases arising from having a small patient sample (n=24) and a short duration of follow-up (48 hours) may be important factors.

The analysis of benefits was based upon the therapeutic equivalence of the treatment alternatives. The economic analysis therefore included only the costs.

The perspective adopted was not reported, but it is likely to have been that of the provider. You should decide if any relevant costs were omitted, and whether these are likely to affect the authors' conclusions when applied in your own setting. The unit costs were reported, but the quantities of resources were not reported for side-effect treatments. Statistical analyses were performed.

The issue of generalisability to other settings was not addressed. The data do not appear to have been used selectively to prove a particular point, and the choice of health outcomes was justified. The authors commented on one limitation of their analysis, the sample size, which may bias the comparison of pain scores.

The authors conclude that postoperative pain control with oral morphine seems to be a relevant alternative for patients undergoing orthopaedic surgery. However, this may need to be supported by prospective randomised studies conducted on a wider sample study.

None stated.

Chan VW. Impact of patient-controlled analgesia on required nursing time and duration of postoperative recovery. Regional Anesthesia 1995;20:506-14.

McCormack JP. A comparison of regularly dosed oral morphine and on-demand intramuscular morphine in the treatment of post surgical pain. Canadian Journal of Anaesthesia 1993;40:819-24.

Bourke M. A comparison of regularly administrated sustained release oral morphine with intra-muscular morphine for control of postoperative pain. Anesthesia and Analgesia 2000;90:427-30.