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 use of a laparoscopic approach to a Roux-en-Y gastric bypass (GBP) for the treatment of obesity.

Treatment.

Cost-effectiveness analysis.

The study population comprised severely obese adults.

The setting was secondary care. The economic study was carried out in the USA.

The effectiveness data were taken from studies published between 1984 and 2004. The costs were estimated from data from the years 2002 and 2004. All costs were adjusted to 2004 values.

The effectiveness data were derived from a review of published studies.

The outcomes assessed were:

the probability of the 16 most commonly reported major surgical complications,

the probability of conversion from laparoscopic to open GBP,

perioperative mortality,

the average operating time,

the amount of blood loss,

the average length of stay, and

the perioperative body mass index (BMI).

The review assessed journal articles published in English. Case reports, reviews, commentaries, repeated patient populations and studies with any arm smaller than 50 patients were excluded. Studies that reported outcomes for less than 5 of the 16 most common major surgical complications were also excluded.

The National Library of Medicine online journal database was searched for the studies included in the review.

Not reported.

Not reported.

Twenty-four primary studies were included in the review.

The data from the primary studies were pooled with the assumption that the data were directly exchangeable. The total number of events across all studies was reported as a proportion of the total sample size across all studies.

The authors explored differences between the primary studies in the discussion. They discussed the differential reporting of complications by different authors and the differences between participants. They did not perform statistical tests of heterogeneity.

The results of the review indicated that the mean length of stay was 2.5 days for a laparoscopic GBP compared with 3.7 days for an open GBP.

The average operating time was 194.6 minutes for laparoscopic GBP and 106.8 minutes for open GBP.

The average probability of conversion from laparoscopic to open GBP was 2.25% across all studies.

The average estimated blood loss was 104.6 cm3 for laparoscopic GBP compared with 301.5 cm3 for open GBP.

The rates of complications were too numerous to report in this abstract.

Not every study reported every outcome. The authors did not present information about the variability of the review results, but they did report the number of patients on which each estimate was based.

The authors used estimates of perioperative mortality to calculate the average years lost to death, based on average life expectancy adjusted for co-morbidities and BMI.

The estimated years lost to death were 17.8 for laparoscopic GBP compared with 20.3 for open GBP.

No summary health benefit was used for the economic analysis. In effect, a cost-consequences analysis was performed.

The study reported some resource use quantities and average costs separately. The study included only direct costs to the US health care system. These included the cost of the procedures (laparoscopic and open GBP), hospital stay and complications. The resource use data were taken from the effectiveness review, while the cost data were obtained from national databases and other published sources. The cost data were based on charges in the US health care system. Discounting was not relevant for the direct costs, which were incurred during less than one year. The study reported the average costs, adjusted to 2004 values using the medical care component of the Consumer Price Index.

The resource use and cost data were treated as point estimates. The authors stated that many of the studies included in the review reported only point estimates without providing the necessary information on variability. Consequently, they restricted their pooling and calculations to point estimates.

Productivity costs, as the cost of lost income from convalescence in hospital and lost income from years of life lost, were included. The average weekly salary was used to calculate productivity losses to patients. This was based on US national statistics for the year 2004. The quantities of life lost and length of hospital stay were reported. Discounting was relevant, as the study included productivity losses from death until average life expectancy, but it was not conducted.

US dollars ($).

The generalisability of the results was investigated by calculating costs adjusted for gender. No other sensitivity analyses were performed.

See the 'Effectiveness Results' section.

The total cost of laparoscopic GBP was $17,660, compared with $20,443 for open GBP. The results were presented as point estimates and discounting was not conducted.

Not relevant.

The favourable outcomes of laparoscopic gastric bypass (GBP) compared with open GBP, in combination with its lower cost, suggest that laparoscopic GBP is a more cost-effective weight loss method than open GBP.

The choice of open GBP as the comparator was explicitly justified to represent current practice in the study setting. You should decide if this is a widely used health technology in your own setting.

The effectiveness data were derived from a review of published studies. The authors stated that a systematic review of the literature had been undertaken. The methods of the review were reported in detail. The effectiveness estimates were combined very simply by taking the average of the point estimates from all studies that reported each outcome of interest, with the assumption that all studies were directly exchangeable. This method does not weight the estimates according to study size, and does not reflect the uncertainty and variability in the estimates. The authors stated that many studies in the review did not provide the information necessary to incorporate uncertainty in the pooled data. However, variation between studies in point estimates could have been reported, to allow the reader to assess heterogeneity among the pooled studies, but this was not done. The authors considered the heterogeneity of the primary studies in their discussion. The authors reported the results of the study with the caveat that strong assumptions were made in order to pool the data.

The authors did not derive a single measure of health benefit, and the analysis was therefore a cost-consequences study. The results of the review indicated that laparoscopic GBP is associated with an increased frequency of intestinal complications, while open GBP is associated more with extraintestinal complications. As the authors reported the effectiveness results in a disaggregated manner, the relative weight to apply to each outcome was unclear. The study assessed two forms of surgery for weight loss, but an outcome related to weight loss was not a primary outcome in the review. The authors informally assessed that laparoscopic GBP is associated with better outcomes than open GBP.

The authors stated that they would include costs to the US health care system, and these were included in the analysis. The direct average costs, based on charges, were reported separately from the quantities. The cost data used for productivity losses were not reported. Beyond calculating costs adjusted for gender, no sensitivity analyses were performed. This may limit the generalisability of the study results. The direct cost data were based on charges in the US health care system, and you should decide how similar these are likely to be to cost data in your own setting. The cost data were treated as point estimates, so variability and uncertainty were not included in the study results. Although the authors included productivity losses for years of life lost, they failed to discount the costs. The productivity losses were based on the average US salary, which may not accurately reflect lost earnings in this severely obese patient population. The authors did not assess the potential for change in real earnings over time, or with treatment. The price year to which the data related was reported.

The authors made appropriate comparisons of their findings with those from other studies. The issue of generalisability was not addressed in terms of other country settings, but between patient populations it was addressed. The authors did not report information on the variability in point estimates between the primary studies, so the impact of heterogeneity on the study results was unclear. They did, however, discuss the implications of heterogeneity between the primary studies in their discussion. The authors acknowledged a number of limitations to the study. These included the fact that certain resource use elements may be protocol driven, and the success and cost of laparoscopic GBP is highly dependent on the experience of the operating surgeon. They also suggested that, compared with open GBP, laparoscopic GBP may be more frequently carried out by more experienced surgeons. The authors pointed out that the long-term effects of laparoscopic GBP-induced weight loss were not assessed in the primary studies. The authors' acknowledgement of these limitations means that their conclusions reflect the scope of the analysis.

The authors recommended the use of laparoscopic GBP as a surgical intervention for weight loss, despite uncertainty about the long-term effects of the procedure.

None stated.