|
The efficacy of proton pump inhibitors in nonulcer dyspepsia: a systematic review and economic analysis |
Moayyedi P, Delaney B C, Vakil N, Forman D, Talley N J |
|
|
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 proton-pump inhibitors (PPIs) in adult patients with nonulcer dyspepsia (NUD).
Economic study type Cost-effectiveness analysis.
Study population The study population comprised a hypothetical cohort of adult patients with symptoms of dyspepsia, negative findings on endoscopy, and a diagnosis of NUD made by a gastroenterologist.
Setting The setting was secondary care. The economic study was carried out in the USA.
Dates to which data relate The effectiveness data were derived from studies published between 1998 and 2004. Resource use was also derived from published studies, dating from 2002 to 2004. The price year, although not stated, can be inferred to have been 2004.
Source of effectiveness data The effectiveness evidence was derived from a systematic review of the literature.
Modelling A Markov-model (Data Pro TreeAge) was constructed to model the costs and benefits of PPI treatment for a 12-month time horizon. The model considered a 30-day course of treatment and incorporated subsequent outcomes and responses.
Outcomes assessed in the review The main outcome estimated from the literature was the PPI treatment effect, expressed as the relative risk (RR) at 1 month. Other important variables derived from the literature were:
the probability that the patient would remain dyspeptic after 1 month of treatment;
the probability that they would not require further therapy if symptom-free at 1 month; and
the probability that they would remain symptom free after 1 year.
Study designs and other criteria for inclusion in the review Only randomised controlled trials (RCTs) evaluating PPI therapy in NUD were included. The inclusion criteria included:
adult patients;
patients with dyspepsia who fulfilled Rome or Working Party definitions;
negative findings at endoscopy;
the study evaluated intervention with a PPI, using a comparison group that was either placebo, H2-receptor antagonists, prokinetic therapy, antiacids or mucosal protecting agents; and
the outcome assessed was symptoms of dyspepsia.
The exclusion criteria were:
trials that evaluated only patients with symptoms of gastroesophageal reflux disease;
trials that evaluated only patients who were predominantly (>20%) taking non-steroidal anti-inflammatory drugs; and
trials assessing outcomes at less than 1 week.
Sources searched to identify primary studies The systematic review used was conducted using Cochrane Collaboration methodology. The Cochrane Controlled Trials Register, MEDLINE, EMBASE and CINAHL were searched. Grey literature was searched using the SIGLE database. All these databases were searched until September 2002. Experts in the field of dyspepsia and pharmaceutical companies were contacted for unpublished materials. Recent relevant gastroenterology journals were handsearched.
Criteria used to ensure the validity of primary studies The quality of the trials was evaluated according to the BMJ criteria.
Methods used to judge relevance and validity, and for extracting data A single reviewer extracted the data from eligible trials on a standardised form. A second investigator conducted an unblinded check on the accuracy of the information.
Number of primary studies included Six papers, reporting eight international trials, were identified.
Methods of combining primary studies The efficacy data were combined using a random-effects meta-analysis.
Investigation of differences between primary studies There was statistically significant heterogeneity between the trials (heterogeneity chi-squared 30.05, d.f.=7, p<0.001). The variation in results between studies was not associated with the prevalence of Helicobacter pylori, gastroesophageal reflux symptoms, or duration of treatment. No statistically significant heterogeneity was associated with different doses of PPI.
Results of the review Eight trials comparing PPI therapy with placebo in 3,293 patients were identified. The RR of remaining dyspeptic with PPI therapy versus placebo was 0.86 (95% confidence interval, CI: 0.78 - 0.95; p=0.003) when using a random-effects model, with a number-needed-to-treat of 9 (95% CI: 5 - 25).
The probability that the patient would remain dyspeptic after 1 month of treatment was 0.77 (95% CI: 0.68 - 0.84).
The probability that the patient would not require further therapy if symptom-free at 1 month was 0.20 (95% CI: 0.13 - 0.29).
The probability that the patient would remain symptom-free after 1 year was 0.80 (95% CI: 0.71 - 0.87).
Measure of benefits used in the economic analysis The benefit measure used, months free from dyspepsia, was disease specific.
Direct costs Both productivity and health service costs were reported. The key resource use categories included were costs of medication, visits to the gastroenterologist, patient travel costs, and time off work or loss of leisure time. Resource use was reported separately from the costs. Discounting was not necessary for a 1-year time horizon. The price year can be inferred to have been 2004.
Statistical analysis of costs No statistical analysis of the costs was reported.
Indirect Costs Productivity costs were included, being conservatively valued using the US legal minimum wage and also at $25 per hour. Data on time off work was obtained from a community survey of dyspepsia in 8,473 individuals that was published in 2002. Discounting was not necessary for a 1-year time horizon.
Sensitivity analysis A probabilistic analysis was conducted using 5,000 Monte Carlo simulations. A sensitivity analysis was conducted in variables considered to vary in different settings (i.e. the cost of medication and the value of time off work). The distributions used in the model were either beta, gamma or log-normal. In some cases, distributions were not applicable.
Estimated benefits used in the economic analysis From a health care perspective, the estimated months free from dyspepsia were 3.73 for the PPI strategy, compared with 2.56 for the antacid strategy. This gives an incremental gain of 1.18 months symptom-free per patient.
Some variation could be observed by symptom sub-groups. PPI therapy was more effective in the case of reflux predominant symptoms and epigastric pain (4.57 and 3.82 expected months free from dyspepsia, respectively, versus 2.56). PPI therapy was less effective than antacid therapy in the case of dysmotility predominant symptoms (2.38 versus 2.56 free months with antacid).
Cost results From a health care perspective, the mean cost of the PPI strategy was $465 if the drug cost $90/month, and $205.9 if PPI medication was available over the counter ($19.99).
The analysis by symptom sub-groups suggested that, with the PPI costed as a generic drug, the mean cost per patient ranged from $166.4 to $230.1, but the difference in terms of marginal cost were more marked in the case of epigastric pain disorder ($69.6 incremental cost compared with the antacid option).
Synthesis of costs and benefits The cost-effectiveness results were first calculated in a deterministic manner. The PPI strategy would cost, on average, an extra $326 for an extra 1.18 months free from dyspepsia. This resulted in an incremental cost-effectiveness ratio (ICER) of $278 per month free from dyspepsia. Probabilistic results showed that the probability of PPI being cost-effective was 0.95 for a willingness-to-pay of $517 per month free of dyspepsia.
The results of the sensitivity analyses suggested that if over-the-counter prices were used (i.e. $19.99 instead of $90), then a PPI strategy would cost, on average, an extra $67 for 1.18 months free from dyspepsia gained. Sub-group analyses suggested that the PPI therapy remained cost-effective from a health care perspective in patients with predominant heartburn and epigastric pain. The base-case suggested that PPI therapy costs $45/month free from dyspepsia in patients with NUD with reflux symptoms.
Authors' conclusions Compared with no treatment, proton-pump inhibitor (PPI) therapy may be a cost-effective therapy in nonulcer dyspepsia (NUD) provided generic prices are used (incremental cost-effectiveness ratio, ICER=$56.98 per month free from dyspepsia).
CRD COMMENTARY - Selection of comparators The authors compared the use of PPI therapy with antacid therapy, which was assumed to act as an inexpensive placebo. However, the authors did not provide a detailed justification for this assumption. It was unclear whether this choice was appropriate, but it can be inferred that antacid therapy is the standard of care. You should decide whether it represents a valid comparator in your own setting.
Validity of estimate of measure of effectiveness The effectiveness estimates were derived from a well-conducted and appropriately reported systematic review of the literature. The results do not seem to have been presented selectively, and papers were considered regardless of their language and publication status. The authors clearly reported the methods used to combine the estimates of effectiveness. The estimates were investigated through comprehensive sensitivity analyses and a sub-group analysis of predominant symptoms.
Validity of estimate of measure of benefit The benefit measure used in the economic analysis was appropriate for the study question but was disease specific, thus making it difficult to compare with the benefits from other health care interventions, as would be the case with an appropriate utility measure.
Validity of estimate of costs The unit costs and resource use were reported separately. It can be inferred from the text that both therapies only differed in the cost of the medication, assuming the cost of antacid to be equivalent to placebo. However, differences in resource use between the PPI and antacid therapy would require further discussion.
Other issues In terms of comparisons with other studies, the authors stated that this was the first systematic review evaluating the efficacy of PPI therapy in NUD. Although the trials selected were international, the issue of the generalisability of the findings to other settings was not addressed. The authors discussed appropriately the limitations of the meta-analysis and the economic analysis.
Implications of the study The authors showed that treatment with PPIs has a small but statistically significant effect on symptoms of NUD. Hence, a modest proportion of patients with NUD have an acid-related disorder. The economic evaluation suggested that PPI therapy is not cost-effective from a health service perspective if the drug cost is $90/month, so the use of generic drugs would be advisable.
Source of funding Supported by a UK National Health Service R&D Primary Care Career Scientist Award.
Bibliographic details Moayyedi P, Delaney B C, Vakil N, Forman D, Talley N J. The efficacy of proton pump inhibitors in nonulcer dyspepsia: a systematic review and economic analysis. Gastroenterology 2004; 127(5): 1329-1337 Indexing Status Subject indexing assigned by NLM MeSH Clinical Trials as Topic; Costs and Cost Analysis; Dyspepsia /drug therapy /economics; Humans; Placebos; Probability; Proton Pump Inhibitors AccessionNumber 22004009241 Date bibliographic record published 31/12/2005 Date abstract record published 31/12/2005 |
|
|
|