|
Impact of a formulary change in proton pump inhibitors on health care costs and patients' symptoms |
Raisch D W, Klaurens L M, Hayden C, Malagon I, Pulliam G, Fass R |
|
|
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 proton-pump inhibitors, lansoprazole and omeprazole, were studied. No further details were given.
Type of intervention Other: Impact of a formulary change that concerned treatments.
Economic study type Cost-effectiveness analysis.
Study population The study population comprised patients in the study setting who were receiving omeprazole and who were switched to lansoprazole because of the formulary change. No other inclusion or exclusion criteria were stated.
Setting The setting was secondary care. The economic study was carried out in Tucson (AZ), USA.
Dates to which data relate The effectiveness data and cost data were collected between August 1996 and August 1998. The price year was 1999.
Source of effectiveness data The effectiveness data were derived from a single study.
Link between effectiveness and cost data The costing was carried out retrospectively using the same sample of patients as that used in the effectiveness study.
Study sample Patients whose outcomes were successful, or who failed in the switch from lansoprazole to omeprazole, were identified in another study (see Other Publications of Related Interest). Power calculations, to estimate the influence of chance on the results, were not reported to have been carried out. Failure patients were identified and a matched comparison of successful patients was selected from all patients who were successfully switched. The patients were matched on the basis of age, gender and the amount of time they had been treated with proton-pump inhibitors. There were 51 patients in each study group. The mean age was 62.2 (+/- 11.6) years in the successful group and 62.4 (+/- 12.5) years in the failure group. The proportion of men was 90.1% in both groups. The initial sample was appropriate for the study question since it included patients who had both successfully and unsuccessfully undergone the change from lansoprazole to omeprazole. There were no reports of any patients refusing to participate or being excluded for any reason.
Study design The study was a nested case-controlled study with groups defined by their successful or unsuccessful switch to omeprazole. The study was carried out in a single centre, the Southern Arizona Veterans Affairs (VA) Health Care System in Tucson (AZ), USA. Successful patients were followed for 6 months prior to the switch and 6 months after the switch. Unsuccessful patients were followed for 6 months prior to the switch, the period of the unsuccessful switch, and for 6 months after the return to omeprazole. No loss to follow-up was reported. The authors did not report whether there were any attempts to blind the clinicians or researchers.
Analysis of effectiveness The analysis was based on the patients' response to the switch. Response was considered successful or not successful. The primary outcomes were the numbers of successful and unsuccessful patients, and the reasons for failure. The authors also measured the severity and frequency of both heartburn and acid regurgitation. No significant differences in the characteristics of the two groups were reported.
Effectiveness results Among the 722 patients who were affected by the formulary change, 62 (8.6%) were unsuccessful. Sufficient data were available for 51 of these patients. The reasons for failure were poor symptom control on lansoprazole (37 patients, 72.5%), side effects that resolved after switching back (13 patients, 25.5%), and an allergy to lansoprazole (1 patient, 2%).
The severity of heartburn was 3.2 (+/- 1.2) during lansoprazole use and 1.2 (+/- 0.68) during the return to omeprazole, (p<0.001).
The frequency of heartburn was 1.8 (+/- 1.7) during lansoprazole use and 0.9 (+/- 3.8) during the return to omeprazole, (p<0.172).
The severity of acid regurgitation was 3.0 (+/- 1.3) during lansoprazole use and 1.2 (+/- 0.6) during the return to omeprazole, (p<0.001).
The frequency of acid regurgitation was 0.43 (+/- 1.19) during lansoprazole use and 0.09 (+/- 0.33) during the return to omeprazole, (p<0.001).
Clinical conclusions The authors did not draw any clinical conclusions independently to the cost conclusions. However, they did express a concern that failed patients may need to use additional health care resources. It was this concern that led to the current study.
Measure of benefits used in the economic analysis The authors did not estimate a summary measure of health benefit. Therefore, the study was effectively a cost-consequences analysis.
Direct costs A perspective for the costing analysis was not reported, but it would appear to have been that of the payer. As such, all the relevant costs were included. The authors did not report that discounting was carried out. However, as the study concerned costs incurred 6 months prior to the switch and 6 months after, the total time horizon was one year. Therefore, no discounting was required. Resource use data were collected from the patients' charts and the VA Information System and Technical Architecture database. The data included information on hospitalisations, outpatient medications, clinical visits, and laboratory and diagnostic studies. The unit costs were derived from the 1999 Health Care Financing Administration reimbursement rates for Tucson. Medications were priced at VA pricing rates. The authors estimated the difference between the 6-month pre- and 6-month post-switch costs for both groups, then compared the differences between the groups. The price year was 1999.
Statistical analysis of costs The cost data were compared using the Wilcoxon-signed rank test, to eliminate the impact of any skew in the data. The costs themselves were treated deterministically.
Indirect Costs The authors did not report estimating the indirect costs.
Currency The currency was not reported. However, it is likely to have been US dollars ($).
Sensitivity analysis No sensitivity analysis was reported.
Estimated benefits used in the economic analysis See the 'Effectiveness Results' section.
Cost results The change in the total monthly health care costs for the pre-post study period was -46.34 (+/- 266.64) for successful patients and -19.44 (+/- 256.81) for unsuccessful patients, (p=0.800). The currency was not reported. However, it is likely to have been US $.
The change in total monthly health care costs for the pre-during study period was -46.34 (+/- 266.64) for successful patients and -22.57 (+/- 250.61) for unsuccessful patients, (p=0.626).
The change in monthly gastrointestinal-related health care costs for the pre-post study period was -25.36 (+/- 128.62) for successful patients and -16.87 (+/- 59.04) for unsuccessful patients, (p=0.687).
The change in monthly gastrointestinal-related health care costs for the pre-during study period was -25.36 (+/- 128.62) for successful patients and -21.57 (+/- 77.37) for unsuccessful patients, (p=0.687).
The authors also reported that the formulary change resulted in savings of $300,000 in drug costs.
Synthesis of costs and benefits The costs and benefits were not combined.
Authors' conclusions The authors concluded "the formulary change was cost-efficient, at least in the short term".
CRD COMMENTARY - Selection of comparators The authors compared patients who successfully changed formulary from omeprazole to lansoprazole with patients who were unsuccessful. This paper lacked a specific hypothesis such as 'do the long and short run benefits of the formulary switch outweigh the long and short run costs of the switch' (which requires comparison before and after the drug change). This hypothesis is then a small aspect of the wider question - is lansoprazole more cost-effective than omeprazole? The comparison between successful and unsuccessful patients, although interesting, is then not relevant to answering this new study question. Success or not is relevant only to the question of whether patients likely to be successful can be identified before the treatment so that only these patients can undergo the switch. This is a clinical question and not one for economic evaluation, and certainly did not appear to be the authors' focus. At the outset of the study omeprazole represented standard practice in the authors' setting.
Validity of estimate of measure of effectiveness The analysis used a nested case-controlled study with equal proportions of successful and unsuccessful patients. However, over 91% of the patients actually successfully switched drug. To give a true picture of cost-effectiveness these proportions should have been reflected in the study sample. The patients groups were shown to be comparable at analysis, suggesting that no a priori distinction could be made between successful and unsuccessful patients. The time horizons for analysis for successful and unsuccessful patients were not equal, thus introducing potential bias when the two groups were compared on either cost or effectiveness grounds (the authors acknowledged this weakness).
It was unclear whether the adverse symptoms were withdrawal effects from lansoprazole, which may have subsided over time, or were actually due to the side effects of omeprazole. This distinction may significantly alter the conclusions of the study. Moreover, the poor health outcomes experienced by failed patients were at no point reported to have been balanced against potential improved outcomes from successful patients, especially in the context that over 91% of the patients were successful in the switch.
Validity of estimate of measure of benefit The authors did not derive a summary measure of health benefit. Therefore, the study was effectively a cost-consequences analysis.
Validity of estimate of costs The costs were estimated from the perspective of the payer and the costs relevant to this perspective were included. However, the authors focused on the costs of switching. These costs are transitory and play a small part in the overall costs of treating patients in the long run. Moreover, switching costs are not relevant for new patients. Therefore, the costs considered are a relatively minor part of the budgetary impact on the payer in the long run. Since there were no significant differences in cost between successful and unsuccessful patients, and there was little actual difference between pre- and post-switch costs, small omissions in the costs may greatly affect the principle results and conclusions. The unit costs and the quantities were not reported separately. Discounting was not relevant due to the timeframe of the study and, appropriately, was not conducted. The price year was reported, which will aid any future reflation exercises.
Other issues The authors made some appropriate comparisons of their results with those from other authors. The issue of generalisability to other settings was not addressed. The results seem to have been both observed and reported selectively, with no discussion of the outcomes for successful patients. The authors claimed that the formulary change was cost-efficient in the short term, yet their principle comparison and calculations of statistical significance concerned the difference between successful and unsuccessful patients, not pre- and post-switch costs and outcomes. A number of limitations were presented. For example, health-related quality of life was not measured and there was a lack of randomisation.
Implications of the study The authors did not make any recommendations for policy or practice as a result of their study. They proposed that future studies should examine health-related quality of life and proactive identification of failure patients.
Source of funding Supported in part by AstraZeneca.
Bibliographic details Raisch D W, Klaurens L M, Hayden C, Malagon I, Pulliam G, Fass R. Impact of a formulary change in proton pump inhibitors on health care costs and patients' symptoms. Digestive Diseases and Sciences 2001; 46(7): 1533-1539 Other publications of related interest Hatton BN, Gerson L, Jones W, et al. Assessment of patients at a VA medical centre with omeprazole due to lansoprazole failure. Gastroenterology 1999;116:G0797.
Indexing Status Subject indexing assigned by NLM MeSH 2-Pyridinylmethylsulfinylbenzimidazoles; Female; Formularies as Topic /standards; Gastroesophageal Reflux /drug therapy /etiology; Health Care Costs; Heartburn; Hospitals, Veterans; Humans; Lansoprazole; Male; Medical Records; Middle Aged; Omeprazole /analogs & Outcome Assessment (Health Care); Patient Satisfaction; Proton Pump Inhibitors; United States; derivatives /therapeutic use AccessionNumber 22001001507 Date bibliographic record published 31/07/2004 Date abstract record published 31/07/2004 |
|
|
|