Economic evaluation of three surgical interventions for menorrhagia

Size: px

Start display at page:

Download "Economic evaluation of three surgical interventions for menorrhagia"

Teresa Patrick
5 years ago
Views:

1 Human Reproduction Vol.18, No.3 pp. 583±587, 2003 DOI: /humrep/deg141 Economic evaluation of three surgical interventions for menorrhagia HerveÂ Fernandez 1,3, GiseÂla Kobelt 2 and AmeÂlie Gervaise 1 1 Service de GyneÂcologie-ObsteÂtrique, UniversiteÂ Paris-Sud, HoÃpital Antoine BeÂcleÁre (AP-HP), 157, Rue de la Porte de Trivaux, Clamart Cedex and 2 HDI France, 492 chemin des Laurens, SpeÂraceÁdes, France 3 To whom correspondence should be addressed. herve.fernandez@abc.ap-hop-paris.fr BACKGROUND: The study was carried out to compare the overall effectiveness and direct economic costs of vaginal hysterectomy (VH), endometrial ablation (EA) and thermo-coagulation (TC) for the treatment menorrhagia. METHODS: We treated 50, 50 and 47 women with menorrhagia (>150 points on the Higham pictorial chart) by VH, EA and TC respectively. The patients were treated consecutively by the same surgeon and the choice between the three procedures depended on the desire of the patients. Resource utilization for the interventions was collected retrospectively from the hospital charts. A study questionnaire was mailed to the patients 24±36 months after the primary surgery. Patients who reported that they had undergone a second procedure or who were still menorrhagic were considered as treatment failures. RESULTS: As expected, the failure rate was lowest for VH. The total cost (without re-intervention for persistent menorrhagia) was C= 5315 for VH, C= 1098 for EA and C= 921 for TC. The total cost with re-intervention was calculated based on therapeutic strategies used in 2001 and estimated at C= 5321 for VH, C= 1263 for EA and C= 1320 for TC. CONCLUSIONS: The two out-patient procedures are very comparable in terms of success rates and costs. Choices will depend on budgeting considerations, surgeon skill and patient preference. The results may give guidance to investment decisions. Key words: direct costs/endometrial ablation/hysterectomy/menorrhagia/thermo-coagulation Introduction Menorrhagia affects 20±30% of women with signi cant effects on their quality of life (Cooper et al., 1997). Medical treatment is usually the rst intent treatment but it reduces menstrual blood loss by only 50% (Preston et al., 1995). Up to 50% of women with menorrhagia undergo surgical treatment within 5 years of their referral to a gynaecologist (Coulter et al., 1991). Several different surgical interventions for treatment of severe menorrhagia are routinely performed. They differ in their effectiveness and the rate of post-surgical complications (Overton et al., 1997; NHSCRD, 1995), but also, as regards hysterectomy, in the burden of the procedure, in terms of surgery duration and hospital costs. In the current environment of limited resources, choices need to be made to allocate resources ef ciently, and different treatment alternatives should therefore be compared in terms of clinical effectiveness, costs, and estimated cost-effectiveness. The aim of this study was to compare vaginal hysterectomy, hysteroscopic endometrial ablation and endometrial thermocoagulation (Thermachoice; GynCare Inc, MenloPark, CA, USA) for the treatment of dysfunctional uterine bleeding and to evaluate the direct medical costs for each of these three different procedures. Materials and methods Patients Patients who had undergone one of the three procedures between 1995 and 1997 were selected from the patient charts in a hospital in the region of Paris (France). All patients that were selected had been operated on by the same surgeon, consecutively, during the inclusion period. All patients included had only menorrhagia evaluated on a pictorial chart (>150 for Higham score; Higham et al., 1990). Patients with myomas were excluded from this study. Every day of menstruation, the patients recorded the number of lightly, moderately or heavily soiled pads and the passage of clots and periods of ooding on a pictorial chart, which were then scored. The total numerical score for each pictorial chart showed the extent of menorrhagia. Usually a score >100 is taken to correlate with menorrhagia >80 ml. We included in this series patients with a score >150. Indeed, we decided to treat surgically only severe menorrhagia de ned by this score >150. These women had either failed medical therapy with progestins or were unwilling or unable to carry on with medical treatment with an anamnesis of severe menorrhagia over 1 year. The choice between the three procedures was based on patient preference: vaginal hysterectomy to obtain 100% of success and amenorrhoea, hysteroscopic endometrial ablation as usually per- ã European Society of Human Reproduction and Embryology 583

2 H.Fernandez, G.Kobelt and A.Gervaise Table I. Operating theatre staff (hospital perspective) Cost/hour C= Present in the operating theatre Thermo-coagulation Surgeon X X X Assistant (resident) X X Anaesthesiologist X X X Anaesthesiology nurse (8.38) 1 X X X Nurse (8.38) 1 X X X Nurse (8.23) 1 X X X Medical student (8.23) 1 X 1 Numbers in brackets represent unit cost based on national statistics rather than hospital accounting. formed in the department as primary conservative surgical treatment, endometrial thermo-coagulation when local anaesthesia was desired and when the device was in demonstration. Patients were then contacted 24±36 months later to collect data on the level of treatment success, on subsequent interventions, medical visits, and gynaecological problems if any, and to assess their satisfaction with the procedure. A questionnaire was mailed to 60 patients for each procedure and patients who had not answered after 1 month received a reminder. Resource utilization and costs Resource utilization for the interventions were collected from the hospital charts. These provided detailed data on primary intervention duration, occupation time of the operating theatre, staff present, product consumption and number of in-patient days. The number of out-patient visits after the intervention, which were related to the indication, were collected from patients. Resources were valued from the point of view of society, i.e. all resources were valued at their opportunity cost as far as possible, regardless of who ultimately paid for the resource. Unit costs for resources in the hospital were obtained from hospital accounting. The cost of out-patient visits was based on tariffs of the national health insurance. Drug treatment was excluded from the analysis, as no detailed data were available. The main analysis presents total costs based on hospitalization and out-patient procedures, including costs for re-interventions as reported by patients. As no detailed data on resource consumption for reinterventions were available, we based the cost on the average cost for out-patient procedures and the mean length of stay for in-patient procedures. We also estimated the cost of the procedures alone (excluding hospitalization) from the hospital perspective, based on the average duration of the intervention, operating theatre use and staff, sterilization, devices and materials and post-surgical visits. Table I shows the presence of different staff in the operating theatre as well as hourly staff costs. The rate for nurses obtained from the hospital administration appeared high compared with average costs in national accounting records. We therefore also present a sensitivity analysis using these national costs. Table II presents the cost-estimates for the procedures from the perspective of the hospital. Treatment outcome As this study was not based on a randomized clinical trial, effectiveness was estimated from patients' answers to the follow-up questionnaire. Treatment success was de ned as patients free of menorrhagia and not having undergone a second procedure: Patients 584 Table II. Intervention cost (Euro) Operating theatre (per min) Staff (per min) Surgeon (per min) Average surgical time (min) Average intervention time (min) Average time cost Sterilization cost ± Devices and materials Average total cost Sensitivity staff (per min) average total cost Indicative cost only; the actual duration of the intervention for each patient is used in the analysis. 2 Sensitivity using national average costs for nursing time (see text). who reported having undergone a second procedure, or who were still menorrhagic were considered as treatment failures. Satisfaction with the procedure was scored as 1 (fully satis ed), 2 (partially satis ed) and 3 (not at all satis ed). Ongoing pain (outside menstruation) was reported by patients as 1 for pain and 0 for no pain. This study was not set up to perform a cost-effectiveness analysis, particularly as outcome was assessed through patient questionnaires between 2 and 3 years after the intervention. Also, our scoring de nition of the satisfaction level was not based on a validated questionnaire. Lastly, the re-intervention rates after thermo-coagulation may re ect the introduction of a new procedure, rather than current clinical practice. For illustrative purposes, we present the incremental cost effectiveness for hysterectomy based on the total number of failures (re-interventions and persistent severe menorrhagia). Results Demographics patient sample A total of 147 patients were included in the study, of which 47 had a thermo-coagulation, 50 a vaginal hysterectomy and 50 an endometrial ablation. The mean age of patients undergoing a thermo-coagulation was 46.8 years (SD 6.2), of those undergoing vaginal hysterectomy 54.9 years (SD 10.1) and of those undergoing ablation 50.4 years (SD 7.66). It appears that in our hospital, thermo-coagulation is performed in younger patients than the other procedures.

3 Economic evaluation of surgery for menorrhagia Table III. Effectiveness of the procedures Table V. Mean cost per patient C= Outcome Resources Patients Initial re-hospitalization 0 8 (16%) 0 Re-intervention 8 (15.2%) 0 6 (12.0%) Complication 0 8 (16%) 0 Total failure (persistent menorrhagia, re-intervention for menorrhagia) 9 (19%) 0 10 (20%) Table IV. Patient satisfaction Outcome Patients Pain 2 (4.5%) 6 (12%) 12 (24.0%) Satisfaction fully 35 (74.5%) 44 (88%) 44 (88.0%) partially 5 (10.6%) 2 (4%) 3 (6.0%) not at all 7 (14.9%) 4 (8%) 3 (6.0%) Satisfaction score (SD 0.371) (SD 0.221) (SD 0.261) Effectiveness As expected, with our de nition of success and failure, the proportion of patients considered in our analysis as successfully treated was highest for hysterectomy, followed by thermo-coagulation and endometrial ablation (Table III). Signi cantly more patients reported pain after ablation and hysterectomy than after thermo-coagulation, but the overall satisfaction level was higher for ablation and hysterectomy (88% fully satis ed) than for thermo-coagulation (74.5% fully satis ed) (Table IV). Overall, in terms of clinical effectiveness, thermo-coagulation and endometrial ablation can be considered equal, and differences relate to costs rather than effectiveness. As expected hysterectomy is more effective in terms of bleeding control, but the procedure is far more costly and will not be the best choice for some patients. Economic analysis All patients undergoing hysterectomy were hospitalized and the mean length of stay (LOS) was 5.5 days. The corresponding gures for ablation were 14 (28%) patients (mean LOS 0.34 days) and for thermo-coagulation four (9%) patients (mean LOS 0.09 days). Re-hospitalizations were required for eight (16%) patients after hysterectomy, and re-interventions were required for six (12.0%) after ablation and eight (15.2%) after thermo-coagulation. It is, however, dif cult to compare these results, as all eight re-hospitalizations after hysterectomy were complications (pelvic abscess, retention of urine, vaginal haemorrhage, occlusion) treated by medical treatments without re-interventions. The mean operating time was ~2 h for hysterectomy, ~1 h for ablation and 46 min for thermo-coagulation. Patients who had undergone thermo-coagulation had the highest number of follow-up visits, 5.9, explained by the fact that the procedure Patients (n) Primary procedure only (excl. devices and materials) Primary procedure only (incl.devices and materials) Medical visits Total cost (without re-intervention) Total cost (with re-intervention) 1704 (1320) a a With therapeutic strategies used in was new, compared with 3.9 for ablation and 2.2 for hysterectomy. The total cost for hysterectomy was, as expected, higher than for the other two procedures, due to the need for all patients to be hospitalized. The two other procedures were predominantly out-patient procedures and hence less costly. When the procedures alone are considered, excluding devices, thermocoagulation had the lowest cost. When devices are included, ablation had the lowest cost. Hospitalization cost was lowest for thermo-coagulation, as only four of 47 patients required an overnight stay, compared with 14 of 50 patients for ablation. Patients undergoing thermocoagulation had slightly more follow-up visits than those undergoing ablation or hysterectomy. Also, the cost of re-interventions was higher for thermocoagulation than for ablation, as all seven re-interventions were hysterectomies, while the preferred intervention after failure of ablation was a second ablation. Thus, the total cost per patient was increased by C= 782 per patient for re-interventions after thermo-coagulation, but only by C= 165 after ablation (Table V). is more effective than either of the two outpatient procedures, but is considerably more expensive. The incremental cost per patient is C= 3617 compared with thermocoagulation and C= 4058 compared with endometrial ablation. The two out-patient procedures have basically the same cure rate (~80%), but within the conditions of this study, ablation had a lower cost (by C= 441), and would therefore be preferred on cost grounds. It has however to be borne in mind that the study represents experimental conditions for Thermachoice and that therefore the cost of re-intervention may be overstated. The incremental effectiveness of hysterectomy compared with ablation (i.e. the proportion of additional patients that are cured) is 20%. Thus, the incremental cost-effectiveness (i.e. the cost per additional patient cured with hysterectomy compared with ablation) is high, C= Discussion New surgical procedures should be assessed for their cost and their effectiveness, before a decision regarding their value can be reached. Often, new procedures include new devices, generally single-use devices, which increase the initial cost of 585

4 H.Fernandez, G.Kobelt and A.Gervaise the intervention. A good example of this is laparoscopic surgery (Simon et al., 1999). However, the acquisition cost of any device must be judged by including all other costs, such as the staff and time required, patient management such as hospitalization, consultations and other related costs that may differ from other procedures. Most importantly, however, the success rates of the procedures must be compared, and the cost of treating failures (e.g. with a second intervention) included. This study illustrates very well the importance of such an analysis. Thermo-coagulation requires a rather costly singleuse device, which makes the procedure expensive compared with, for example, ablation. However, the procedure is less resource-intensive and overnight stays are rarely required. Thus, when all costs relating to the primary procedure are included, thermo-coagulation has the lowest mean cost per patient (C= 922) followed by C= 1098 for ablation and by C= 5315 for hysterectomy. This ranking is changed when reinterventions or re-hospitalizations are included, and ablation has the lowest costs (C= 1263) followed by C= 1704 for thermo-coagulation and C= 5321 for hysterectomy. The number of successes (no eumenorrhoea or less or no reintervention for persistent bleeding) were, as expected, similar for thermo-coagulation and ablation, ~80%. However, all patients, where thermo-coagulation was not successful, underwent hysterectomy, leading to a very high cost of reintervention. As the procedure was new when this study was performed, it could be argued that as the procedure becomes more common, some patients may undergo a second thermocoagulation when the rst intervention fails. This hypothesis is supported by the pattern of re-interventions after ablation, where three of the six patients who required a re-intervention, had a second ablation. The novelty of thermo-coagulation could explain the nding that only 74.5% of patients were fully satis ed by the procedure in spite of the fact that the pain reported was lower. Moreover, the subjectivity of pain might also explain the difference between the observed results. Under such an assumption, where only 40±50% of the failures would be treated with a hysterectomy, (compared with only 1 in 6 after ablation), the total costs for thermocoagulation is reduced to a similar level as ablation, i.e. C= Thus, the two out-patient procedures would be equivalent both in terms of cost and effectiveness. Overall satisfaction appeared slightly lower for thermocoagulation, which was a new surgical technique in 1995 but far more patients reported pain after ablation. Thus, the two out-patient procedures are very comparable in terms of success rates, and choices will depend on budgetary considerations and patient preference. This analysis has several shortcomings that must be considered. First, the database includes interventions performed in the years 1995±1997 and patient management patterns may have changed since then, particularly as far as inpatient stays are concerned. Second, the data concerning reinterventions and medical visits have been obtained from patients 2±3 years after the procedure and could therefore include a large recall bias. Third, it was not possible to include all costs relating to the procedures, such as follow-up 586 treatments nor the cost of treating complications, which are more frequent after hysterectomy. Fourth, information on sick leave and other indirect costs of the treatments were not available and the analysis will thus underestimate the costs from a societal perspective, as patients in this cohort were rather young, particularly patients in the thermo-coagulation group. Lastly, the signi cant difference in age between the three groups introduces a bias in success rate estimation because the younger patients of the thermo-coagulation group have a longer time to menopause and thus a longer exposure to relapse risk than in the other two groups. An adjustment for age should be performed. However, the key issues for a comparison of different procedures (success and failure rates), as well as the main costs (in-patient and procedure costs) may give some guidance to investment decisions. Moreover, not all gynaecologists will be able, for various reasons, to acquire skill in all procedures designed to treat menorrhagia. However, skill-intensive procedures such as vaginal hysterectomy or hysteroscopic endometrial ablation may well be replaced by the second generation of endometrial ablation such as Thermachoice that appears to require minimal training and experience for effective outcomes, less complications and obviously less cost. This study reinforces the economic information included in trials comparing hysterectomy with endometrial ablation (Gannon et al., 1991; Dwyer et al., 1993; Sculpher et al., 1993, 1996; Pinion et al., 1994; Croscignani et al., 1997; Research Council, 1997). Although our analysis cannot be directly applied to health care systems of other countries, it appears that in view of our results conservative out-patient surgery should be the preferred option in most cases. Acknowledgements The authors wish to thank Laurent Metz, M.D. (Ethicon SA) for the support brought to this health economic study. References Cooper, K.G., Parkin, D.E., Garratt, A.M. and Grant, A.M. (1997) A randomised comparison of medical and hysteroscopic management in women consulting a gynaecologist for treatment of heavy menstrual loss. Br. J. Obstet. Gynaecol., 104, 1360±1366. Coulter, A., Bradlow, J., Agass, M., Martin-Bates, C. and Tulloch, A. (1991) Outcome of referrals to gynaecology outpatient clinics for menstrual problems: an audit of general practice records. Br. J. Obstet. Gynaecol., 98, 789±796. Croscignani, P.G., Vercellini, P., Apolone, G., De Giorg, O., Cortesi, I. and Meschia, M. (1997) Endometrial resection versus vaginal hysterectomy for menorrhagia: long term clinical and quality of life outcomes. Am. J. Obstet. Gynaecol. 177, 95±101. Dwyer, N., Hutton, J. and Stirrat, G.M. (1993) Randomised controlled trial comparing endometrial resection with abdominal hysterectomy for the surgical treatment of menorrhagia. Br. J. Obstet. Gynaecol., 100, 237±243. Gannon, M.J., Holt, E.M., Fairbank, J., Fitzgerald, M., Milne, M.A. and Crystal, A.M. (1991) A randomised trial comparing endometrial resection and abdominal hysterectomy for the treatment of menorrhagia. BMJ, 303, 1362±1364. Higham, J.M., O'Brien, P.M.S. and Shaw, R.W. (1990) Assessment of menstrual blood using a pictorial chart. Br. J. Obstet. Gynaecol., 9, 734±739. NHSCRD, (1995) The management of menorrhagia what are effective ways of treating excessive regular menstrual blood loss in primary and secondary care? Effect. Health Care Bull. 9, 15p.

5 Economic evaluation of surgery for menorrhagia Overton, C., Hargreaves, J. and Maresh, M. (1997) A national survey of the complications of endometrial destruction for menstrual disorders: the MISTLETOE study. Br. J. Obstet. Gynaecol., 104, 1351±1359. Pinion, S.B., Parkin, D.E. and Abramovich, D.R. et al. (1994) Randomised trial of hysterectomy, endometrial laser ablation, and transcervical endometrial resection of dysfunctional uterine bleeding. BMJ, 309, 979±983. Preston, J.T., Cameron, I.T., Adams, E.J. and Smith, S.K. (1995) Comparative study of tranexamic acid and norethisterone in the treatment of ovulatory menorrhagia. Br. J. Obstet. Gynaecol., 102, 401±406. Research Council (1997) Randomised trial of endometrial resection versus hysterectomy in the management of menorrhagia. Lancet, 349, 897±901. Sculpher, M.J., Bryan, S., Dwyer, N., Hutton, J. and Stirrat, G.M. (1993) An economic evaluation of transcervical endometrial resection versus abdominal hysterectomy for the treatment of menorrhagia. Br. J. Obstet. Gynaecol., 100, 244±252. Sculpher, M.J., Dwyer, N., Byford, S. and Stirrat, G.M. (1996) Randomised trial comparing hysterectomy and transcervical endometrial resection: effect on health related quality of life and costs two years after surgery. Br. J. Obstet. Gynaecol., 103, 142±149. Simon, N.V., Laveran, R.L., Cavanaugh, S., Gerlach, D.H. and Luckson, J.R. (1999) Laparoscopic supracervical hysterectomy vs abdominal hysterectomy in a community hospital: a cost comparisons. J. Reprod. Med., 44, 339±45. Submitted on June 10, 2002; resubmitted on September 24, 2002; accepted on December 3,

REPRODUCTIVE ENDOCRINOLOGY

REPRODUCTIVE ENDOCRINOLOGY FERTILITY AND STERILITY VOL. 82, NO. 1, JULY 2004 Copyright 2004 American Society for Reproductive Medicine Published by Elsevier Inc. Printed on acid-free paper in U.S.A. Treatment