Objective
Menorrhagia is a common problem impairing the quality of life (QOL) of many women. Both levonorgestrel-releasing intrauterine system (LNG-IUS) and hysterectomy are effective treatment modalities but no long-term comparative studies of QOL and costs exist. The objective of this study was to compare QOL and costs of LNG-IUS or hysterectomy in the treatment of menorrhagia during 10-year follow-up.
Study design
A total of 236 women, aged 35-49 years, referred for menorrhagia to 5 university hospitals in Finland were randomly assigned to treatment with LNG-IUS (n = 119) or hysterectomy (n = 117) and were monitored for 10 years. The main outcome measures were health-related QOL (HRQOL), psychosocial well-being, and cost-effectiveness.
Results
A total of 221 (94%) women were followed for 10 years. Although 55 (46%) women assigned to the LNG-IUS subsequently underwent hysterectomy, the overall costs in the LNG-IUS group ($3423) were substantially lower than in the hysterectomy group ($4937). Overall, levels of HRQOL and psychosocial well-being improved during first 5 years but diminished between 5 years and 10 years and the improved HRQOL returned close to the baseline level. There were no significant differences between LNG-IUS and hysterectomy groups.
Conclusion
Both LNG-IUS and hysterectomy improved HRQOL. The improvement was most striking during the first 5 years. Although many women eventually had hysterectomy, LNG-IUS remained cost-effective.
Nearly one-third of women have heavy menstrual bleeding at some point during their reproductive years. Menorrhagia markedly interferes with daily activities and impairs the quality of life (QOL). Menorrhagia causes iron deficiency anemia, substantial work loss, and significant health care costs.
Hysterectomy has been the preferred treatment option for menorrhagia. However, the associated surgical risks, adverse long-term effects, and substantial costs can not be ignored. Less invasive alternatives to hysterectomy such as levonorgestrel-releasing intrauterine system (LNG-IUS) (Mirena; Bayer Oy, Turku, Finland) and endometrial ablation (EA) have become increasingly popular, reducing the need for hysterectomy. LNG-IUS and EA have similar effects on patient satisfaction and health-related QOL (HRQOL).
The majority of the studies comparing LNG-IUS and hysterectomy are modeling studies analyzed by Markov model or metaanalyses. Most, but not all, of these studies demonstrate LNG-IUS as the most cost-effective treatment option. Two recent reviews of menorrhagia treatment advocate the use of LNG-IUS. We conducted a randomized controlled trial comparing LNG-IUS and hysterectomy in the treatment of menorrhagia and showed that LNG-IUS is more cost-effective than hysterectomy after 1 year and 5 years, with no difference in patient satisfaction or HRQOL. We now report the results of clinical outcomes, HRQOL, psychosocial well-being, and costs after 10-year follow-up.
Materials and Methods
Trial design
Of 598 women referred for menorrhagia to the 5 university hospitals in Finland from 1994 through 1997, 236 were eligible for the study and agreed to participate ( Appendix ; Supplementary Figure ). Women were randomly assigned to treatment with LNG-IUS (n = 119) or hysterectomy (n = 117) ( Figure 1 ). The women were 35-49 years old (mean, 43.1 years; median, 43.0 years), were menstruating, had completed childbearing, and were eligible for both treatments.
LNG-IUS was inserted during the randomization visit. Hysterectomy was performed abdominally, vaginally, or laparoscopically at the physician’s discretion. The follow-up visits took place 6 and 12 months after the treatment, and again 5 and 10 years after the randomization. Questionnaires were completed before the randomization and at each follow-up visit. Participants completed a questionnaire at home containing HRQOL instruments and questions on health care use, sick-leave days, and travel costs. Study gynecologists completed a form containing detailed data on symptoms and signs.
The study was approved by ethics committees of all university hospitals and the National Research and Development Center for Health and Welfare in Finland (ethics reference no. 249/E8/04; date of approval Aug. 26, 2004; ethical approval given from the Institutional Review Board of the Helsinki University Central Hospital).
All participants provided written informed consent.
Health-related QOL
The 5-Dimensional EuroQol (EQ-5D) was selected as the primary measure of effectiveness. The EQ-5D includes 5 3-level dimensions: morbidity, self-care, usual activities, pain, and mood and has been validated in the Finnish general population. The validated Finnish version of the RAND 36-Item Health Survey (RAND-36) including 8 multi-item dimensions was also used. The general health assessment was recorded by a visual analog scale.
Other psychosocial measures
Anxiety was measured by the validated Finnish version of Spielberger 20-Item State-Trait Anxiety Inventory. Depression was measured by the 13-item version of the Beck Depression Inventory. Sexuality-related factors were assessed by the McCoy Sex Scale including 3 subscales, as modified by Wiklund.
Cost analysis
Data on direct costs, including use of hospital services and on indirect costs including sick-leave days as markers of productivity losses, were obtained from medical records and questionnaires. Information on other appointments for menorrhagia, Papanicolaou tests, and out-of-pocket costs related to menorrhagia was obtained from questionnaires. The costs of hospital services were obtained through 10 years, whereas the costs of health care out-of-hospital, out-of-pocket costs, and indirect costs were obtained from first and fifth study years and from 5-10 years.
A pricing system based on diagnostic related groups used by the Helsinki University Central Hospital was used for pricing the hospital procedures. The first-year costs were based on 1996 price levels, the costs over 1-5 years on 2001 price levels, and the costs over 5-10 years on 2006 price levels. The unit cost of hysterectomy included 1 preoperative visit, the actual operation, and 1-5 inpatient days ($1864 in 1996, $2055 in 2001, and $3187 in 2006). If longer stay in hospital was needed, additional days were priced according to the average bed day price ($247 in 1996, $297 in 2001, and $363 in 2006). The productivity loss per sick-leave day was defined as the average daily gross wage of women in Finland including social security ($71, $85, and $142). The costs were discounted by the commonly recommended rate of 3% per year to 1996 level (average year for treatment decisions). The currency conversion had its basis in purchasing power parities in 1996 (US$1 = FIM 5.89) and in 2002, when Finland joined the euro (FIM 1 = EUR €0.168).
We assessed uncertainty in the model parameters by sensitivity analysis with different discount rate, a lower estimate of productivity loss, different prices of hysterectomy, and without costs of complications of hysterectomies.
Cost-utility analysis
The additional quality-adjusted life years (QALYs) gained by the treatments were calculated by using 12-month, 5-year, and 10-year values of EQ-5D, assuming that without treatment the level of HRQOL would have remained on baseline level over 10 years.
Laboratory investigation
Menstrual blood loss was measured before the randomization, 12 months later, and 5 years later by the alkaline hematin method. At 10 years, none of the women provided menstrual blood loss samples. Blood hemoglobin concentrations were measured using a Coulter Counter T660 (Coulter Electronics Ltd, London, United Kingdom).
Statistical analyses
All analyses were performed according to intention to treat using the statistical software SPSS 19.0 (IBM, Armonk, NY) and STATA 11.1 (StataCorp, College Station, TX). Power calculation was based on EQ-5D SD of 19% (derived from a Finnish 34- to 49-year-old female population ) and α level of 0.05, resulting in a target of 115 patients in each treatment group. The study had 80% power to detect 7.5% difference between the groups. Student t test for paired samples and Wilcoxon signed rank test were used to test changes in outcome measures within the groups before and after treatment and between follow-ups. Student t test for independent samples was used to test differences in score changes between the groups. P values ≤ .05 were considered statistically significant.
To assess the total change trajectories of HRQOL and psychosocial outcomes over the 10-year period, we applied random-intercept multilevel longitudinal modeling by pooling all the 4 measurement times (baseline, 1-year, 5-year, 10-year follow-ups) in a single analysis. The method does not require full data for all measurement time points, so all participants and their available person-observations of the 4 measurement times were included in this analysis.
Results
The study groups were comparable except that women assigned to LNG-IUS had a slightly higher body mass index at baseline ( Table 1 ). Of the 236 randomized women, 221 attended the 10-year visit. At 10 years 12 women had withdrawn and 3 women had died. Thus, the dropout rate was 6%.
| Charactertistic | LNG-IUS (n = 119) | Hysterectomy (n = 117) |
|---|---|---|
| Age, y | 43.1 (3.5) | 43.1 (3.2) |
| BMI, kg/m² | 26.6 (5.1) a | 25.1 (4.5) |
| Parity | 2.1 (1.0) | 2.0 (1.2) |
| Smoker | 27 (26%) | 38 (33%) |
| Menstrual blood loss, mL | 130 (116) | 128 (116) |
| Hemoglobin, g/L | 127 (13) | 125 (12) |
| Education | ||
| Elementary | 39 (32%) | 33 (28%) |
| Lower secondary | 52 (44%) | 47 (40%) |
| Upper secondary | 28 (24%) | 37 (32%) |
| Unemployed | 10 (8) | 8 (7%) |
LNG-IUS arm
Of the 119 women randomized to the treatment with LNG-IUS, 110 (92%) attended the 10-year visit, 8 withdrew, and 1 died ( Figure 1 ). In all, 55 (46%) of the women had hysterectomy, 44 (37%) had LNG-IUS in situ, 21 had no LNG-IUS or hysterectomy, and data were missing for 1 woman ( Figure 1 ). Of the 55 hysterectomies performed in the LNG-IUS group, 24 (44%) were performed during the first year, 26 (47%) between 12 months and 5 years, and 5 (9%) between 5-10 years. Of the hysterectomies, 13 were performed vaginally, 10 abdominally, and 32 laparoscopically. Eight women had bilateral salpingo-oophorectomy with hysterectomy. Perioperative and postoperative complications occurred in 15 (27%) of the 55 women who had hysterectomy. All 5 hysterectomies performed between 5-10 years were because of fibroids and bleeding problems. The main reason for removal of LNG-IUS after 5 years was menopause. The reasons for hysterectomies and LNG-IUS removals during the first 5 years have been reported previously. The main reason was bleeding problems.
Of the 44 women with LNG-IUS in situ at 10 years, 40 (91%) reported amenorrhea or oligomenorrhea, 2 reported hypomenorrhea, 1 reported normal menstrual bleeding, and 1 reported irregular bleeding. Of the 18 women who had LNG-IUS removed but no hysterectomy, 12 reported amenorrhea or oligomenorrhea, 3 reported normal menstrual bleeding, 2 reported irregular bleeding, and 1 woman who had thermoablation reported hypomenorrhea. None of the women reported heavy menstrual bleeding.
Hysterectomy arm
Of the 117 women randomized to hysterectomy, 111 attended the 10-year visit, 4 withdrew, and 2 died ( Figure 1 ). Among the 117 women, 109 had undergone hysterectomy (2 of whom had died), 8 had no hysterectomy, and 1 had LNG-IUS in situ. Of the 109 hysterectomies, 30 (28%) were performed vaginally, 22 (20%) abdominally, and 57 (52%) laparoscopically. Intraoperative or postoperative complications occurred in 37 (34%) women. During 10-year follow-up, 8 women had bilateral salpingo-oophorectomy with hysterectomy. Seven of the 8 women with no hysterectomy reported amenorrhea or oligomenorrhea. One woman reported normal menstrual bleeding.
HRQOL and other psychosocial outcomes
HRQOL and psychosocial well-being improved significantly during the first year and 5 years after the treatment, with no significant difference between the treatment groups (except that women with hysterectomy experienced less pain at 12 months).
At 10 years EQ-5D scores remained at baseline level with no difference between the groups. Between baseline and 10 years many of the dimensions of HRQOL measured by the RAND-36 improved, especially in the LNG-IUS group ( Table 2 , Figure 2 ).
| Variable | Baseline | Change from baseline to 10-y follow-up | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| LNG-IUS (n = 119) | Hysterectomy (n = 117) | LNG-IUS (n = 110) | Hysterectomy (n = 111) | P value e | |||||||
| Mean | (95% CI) | Mean | (95% CI) | Mean | (95% CI) | P value d | Mean | (95% CI) | P value d | ||
| EQ-5D (scale range, 0–1) a | 0.76 | (0.72–0.79) | 0.78 | (0.75–0.81) | −1.1 | (−0.05 to 0.03) | .67 | −0.01 | (−0.05 to 0.03) | .60 | .94 |
| RAND-36 (scale range, 0–100) a | |||||||||||
| General health | 64.0 | (60.7–67.3) | 65.0 | (61.2–68.8) | −2.3 | (−5.8 to 1.2) | .20 | −4.5 | (−8.3 to −0.8) | .02 f | .39 |
| Physical functioning | 83.0 | (79.5–86.6) | 83.9 | (80.7–87.0) | −3.4 | (−7.5 to 0.8) | .11 | −3.8 | (−8.0 to 0.4) | .07 | .88 |
| Emotional well-being | 66.9 | (63.2–70.5) | 69.7 | (66.3–73.1) | 5.7 | (1.3–10.1) | .01 f | 3.2 | (−0.7 to 7.0) | .11 | .40 |
| Social functioning | 71.9 | (67.6–76.2) | 75.6 | (71.9–79.4) | 7.9 | (2.3–13.4) | .01 f | 1.8 | (−3.3 to 7.0) | .49 | .12 |
| Energy | 55.3 | (51.1–59.4) | 57.1 | (53.3–61.0) | 6.0 | (1.7–10.3) | .01 f | 5.3 | (0.6–10.0) | .03 f | .83 |
| Pain | 62.7 | (58.2–67.1) | 61.8 | (57.6–66.1) | 4.4 | (−0.4 to 9.2) | .07 | 4.0 | (−2.1 to 10.0) | .20 | .91 |
| Role functioning | |||||||||||
| Physical | 64.6 | (57.5–71.8) | 65.9 | (58.8–73.1) | 8.2 | (−0.53 to 16.9) | .07 | 3.2 | (−5.7 to 12.2) | .48 | .4 |
| Emotional | 61.5 | (54.1–69.0) | 66.1 | (58.9–73.3) | 9.1 | (−1.4 to 19.6) | .09 | 4.9 | (−5.1 to 14.1 | .33 | .57 |
| General health (VAS 0–100) a | 73.5 | (70.2–76.9) | 75.2 | (72.1–78.3) | −4.4 | (−7.7 to −1.1) | .01 f | −7.4 | (−12.1 to −2.7) | < .001 f | .32 |
| Anxiety (STAI [scale range, 20–80]) b | 32.5 | (31.3–33.7) | 31.4 | (30.4–32.5) | −2.2 | (−3.4 to −1.0) | < .001 f | −1.2 | (−2.5 to 0.1) | .07 | .3 |
| Depression (BDI [scale range, 0–39]) b | 5.1 | (4.2–6.1) | 4.2 | (3.5–5.0) | −0.5 | (−1.4 to 0.5) | .33 | 0.12 | (−0.9 to 1.1) | .82 | .40 |
| Sexuality (MSS) | |||||||||||
| Sexual satisfaction (subscale range, 5–35) c | 23.8 | (22.7–24.9) | 23.6 | (22.8–24.5) | −1.6 | (−2.7 to −0.4) | .009 f | −1.4 | (−2.5 to 3.8) | .01 f | .86 |
| Sexual problems (subscale range, 2–14) b | 4.4 | (4.0–4.8) | 4.5 | (4.1–4.9) | 1.2 | (0.6–1.9) | < .001 f | 0.7 | (0.17–1.3) | .01 f | .25 |
| Satisfaction with partner (subscale range, 3–21) c | 16.5 | (15.8–17.1) | 16.5 | (16.0–17.1) | −1.3 | (−2.1 to −0.5) | < .001 f | −0.7 | (−1.4 to −0.1) | .02 f | .30 |
| 5-y follow-up | Change from 5-y follow-up to 10-y follow-up | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| LNG-IUS (n = 117) | Hysterectomy (n = 115) | LNG-IUS (n = 110) | Hysterectomy (n = 111) | P value e | |||||||
| Mean | (95% CI) | Mean | (95% CI) | Mean | (95% CI) | P value d | Mean | (95% CI) | P value d | ||
| EQ-5D (scale range, 0–1) a | 0.81 | (0.77–0.85) | 0.85 | (0.82–0.89) | −0.07 | (−0.10 to −0.03) | < .001 f | −0.08 | (−0.1 to −0.04) | < .001 f | .57 |
| RAND-36 (scale range, 0–100) a | |||||||||||
| General health | 66.7 | (63.0–70.5) | 69.3 | (65.5–73.2) | −6.0 | (−9.2 to −2.8) | < .001 f | −8.7 | (−12.3 to −5.1) | < .001 f | .27 |
| Physical functioning | 85.4 | (81.2–89.6) | 86.3 | (83.0–89.5) | −5.9 | (−9.2 to −2.6) | < .001 f | −5.6 | (−9.4 to −1.7) | .01 f | .89 |
| Emotional well-being | 75.0 | (71.6–78.4) | 77.5 | (74.8–80.3) | −3.2 | (−6.6 to 0.2) | .06 | −4.7 | (−7.9 to −1.5) | < .001 f | .52 |
| Social functioning | 81.5 | (76.5–86.4) | 84.9 | (80.8–89.0) | −3.2 | (−8.6 to 2.3) | .26 | −6.3 | (−11.3 to −1.2) | .02 f | .41 |
| Energy | 64.9 | (60.7–69.0) | 66.5 | (62.9–70.2) | −4.7 | (−8.7 to −0.7) | .02 f | −3.8 | (−8.0 to 0.4) | .08 | .76 |
| Pain | 76.1 | (71.3–80.8) | 75.6 | (70.6–80.7) | −10.0 | (−14.2 to −5.8) | < .001 f | −9.2 | (−14.7 to −3.8) | < .001 f | .84 |
| Role functioning | |||||||||||
| Physical | 75.2 | (68.1–82.3) | 77.0 | (70.4–83.6) | −5.9 | (−13.1 to −1.2) | .10 | −5.7 | (−14.2 to 2.8) | .19 | .97 |
| Emotional | 78.1 | (71.6–84.5) | 78.9 | (72.3–85.4) | −8.8 | (−16.9 to −0.7) | .03 f | −6.0 | (−12.8 to 0.7) | .08 | .60 |
| General health (VAS 0–100) a | 74.4 | (70.9–70.0) | 78.7 | (75.6–81.8) | −5.6 | (−9.4 to −1.7) | .01 f | −9.9 | (−14.6 to −5.2) | < .001 f | .16 |
| Anxiety (STAI [scale range, 20–80]) b | 30.1 | (28.9–31.3) | 29.7 | (28.6–30.8) | 0.9 | (−0.2 to 2.1) | .10 | 0.5 | (−0.9 to 1.9) | .47 | .65 |
| Depression (BDI [scale range, 0–39]) b | 4.0 | (3.1–4.9) | 3.0 | (2.3–3.7) | 0.9 | (0.2–1.6) | .02 f | 1.4 | (0.5–2.2) | < .001 f | .40 |
| Sexuality (MSS) | |||||||||||
| Sexual satisfaction (subscale range, 5–35) c | 22.8 | (21.8–23.9) | 24.0 | (23.0–25.1) | −1.0 | (−2.0 to −0.1) | .04 f | −1.0 | (−2.2 to 0.2) | .10 | .99 |
| Sexual problems (subscale range, 2–14) b | 4.4 | (3.9–4.9) | 4.4 | (3.9–4.9) | 1.2 | (0.6–1.9) | < .001 f | 0.5 | (−0.1 to 1.2) | .09 | .15 |
| Satisfaction with partner (subscale range, 3–21) c | 15.3 | (14.5–16.1) | 15.9 | (15.3–16.6) | −0.4 | (−1.2 to 0.4) | .3 | −0.2 | (−0.8 to 0.5) | .64 | .61 |
a Higher scores indicate better health-related quality of life
b Higher scores indicate more symptoms or problems
c Higher score indicate more satisfaction
d Within group difference (Student t test for paired samples)
e Between group difference (Student t test for independent samples).
In both groups sexual satisfaction and satisfaction with partner decreased and sexual problems increased ( Table 2 ). HRQOL declined by most measures between 5-10 years ( Table 2 , Figure 2 ). No significant differences were found between the study groups between baseline and 10 years or between 5-10 years ( Table 2 ).
The multilevel models suggested fairly steep improvement within the first year but then declining levels of HRQOL during the rest of the follow-up ( Figure 2 ). Despite this decline the average levels of HRQOL tended to be higher after 10 years than at baseline, although not significantly ( Table 2 ). The LNG-IUS and hysterectomy groups followed similar trajectories, and the few significant differences at the first year mostly resulted from small baseline differences combined with nonsignificant differences in change between baseline and 12 months’ follow-up. Most importantly, there were no differences between LNG-IUS and hysterectomy groups over time ( P values > .07) ( Appendix ; Supplementary Table ).
Laboratory results
A significant increase in blood hemoglobin concentration occurred in both study groups. In the LNG-IUS group hemoglobin concentration increased from 126.5 g/mL (SD 12.5) at baseline to 137.9 g/mL (SD 9.3) at 5 years, and 140.4 g/mL (SD 8.9) at 10 years and in the hysterectomy group from 124.5 g/mL (SD 12.0) to 134.5 g/mL (SD 8.4) and 137.8 g/mL (SD 10.8) with no difference between the study groups.
Costs and cost-utility outcomes
Both discounted direct costs and discounted indirect costs were significantly lower in the LNG-IUS group than in the hysterectomy group ( Table 3 ). The discounted total cost per participant was $3423 in the LNG-IUS group and $4937 in the hysterectomy group. The total costs per participant in sensitivity analyses were again lower in the LNG-IUS group than in the hysterectomy group ( Table 3 ). LNG-IUS produced 0.45 QALYs, and the incremental cost-effective ratio (ICER) was $7607. Hysterectomy resulted in 0.51 QALYs (ICER $9680).
| Variable | Unit cost, US$ | LNG-IUS | Hysterectomy | ||||
|---|---|---|---|---|---|---|---|
| Cost component | 1996 | 2001 | 2006 | n over 10 y (5-10 y) | Cost, US$ | n over 10 y (5-10 y) | Cost, US$ |
| Direct costs | |||||||
| LNG-IUS | |||||||
| First | 185 | 165 | 117 | 21,675 | 2 | 331 | |
| Reinserted | 185 | 165 | 216 | 84 (26) | 15,307 | 1 (1) | 216 |
| Hysterectomy (includes 1-5 inpatient days) | 1864 | 2055 | 3187 | 55 (5) | 114,102 | 109 | 203,601 |
| Extra inpatient days | 247 | 297 | 363 | 21 (1) | 6058 | 45 | 11,097 |
| Relaparoscopy | 1502 | 1569 | 1 | 1569 | |||
| Readmissions because of complication | |||||||
| Infection (inpatient days) | 247 | 297 | 10 | 2668 | 35 | 8631 | |
| Urinary retention (inpatient days) | 247 | 297 | 3 | 740 | |||
| Intestinal occlusion (inpatient days) | 247 | 297 | 12 | 3569 | 17 | 4886 | |
| Secondary hemorrhagia (operation) | 1527 | 2 | 3054 | ||||
| Laparoscopy because of pain | 1475 | 1 | 1475 | ||||
| Suture of Ileum (with 11 d in intensive care) | 1 | 11,000 | |||||
| Laparotomy because of occlusion | 1 | 3102 | |||||
| Nephrostoma (with 2 inpatient days) | 1 | 1273 | |||||
| Ureterneocystostomia and oophorectomy (with 8 inpatient days) | 1 | 5494 | |||||
| Curettage/hysteroscopy | 542 | 798 | 823 | 8 (2) | 5921 | ||
| Thermoablation | 1225 | 1 | 1225 | ||||
| Laparoscopic oophorectomy for ovarian cyst | 1475 | 1503 | 5 | 7429 | 3 | 4508 | |
| Abdominal salpingo-oophorectomy for ovarian cyst | 3187 | 1 (1) | 3187 | ||||
| Outpatient visits (controls a and complications) | 110 | 124 | 253 | 775 (123) | 105,621 | 713 (113) | 96,729 |
| Health care use out of hospital | |||||||
| Visits to general practitioner at health center | 46 | 58 | 82 | 62 (33) | 4186 | 18 (10) | 1186 |
| Visits to private physician | 27 | 40 | 54 | 10 (8) | 486 | 9 (8) | 459 |
| Visits to private gynecologist | 42 | 53 | 112 | 46 (16) | 3355 | 22 (8) | 1625 |
| Papanicolaou test | 31 | 20 | 263 (180) | 6137 | 185 (146) | 4112 | |
| Out-of-pocket costs | |||||||
| Medication | 1033 | 3265 | |||||
| Travel | 552 | 986 | |||||
| Indirect costs | |||||||
| Sick-leave days | 71 | 85 | 142 | 1718 (234) | 148,975 | 3080 (30) | 224,719 |
| Summary of health care costs | |||||||
| Direct costs | |||||||
| Total health care costs | 300,893 | 370,657 | |||||
| Discounted total health care costs per participant b | 2291 | 3036 | |||||
| Indirect costs | |||||||
| Discounted productivity losses per participant b | 1133 | 1900 | |||||
| Total costs | 449,868 | 595,376 | |||||
| Total costs per participant | 3780 | 5089 | |||||
| Discounted total costs b | 407,355 | 577,573 | |||||
| Discounted total costs per participant b | 3423 | 4937 | |||||
| Sensitivity analysis | |||||||
| Total cost per participant, US$ | |||||||
| Variable used in analysis | LNG-IUS | Hysterectomy | |||||
| Base case b | 3423 | 4937 | |||||
| Discount rates for costs | |||||||
| No discounting | 3780 | 5089 | |||||
| Discount rate 5% | 3247 | 4862 | |||||
| Productivity loss (indirect cost) | |||||||
| Lower estimate c | 2668 | 3669 | |||||
| Cost of hysterectomy | |||||||
| Lower estimate d | 3324 | 4590 | |||||
| Mean hysterectomy cost in US in 1996 e | 3861 | 6917 | |||||
| Laparoscopic hysterectomy cost in US in 2007 f | 4876 | 8663 | |||||
| Costs of complications excluded | 3360 | 4608 | |||||
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
