Cost-utility analysis of great saphenous vein ablation with radiofrequency, foam and surgery in the emerging health-care setting of Thailand - PDF Document

Presentation Transcript

  1. Original Article Phlebology 2016, Vol. 31(8) 573–581 ! The Author(s) 2015 Reprints and permissions: DOI: 10.1177/0268355515604258 Cost-utility analysis of great saphenous vein ablation with radiofrequency, foam and surgery in the emerging health-care setting of Thailand Boonying Siribumrungwong1, Pinit Noorit2, Chumpon Wilasrusmee3, Pattara Leelahavarong4, Ammarin Thakkinstian5and Yot Teerawattananon4 Abstract Objectives: To conduct economic evaluations of radiofrequency ablation, ultrasound-guided foam sclerotherapy and surgery for great saphenous vein ablation. Method: A cost-utility and cohort analysis from societal perspective was performed to estimate incremental cost- effectiveness ratio. Transitional probabilities were from meta-analysis. Direct medical, direct non-medical, indirect costs, and utility were from standard Thai costings and cohort. Probabilistic sensitivity analysis was performed to assess parameter uncertainties. Results: Seventy-seven patients (31 radiofrequency ablation, 19 ultrasound-guided foam sclerotherapy, and 27 surgeries) were enrolled from October 2011 to February 2013. Compared with surgery, radiofrequency ablation costed 12,935 and 20,872 Baht higher, whereas ultrasound-guided foam sclerotherapy costed 6159 lower and 1558 Bath higher for out- patient and inpatient, respectively. At one year, radiofrequency ablation had slightly lower quality-adjusted life-year, whereas ultrasound-guided foam sclerotherapy yielded additional 0.025 quality-adjusted life-year gained. Because of costing lower and greater quality-adjusted life-year than other compared alternatives, outpatient ultrasound-guided foam sclerotherapy was an option being dominant. Probabilistic sensitivity analysis resulted that at the Thai ceiling threshold of 160,000 Baht/quality-adjusted life-year gained, ultrasound-guided foam sclerotherapy had chances of 0.71 to be cost-effective. Conclusions: Ultrasound-guided foam sclerotherapy seems to be cost-effective for treating great saphenous vein reflux compared to surgery in Thailand at one-year results. Keywords Radiofrequency ablation, foam sclerotherapy, varicose vein, cost-utility, economic evaluation 1Department of Surgery, Faculty of Medicine, Thammasat University Hospital, Thammasat University, Pathumthani; Center of Excellence in Applied Epidemiology, Faculty of Medicine, Thammasat University Hospital, Thammasat University, Pathumthani, Thailand 2Department of Surgery, Chonburi Hospital, Chonburi, Thailand 3Department of Surgery, Faculty of Medicine, Ramathibodi Hospital, Mahidol University, Bangkok, Thailand 4Health Intervention and Technology Assessment Program (HITAP), Department of Health, Ministry of Public Health, Nonthaburi, Thailand 5Section for Clinical Epidemiology and Biostatistics, Faculty of Medicine, Ramathibodi Hospital, Mahidol University, Bangkok, Thailand Introduction Varicose veins (VVs) affect up to 20–30% of adults1,2in which great saphenous vein (GSV) is the most common site of venous reflux worldwide, including Thailand.3 Endovenous procedures are emerging which include endovenous laser ablation (EVLA), radiofrequency ablation (RFA), and ultrasound-guided foam sclero- therapy (UGFS). Compared to the standard surgery (i.e. saphenofemoral ligation and GSV stripping), these procedures are associated with less post-operative pain, fewer complications (e.g. wound infection, and hematoma) and shorter times for return to normal activities with similar efficacy for RFA and EVLA Corresponding author: Boonying Siribumrungwong, Department of Surgery, Faculty of Medicine, Thammasat University Hospital, Thammasat University, Pathumthani 12120, Thailand. Email:

  2. 574 Phlebology 31(8) but less efficacy for UGFS.4However, UGFS has a lower cost, safety and can also be easier to repeat. As a result, these endovenous procedures have become increasingly popular. There has been a threefold increase in patients undergoing sclerotherapy and almost a doubling of endovenous procedures being per- formed in the United Kingdom (UK).5 A cost-effectiveness analysis in UK has shown that day case surgery of EVLA and RFA is likely to be cost-effective.6However, this conclusion may not be generalisable to other settings, especially in developing countries like Thailand, due to the differences in health and economic infrastructures. Currently, the costs for VV surgery and UGFS are reimbursable by the Thai Universal Health Coverage scheme but not for RFA and EVLA. The latter may be explained by the lack of evidence to justify the increased costs of RFA and EVLA to the Thai health-care system. Therefore, this cost-utility analysis was conducted to assess costs and consequences of RFA, UGFS, and surgery to inform health benefit package development in Thailand. of interventions (i.e. RFA, UGFS, and standard sur- gery) (Figure 1). Health consequences included in the model were wound infection and incomplete ablation of GSV (primary treatment failure). This was defined as recanalisation diagnosed by duplex scan for endove- nous procedures and incomplete stripping, incomplete removal of an intended vein, or recanalisation for surgery.4 Other significant different outcomes are post-operative pain, hematoma, thrombophlebitis were not included in cost estimation because most of them are treated conservatively and already incorporated in the QoL measure.4Both out- patient and inpatient management were analysed. Decision to choose inpatient or outpatient management depended on necessary to do concomitant phlebectomy and patient’s preference. Outpatient interventions were done using local tumescent anesthesia with monitored anesthetic care and inpatient interventions used spinal or general anesthesia. The time horizon was one year due to limited data of long-term outcome in randomized controlled trials (RCTs), which were pooled in the meta-analysis.4 Primary failure was a surrogate outcome of clinical recurrence8and might be symptomatic or asymptom- atic.9Therefore, it might not be associated with a utility change until there was symptomatic clinical recurrence, when retreatment would be recommended and done by UGFS at one year follow up.10 and superficial Methods Model The study was conducted according to the Thai Health Technology Assessment approval by research committees of Thammasat University, Ramathibodi, and Chonburi hospitals. A decision tree model was constructed in order to esti- mate costs and health consequences in terms of quality- adjusted life years (QALYs) of performing three types Guidelines7 with ethical Input parameters Transitional probabilities of wound infection and primary failure. For clinical outcomes, we used results from Figure 1. Decision tree analytic model.

  3. Siribumrungwong et al. 575 systematic review and meta-analysis.4 biases from different techniques and settings in the pooled studies used in the meta-analysis,7the risk ratio was used. The incidence of primary failure and wound infection in the surgery group were taken from data in Thai patients. A transitional probability for each decision tree branch was then calculated by multi- plying incidence of events in the surgery group from the cohort with the pooled risk ratio.4The proportions of primary failure with symptoms at one year were assumed to be the same as those obtained by Shadid et al.9 respectively, accounted for 0.5 month (1/24¼4.2%), 0.5 month (4.2%), and 11 months (11/12¼91.6%) from a one-year time horizon. The utility of wound infection was accounted for within two weeks because wound infections have generally healed within this time. Primary failure without symptoms was not con- sidered to affect to the utility. However, primary fail- ure with symptoms was considered to have utility equal to patients with GSV reflux, which accounted for 11 months in a one-year time horizon. To prevent Statistical analysis Resources used, costs, and utility. A prospective cohort study was carried out in Thammasat University, Ramathibodi and Chonburi hospitals to collect direct medical costs, non-medical costs (i.e. traveling cost and accommodation for seeking care, and costs of informal care), indirect costs (i.e. productivity loss due to sick leave), and utility data across different types of inter- ventions. Adult patients presenting with VVs with/ without ankle edema (Clinical Etiologic Anatomic Pathophysiologic: CEAP class 2, 3) between October 2011 and February 2013 were enrolled if they met all of the following criteria: (i) had isolated unilateral GSV reflux diagnosed by duplex scan, (ii) had no history of deep vein thrombosis or superficial thrombophlebitis, (iii) had no peripheral arterial occlusive disease, and (iv) were not pregnant. Patients with reflux in tributaries were not considered. All patients signed informed consent before receiving intervention. Patients were informed about treatment choices (i.e. RFA, UGFS, and surgery), advantages, disadvantages, and cost of each procedure. Decision in choosing treatment proced- ure was done by patient under suggestion by physicians. RFA (Covedien ClosureFASTTM, California, USA) was performed with tumescent anes- thesia begun at 2–3cm distal to saphenofemoral junc- tion to GSV at knee. One session of UGFS was done with saphenofemoral ligation.11,12Foam sclerosant (Tessari’s technique; 1cc of 1% Aethoxysklerol mixed with 3cc of air) about 6–8cc was injected to the GSV just below the knee. Surgery was done with saphenofe- moral ligation with invagination stripping to GSV just below the knee. The EuroQol (EQ) 5DTM, which has been vali- dated in Thailand, was used to measure health- related QoL and utility score.13Utility was measured pre-intervention (U0), at one to two weeks (U1), one month (U2), and six months (U3) post-intervention. Post-intervention utility gained at a given time was calculated by subtracting utility at that time with pre- intervention utility (i.e. utility gain at one to two week¼U1–U0). For each individual patient, utility gain at one week, one month, and six months were, Baseline characteristics, clinical outcomes, and costs for the Thai cohort were described using mean with stand- ard deviation and frequencies. Baseline characteristics and outcomes were compared between intervention groups using Chi-square test for categorical data, and one-way ANOVA (or quantile regression analysis when non-normal distribution was found) for continuous data. A multivariate linear appropriate) regression analysis was applied to assess the relationship between intervention groups and out- comes. Covariables that were significantly associated with the outcomes were also included in the regressive model. Analyses were performed using STATA version 11.0. A p-value of less than 0.05 was considered statis- tically significant. Cost-utility analysis was conducted based on a soci- etal perspective by comparing RFA or UGFS to sur- gery. The incremental cost-effectiveness ratios (ICERs) were estimated to inform additional costs per one QALY gained. One intervention was claimed to be dominant over another intervention if it had lower cost but higher QALY, or it was dominated if it had higher cost but lower QALY. An intervention was cost- effective if its estimated ICER was lower than 160,000 Baht per QALY (a Thai cost-effectiveness threshold).14 For the reason to inform health benefit package devel- opment, the minimum daily wage of 300 Baht per day in Thailand was used. A probabilistic sensitivity analysis was conducted to examine uncertainty of estimated ICERs using a Monte Carlo simulation bootstrap. Probabilistic results were plotted as a cost-effectiveness acceptability curves. The proportion of ICERs that were lower than cost-effectiveness threshold demonstrated probability of that interven- tion to be cost-effective at that threshold, which would be demonstrated acceptability curve in which Y axis is the probability of each intervention to be cost-effective and X axis is a value of ceiling ratios.15All cost analyses were per- formed using user-own Microsoft Excel 2010.16 (or quantile where San Jose, with a 1000-replication as the cost-effectiveness written commands in

  4. 576 Phlebology 31(8) RFA had significantly higher incomes with about half of them paying for treatment by themselves contrast to the other two. Most RFA treatments were done as out- patient (74%) compared to UGFS (5%) and surgery (4%) and also had significantly less proportion of con- comitant treatment of varicosities (p<0.001), see Table 1. Results Seventy-seven patients, of which 31, 19, and 27 under- went RFA, UGFS, and surgery, respectively, were enrolled. Baseline clinical characteristics and post-inter- vention venous clinical severity scores were not signifi- cantly different between groups (Table 1). However, Table 1. Baseline characteristics and outcomes between endovenous and surgery groups. RFA UGFS Surgery N¼31 N¼19 N¼27 Variables P Baseline characteristics Age, year, mean (SD) Incomes (Baht/month) – ?15,000 – >15,000 Types of payment 55 (9.9) 49 (13.4) 52 (14.8) 0.25 <0.001 7 (23%) 13 (68%) 22 (81%) 24 (77%) 6 (32%) 5 (19%) <0.001 – – Reimbursable by health-care provider Not reimbursable by health-care provider 15 (48%) 16 (52%) 17 (89%) 2 (11%) 27 (100%) 0 GSV diameter, cm, mean (SD) CEAP C classification 0.79 (0.14) 0.69 (0.22) 0.85 (0.36) 0.39 – – C2 C3 11 (35%) 20 (65%) 9 (47%) 10 (53%) 5 (19%) 22 (81%) 0.12 VCSS, median (75% IQR) Pre-intervention utility (U0), mean (SD) Type of management 6 (4,7) 5 (3,6) 6 (3,9) 0.68 0.12 0.680 (0.190) 0.562 (0.248) 0.592 (0.227) <0.001 – – Inpatient Outpatient 8 (26%) 23 (74%) 18 (95%) 1 (5%) 26 (96%) 1 (4%) Concomitant treatment of varicosities – Yes ? Phlebectomies ? Foam sclerotherapy – No Outcomes Post-intervention individual VCSS improvementa, median (75% IQR) – one week <0.001 13 (42%) 17 (89%) 24 (89%) 10 (77%) 3 (23%) 11 (65%) 6 (35%) 5 (21%) 19 (79%) 18 (58%) 2 (11%) 3 (11%) 0.163b, 1.000c 0.246b, 1.000c 0.101b, 1.000c 0.494b, 0.429c 3 (3,4) 2 (2,4) 2 (2,4) – – one month six month 4 (3,5) 4 (4,5) 2 (1,3) 4 (2,5) 3 (2,5) 2 (1,7) 3 (2,5) 3 (1,5) 3.5 (2,6.5) Time to return to normal activities (day) With wound infection Time to return to work (day) – 5 (2.5,7) 2.5 (2,4) 5 (2.5,7) 1 (1,5) 5 (2.5,7) 5 (3,8.5) 0.199b, 0.028c With wound infection Caring time (by relatives) – 5 (3,9.5) 2 (1,5.5) 5 (3,9.5) 4 (1.5,7) 5 (3,9.5) 6.5 (3,14) 0.197b, 0.441c 1GBP approximately equal to 49 Baht. Cm: centimeters; GSV: great saphenous vein; IQR: interquartile range; RFA: radiofrequency ablation; SD: standard deviation; UGFS: ultrasound-guided foam sclerotherapy; VCSS: venous clinical severity scores. acalculated by subtracting VCSS at follow up with the baseline VCSS. bP values of RFA compared to surgery. cP values of UGFS compared to surgery.

  5. Siribumrungwong et al. 577 had the probability of 0.15 and 0.71 to be cost-effective compared to surgery, respectively. Univariate analysis found time to return to normal activities and work and post-intervention caring time by relatives were significantly longer after surgery than RFA and UGFS with two (p¼0.131) and three days (p¼0.005) less time to work and five (p¼0.002) and three days (p¼0.085) less caring time by relatives for RFA and UGFS, respectively. This resulted in less direct non-medical and indirect costs (Table 2). However, when adjusted for unbalanced covariables that were considered confounders (income/month, con- comitant treatment of varicosities), only time to return to work after UGFS was significantly shorter than sur- gery, see Table 1. For direct medical costs, data on resources used was multiplied by unit costs identified from a standard cost- ing menu.17Procedural-related costs were significantly higher in: (i) RFA (26,417 Baht) compared to UGFS (5,556 Baht) and (ii) surgery (5096 Baht), (p<0.001), see Table 2. These included cost of intraoperative ultra- sound (1417 Baht17), catheter and generator (21,000 Baht), and operative theater fee (4000 Baht17) for RFA; intraoperative ultrasound Aethoxysklerol (139 Baht), and operative theater fee (4000 Baht17) for UGFS; and standard costs for saphe- nofemoral ligation with stripping (5096 Baht17) for surgery. Utility gained for each individual at each time has been shown in Figure 2 and Table 2. At one to two week post-intervention, RFA seemed to have higher util- ity gained, but it was diluted later and became to be similar to UGFS and surgery at one and six months post-intervention. For those four patients with wound infection, post-operative utility scores at one week, one month, and six months were, respectively, 0.597 (SD¼0.118), 0.905 (SD¼0.191) and 1.000 (SD¼0) with utility gain of 0.047 (SD¼0.430) at one week post-intervention. Probabilistic results of costs, QALYs, and incremen- tal analysis of each intervention compared with surgery for inpatient and outpatient management have been illustrated in Table 3. Outpatient UGFS had the lowest costs and yielded the highest QALYs gained, so it was most cost-effective. Inpatient UGFS had about 1558 higher costs and yielded 0.025 QALY gained, resulting in an ICER of about 62,320 Baht/ QALY that was cost-effective for Thailand’s cost-effec- tiveness threshold. RFA had highest costs with slightly less QALY gained. Sensitivity analysis by treating all patients with primary failure was done for UGFS vs. surgery with the ICER of 87,040 Baht/QALY for inpa- tient and UGFS dominant to surgery for outpatient. The cost-effectiveness acceptability curve has been shown in Figure 3 with demonstrated three options of treatment. At ceiling threshold of 160,000 Baht/QALY gained, RFA and UGFS as outpatient management Discussion To our best knowledge, this is the first economic evalu- ation of endovenous procedures from a developing country. This study has shown that UGFS was likely to be cost-effective and was dominant to surgery in outpatient setting. RFA was deemed not to be cost- effective for both inpatient and outpatient settings due to its limited time of benefit and much higher direct medical costs. We used societal perspective which included both direct and indirect costs. Furthermore, the QALY gained was collected from a real clinical setting in Thailand. For treatment efficacy, we used results from most recent meta-analysis of RCTs,4which should lead to valid results for the analysis. Disease-specific quality of life measures are more sensitive to the detect changes of outcomes after intervention. To compare interven- tions between different diseases for informing policy- maker, EQ-5D was used to estimate utility for analysis. RFA had approximately a 21,000 Baht higher proced- ural cost than surgery but had advantages over surgery by reducing the care time spent by relatives by 4.5 days and reducing productivity loss by 2.5 days. Considering RFA and surgery in condition with similar efficacy, com- plications, and type of management (outpatient vs. inpa- tient); RFA would be dominant to surgery if either RFA costed less or Thailand’s minimum wage increased, so the advantages costs from RFA would be more than the difference of procedural cost. UGFS cost approxi- mately 500 Baht more than surgery but was associated with fewer days of post-intervention care by relative (2.5 days) and a gain of four days of productivity. These advantages made UGFS cost-effective (inpatient care) and dominant (outpatient) to surgery. Utility gained post-intervention was slightly, but not significantly better in the RFA group at one week follow-up and this difference was not seen over one year, consistent with results from recent meta-analysis4 and two large RCTs.18,19Furthermore, the overall util- ity gained after RFA was slightly less than surgery; thus causing RFA to be dominated by surgery. This might have occurred by chance due to the limited number of patients or could be related to the minimal contribution of utility gained at one week post-intervention com- pared to the overall utility gain. We also found signifi- cantly less concomitant treatment of varicosities in RFA. Concomitant treatment of varicosities leads to better gains in QoL in the first three months after treat- ment compared to delayed treatment with no signifi- cant difference in number of further procedures for varicosities.20This could influence to the utility gain Baht17), (1417

  6. 578 Phlebology 31(8) Cohort, Standard cost listsa Standard cost listsa Standard cost listsa Standard cost listsa Standard cost listsa Standard cost listsa Standard cost lists Shadid et al.11 Meta-analysis7 Meta-analysis7 and companya Sources Cohort Cohort Cohort Cohort Cohort Cohort Cohort Cohort Cohort Cohort Cohort 0.071b0.680c 0.392b0.491c 0.893b0.747c 0.28 0.008 0.17 0.02 IQR: interquartile range; NA: not applied; RFA: radiofrequency ablation; RR: relative risk; SD: standard deviation; UGFS: ultrasound-guided foam sclerotherapy. <0.001 NA NA NA NA NA NA NA NA NA NA NA NA NA P 1500 (900,2,850) 1950 (975,3330) 1260 (940,2145) 1500 (900,2550) 600 (415,985) 0.084 (0.277) 0.304 (0.249) 0.210 (0.246) 1619 4669 5096 5839 1917 – – – – 0.07 0.15 0.54 1215?3.0 Surgery 2040 (1000,2350) 1500 (900,2,850) 1200 (600,1200) 800 (510,1200) 300 (300,1500) 700 (520,860) 2.4 (1.6,3.6) 0.3 (0.1,0.7) 0.051 (0.271) 0.195 (0.218) 0.291 (0.261) aHealth Intervention and Technology Assessment Program (HITAP) ( 0.42 1,619 1215?2.6 4669 5556 675 5839 1917 – – UGFS 1625 (1550,1700) 1500 (900,2,850) 600 (300,1650) 741 (500,1400) 750 (600,1200) 700 (520,860) 1.3 (0.7,2.4) 0.3 (0.1,0.7) 0.176 (0.234) 0.225 (0.231) 0.203 (0.244) 1619 4669 675 5839 1917 – – 26,417 0.42 1215?2.4 RFA Table 2. Input parameters for cost-utility analysis. in-patient management) (1215 Baht/day) – Productivity loss of wound infection Indirect cost, Baht, median (75% IQR) – Costs of hospitalization (only for Post-operative utility gain, mean (SD) cP values of UGFS compared to surgery. of operative room, equipment, 1GBP approximately equal to 49 Baht. bP values of RFA compared to surgery. – Wound infection, percentage and drugs for sclerotherapy – Primary failure, percentage – Procedure (included costs – Cost of post-intervention – RR of wound infection symptoms, percentage – Outpatient procedure Direct medical costs, Baht – RR of primary failure and local anesthesia) Direct non-medical cost, – Inpatient procedure – Primary failure with – Costs of anesthesia – Cost per one visit Baht median(75% IQR) cares by relatives – Wound infection – Productivity loss – At one month – At six months – At one week – Preoperative – Retreatment Efficacy and RR ? SB, GA Parameters ? LA

  7. Siribumrungwong et al. 579 Figure 2. Utility gain between procedures. Table 3. Probabilistic results of costs, QALYs, incremental cost effectiveness ratios (ICERs) of five options for treating great saphenous reflux in societal perspective. Probabilistic resultsa Interventions Mean costs (SD) QALYs ICERs Direct medical Direct non-medical Indirect Total Surgery (inpatient)b UGFS (outpatient) 15,129 (141) 11,049 (11,511) 6977 (771) 4449 (519) 1927 (282) 979 (190) 25,688 (1202) 19,529 (9677) 0.836 (0.069) 0.862 (0.090) – Dominantc UGFS (inpatient) RFA (outpatient) 15,524 (201) 31,082 (29,988) 7811 (797) 4438 (542) 979 (190) 1086 (184) 27,246 (2324) 38,623 (26,680) 0.862 (0.090) 0.834 (0.057) 62,320 Dominated by surgery and UGFSd RFA (inpatient) 35,588 (386) 8312 (967) 1086 (184) 46,560 (1463) 0.834 (0.057) Dominated by surgery and UGFSd ICERs: incremental cost-effectiveness ratios; QALY: quality-adjusted life year; RFA: radiofrequency ablation; UGFS: ultrasound-guided foam sclerotherapy. aData of patients with no wound infection and primary failure. bSurgery (inpatient) is a comparator for ICER estimation. cThe option generates less cost and greater effectiveness than other compared intervention. dThe option generates more cost but less effectiveness than other compared intervention. which uses a threshold of £20,000 (& 1,000,000 Thai Baht) per QALY. In Thailand, RFA costs four times more than surgery whereas in the UK NHS, day case RFA costs less than surgery £776 vs. £980 for surgery, a far better cost ratio than in Thailand. Thus, in Thailand, RFA is not suitable for being reimbursable by the Thai universal health coverage scheme and is only available to those who are willing to pay for it. Our study was limited by its small size. The limited time horizon of one year could have biased the results in favor of UGFS because symptomatic failure may occur after one year. Therefore, a sensitivity analysis was done assuming all patients had symptomatic only in the early period, but not to the cost and should not affect much to the results. Adjusted analysis with concomitant treatment of varicosities also found no significant association of this variable to utility in our study (p¼0.693). A systematic reviews, network meta- analyses and exploratory cost-effectiveness model of randomized trials of minimally invasive techniques vs. surgery for VVs have also demonstrated that cost-effec- tiveness is largely dependent on the cost of the treat- ment rather than its efficacy and safety.21 Our results of RFA contrast to those of Gohel et al.6 who found that day case surgery of RFA was cost- effective in the UK National Health Service (NHS)

  8. 580 Phlebology 31(8) Figure 3. Cost-effectiveness acceptability curves demonstrate the probability that treatment is cost-effective provided at each level of cost-effective threshold (THB/QALY gained) based on societal perspective. RFA: radiofrequency ablation; UGFS: ultrasound-guided foam sclerotherapy; OPD: outpatient primary failure with similar results. These strengthen our conclusion that UGFS is more cost-effective in early term. In addition, our study might be prone to selection biases with more high income patients in the RFA group. However, we adjusted for these in the ana- lysis of the outcomes. These preliminary data are useful for our setting, and we intend to conduct larger pro- spective cost-effectiveness studies with longer follow-up. Contributorship BY, PN, CW, and PL reviewed literature, conducted study, analysis and writing manuscript, AT involved in protocol development, analysis, and manuscript writing. YT was involved in protocol development, analysis and critical comments Declaration of Conflicting Interests The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article. Conclusion Endovenous procedures are preferred over surgery and UGFS may be the most suitable alternative to surgery in Thai context now for short-term results. RFA also has advantages in terms of cost of care and productivity loss, but its high cost precludes its use in the public health system at this time. Ethical approval The Human Research Ethics Committee of Thammasat University (no. 1: Faculty of Medicine) (reference number: MTU-EC-SU-4-095/54). Funding Acknowledgement The author(s) disclosed receipt of the following financial sup- port for the research, authorship, and/or publication of this article: Health Intervention Technology Assessment Program of Thailand (HITAP) through Consortium of Thailand (grant number: HITAF 2554-050) funded this study. We would like to thank Dr. Bob Taylor and Stephen Pinder for helping us editing the manuscript and Ms. Waranya Rattanavipapong for kindly helping us to analyse the data and comments.

  9. Siribumrungwong et al. 581 of varicose veins: early results of a randomised controlled trial. Eur J Vasc Endovasc Surg 2006; 31: 93–100. 12. Liu X, Jia X, Guo W, et al. Ultrasound-guided foam sclerotherapy of the great saphenous vein with sapheno- femoral ligation compared to standard stripping: a pro- spective clinical study. Int Angiol 2011; 30: 321–326. 13. Sakthong P. Measurement of clinical-effect: utility. J Med Assoc Thai 2008; 91 (Suppl 2): S43–S52. 14. Mohara A, Youngkong S and Perez Velasco R. Using health technology asesessment for informing coverage decisions in Thailand. J Compar effect Res 2012; 1: 132–146. 15. Fenwick E, O’Brien BJ and Briggs A. Cost-effectiveness acceptability curves – facts, fallacies and frequently asked questions. Health Econ 2004; 13: 405–415. 16. Teerawattananon Y, Tangcharoensathien V. Economic evaluation of palliative management versus peritoneal dialysis and hemodialysis for end-stage renal disease: evidence for coverage deci- sions in Thailand. Value Health 2007; 10: 61–72. 17. Riewpaiboon A. Standard cost lists for health economic evaluation in Thailand. J Med Assoc Thai 2014; 97 (Suppl 5): S127–S134. 18. Rasmussen LH, Lawaetz M, Bjoern L, et al. Randomized clinical trial comparing endovenous laser ablation, radio- frequency ablation, foam sclerotherapy and surgical stripping for great saphenous varicose veins. Br J Surg 2011; 98: 1079–1087. 19. Brittenden J, Cotton SC, Elders A, et al. A randomized trial comparing treatments for varicose veins. N Engl J Med 2014; 371: 1218–1227. 20. Lane T, Onida S, Gohel M, et al. A systematic review and meta-analysis on the role of varicosity treatment in the context of truncal vein ablation. Phlebology 2015; 30: 516–524. 21. Carroll C, Hummel S, Leaviss J, et al. Systematic review, network meta-analysis and exploratory cost-effectiveness model of randomized trials of minimally invasive tech- niques versus surgery for varicose veins. Br J Surg 2014; 101: 1040–1052. References 1. Callam MJ. Epidemiology of varicose veins. Br J Surg 1994; 81: 167–173. 2. Clark A, Harvey I and Fowkes FG. Epidemiology and risk factors for varicose veins among older people: cross- sectional population study in the UK. Phlebology 2010; 25: 236–240. 3. Labropoulos N, Kokkosis AA, Spentzouris G, et al. The distribution and significance of varicosities in the saphe- nous trunks. J Vasc Surg 2010; 51: 96–103. 4. Siribumrungwong B, Noorit P, Wilasrusmee C, et al. A systematic review and meta-analysis of randomised controlled trials comparing endovenous ablation and sur- gical intervention in patients with varicose vein. Eur J Vasc Endovasc Surg 2012; 44: 214–223. 5. Kanwar A, Hansrani M, Lees T, et al. Trends in varicose vein therapy in England: radical changes in the last decade. Ann R Coll Surg Engl 2010; 92: 341–346. 6. Gohel MS, Epstein DM and Davies AH. Cost-effective- ness of traditional and endovenous treatments for vari- cose veins. Br J Surg 2010; 97: 1815–1823. 7. Permsuwan U, Guntawongwan K and Buddhawongsa P. Handling time in economic evaluation studies. J Med Assoc Thai 2008; 91 (Suppl 2): S53–S58. 8. Gloviczki P, Comerota AJ, Dalsing MC, et al. The care of patients with varicose veins and associated chronic venous diseases: clinical practice guidelines of the Society for Vascular Surgery and the American Venous Forum. J Vasc Surg 2011; 53: 2S–48S. 9. Shadid N, Ceulen R, Nelemans P, et al. Randomized clinical trial of ultrasound-guided foam sclerotherapy versus surgery for the incompetent great saphenous vein. Br J Surg 2012; 99: 1062–1070. 10. Bradbury AW, Bate G, Pang K, et al. Ultrasound-guided foam sclerotherapy is a safe and clinically effective treat- ment for superficial venous reflux. J Vasc Surg 2010; 52: 939–945. 11. Bountouroglou DG, Azzam M, Kakkos SK, et al. Ultrasound-guided foam sclerotherapy combined with sapheno-femoral ligation compared to surgical treatment Mugford M and