Using Horsechestnut Seed Extract in the Treatment of Venous Leg Ulcers: A Cost-Benefit Analysis
Venous leg ulceration (VLU) affects approximately 0.6% of the western population1 and accounts for millions of dollars in healthcare expenditures every year. In view of rising budgetary restraints, staff shortages, scarcer resources, spiralling healthcare expenses, and the costs of managing leg ulceration, the net expense of adopting alternative or complementary wound treatments needs to be considered.2-4 Horsechestnut seed extract (HCSE) — Aesculus hippocastanum— is an herbal extract purported to advance venous ulcer healing and, consequently, reduce net healthcare expenditure. To test this assumption, the cost-effectiveness of HCSE and conventional treatment was compared to conventional treatment alone. The findings from this study will not only add to the dearth of economic evaluations in nursing and wound care,4-6 but also will provide a better understanding of the societal benefit of using HCSE as an adjunctive wound care treatment.
Literature Review
Leg ulcer management is expensive.7,8 In France, for instance, the treatment of VLU accounts for an estimated 1.1%, and in the United Kingdom (UK) for approximately 1.3% of total healthcare expenditure.9 Given these figures and the AUD $50.3 billion spent on healthcare in Australia during 1998-99,10 the annual cost of managing VLU in Australia is likely to range between AUD $554 and $655 million. This compares to the AUD $628 million to $1.3 billion spent in the UK every year11 and the AUD $1 to $1.3 billion spent in the US.12
A large part of the cost associated with treating VLU is incurred during inpatient and outpatient management of the condition (which does not include community nursing care).13 According to a cost-analysis of dressing materials used in venous ulcer management, 4 months of conventional outpatient leg ulcer treatment costs an estimated AUD $3,245 per individual.7 By comparison, a 24-day stay in a private Australian hospital for leg ulcer treatment averages AUD $8,734 per admission.14 In addition, many persons with venous ulcers require community nursing care,14 although the cost of this service is still unclear.
Additional financial implications of leg ulceration include personal costs, such as transportation and time lost from work. Patient dependence on social security payments and the possible need for ongoing care in a nursing or residential facility create further economic burdens.15 Phillips, Stanton, Provan, and Lew16 highlight the significance of these personal costs in a study of 62 US patients with leg ulceration: more than 30% reported a mean annual out-of-pocket expense of between USD $100 and USD$1,000 for the treatment of this condition. Therefore, the management of VLU is likely to strain individual and community resources — interventions that reduce the duration and recurrence of ulceration, as well as the frequency of dressing changes, are likely to reduce the burden on these resources. Findings from a recent randomized, controlled trial17 involving 54 subjects found that HCSE can significantly (t = -2.71, P = 0.009) affect the frequency of dressing changes after 12 weeks of treatment, subsequently lowering costs for care.
Horsechestnut Seed Extract
Horsechestnut seed extract, as well as diosmin and 0-(b-hydroxyethyl)-rutosides (HR), are venotonic agents — substances that have been shown to induce a range of vascular effects in both animal and human models, including reducing interstitial exudation and edema formation.18 Although often compared, the actions of these medications are not akin to those exerted by compression therapy. First, randomized clinical trials have shown that the effects brought about by these agents persist after the cessation of treatment.19 Second, both pharmacological and clinical studies have found that venotonics specifically target the pathological processes associated with VLU, including poor venous tone, decreased microcirculation, increased capillary permeability, and edema.20 These agents also reduce inflammation, neutrophil migration, venous damage, and growth factor release in hypoxic venous endothelium.19 These actions are further supported by a Cochrane review of 16 randomized, controlled trials on HCSE that found that HCSE is significantly more effective than placebo in reducing lower leg edema, calf and ankle girth, leg pain, pruritus, and fatigue.21 However, even though venotonic preparations exert direct venous effects, evidence to indicate superior or inferior activity to compression is insufficient. Hence, while venotonic treatment may be utilized as an alternative to compression therapy,19 combining the two treatments is believed to yield a more beneficial effect.18,19
The following study is a cost benefit analysis of the horsechestnut and venous leg ulcer trial — a 12-week prospective, randomized, placebo-controlled trial conducted in South Australia in 2002-200417 — that examines whether the treatment of venous ulceration with both HCSE and conventional therapy is more cost effective than conventional therapy alone.
The Role of Cost-Benefit Analysis
Cost-benefit analysis (CBA) can be used to examine the costs and benefits of alternative interventions from a societal perspective.22 Comparing the estimated total cost of administering an intervention to the estimated benefits the treatment creates23 can inform stakeholders of the inherent worth of an intervention. Calculating net benefits (benefits minus costs) for several different interventions may facilitate the selection of one from among many competing treatment alternatives.24,25 Thus, a CBA does not prescribe choices — it describes the choices available.3-4,26 Therefore, just as evidence can inform clinical practice, decisions regarding the allocation of resources can be guided by cost-benefit analysis.2,27
Despite the proclaimed benefits of CBA, economic evaluations of nursing practice4 and wound management5,6 remain scant, hindering or delaying incorporation of effective interventions into clinical practice. The current study was developed and conducted to contribute to the body of wound care knowledge and to ascertain the societal benefit of using HCSE in the treatment of VLU.
Methods
Design. The current study initially was designed to compare the net benefits of treating venous ulceration with HCSE and conventional therapy to results obtained when using placebo and conventional therapy. However, differential benefits of these interventions were difficult to quantify.4 Indeed, few researchers have attempted such a quantification of benefits in the area of wound management. More importantly, the clinical benefits associated with the two treatments (eg, improvements in ulcer healing and wound topography and the frequency of dressing changes, as discussed elsewhere17), are similar if not better under HCSE treatment.17 Thus, under the conservative assumption of no meaningful difference in clinical benefits between treatments, the current study focused on the costs of treating venous ulceration with HCSE and conventional therapy compared to those using placebo and conventional treatment without attempting to enumerate cost intangible benefits to the patient or service provider.
Sample and setting. Home health care accounts for 69% of the cost of managing venous ulceration6; thus, viewing the current study from the perspective of a large metropolitan community nursing service is appropriate. The large district nursing service within metropolitan Adelaide, South Australia chosen for the study employs 323 nurses over three regional centers and 12 nursing clinics.28 The service treated more than 1,420 clients with acute and chronic wounds in a representative month (June 2002), of which 246 had venous ulcers.
The study sample comprised 54 randomly selected participants from the horsechestnut and venous leg ulcer trial.17 All participants had received wound management from the nursing service for the duration of the trial and satisfied the selection criteria (see Table 1).
In the case where bilateral leg wounds were evident, the ulcer on the leg with an ankle-brachial pressure index (ABPI) nearest to 1.0 was chosen. If multiple wounds were present, the largest wound was selected.
Variables. A CBA utilizes the estimated resources involved in delivering a service or intervention23 (eg, labor, supplies, equipment, transportation, capital, and social service costs22,23). In the context of wound care, costs are incurred primarily through the purchase of necessary dressing materials and the financing of practitioner time and travel.29,30 Therefore, the total outlay of managing VLU in the community includes the cost of staff salaries, organizational infrastructure, Workcover (workers’ compensation), superannuation (employee retirement income), travel, dressing materials, and specific interventions (in this case, HCSE).
Costs considered in this analysis include:
HCSE — The HCSE tablets produced for the study were manufactured by Mediherb Pty Ltd, (Warwick, Queensland). The cost to consumers for the purchase of HCSE was estimated at AUD $20 per 60-tablet bottle — at the administered dosage of one tablet twice daily, one bottle of tablets provided 30 days of treatment.
Dressing materials — The cost of all dressing materials applied to the ulcers of study participants, including all primary and secondary dressings, was tallied for each individual and calculated as dressing costs per visit. The primary dressings applied included either Tricotex™ (Smith & Nephew, UK), Mesorb (Molnlycke, Sweden), Melolin™ (Smith & Nephew), or Viscopaste™ (Smith & Nephew). The secondary dressings applied included either Profore™ (Smith & Nephew), SurePress® (ConvaTec USA), Lestreflex (Seton Healthcare, UK) or graduated Tubigrip (Seton Healthcare). Due to the discounted prices offered to the community nursing service, estimates of these costs were relatively conservative or cautiously moderate.
Service costs — Nurses employed by the nursing service were paid in accordance with state award rates. Because four different levels of nurses visited clients in the community, the hourly rate of pay ranged between AUD $15.36 and AUD $25.56 per hour.31 The rate of pay the nursing service used to calculate visit costs was not disclosed but was incorporated into an overall fee of AUD $69 per client visit, irrespective of visit duration. This fee was also inclusive of organizational infrastructure, Workcover, superannuation, and travel cost.
Data analysis. All costs associated with the management of VLU were calculated in 2004 Australian dollars. To improve the precision of these findings, all inputs were microcosted — ie, the cost of each resource was calculated for each participant.24 Means and standard deviations were used to describe variables with normal distributions, while medians and interquartile ranges were used to describe variables whose distributions were significantly skewed. Differences between group means were evaluated using the t-test for independent groups. In particular, the difference in total costs between the HCSE and placebo groups were calculated and tested for statistical significance.
Sensitivity analysis. Sensitivity analyses also were conducted to determine whether variations in input costs would alter the conclusions made from the analysis.23,32 More specifically, one-way sensitivity analyses were performed on staffing costs and dressing costs, with only one expense manipulated at a time and all other costs remaining constant.32 To ascertain whether total expenditure was sensitive to variations in the base-case estimates of staffing and dressing costs, each expense was generously varied by 50% above and below the mean.
Discounting. Discounting is a strategy used in some applications of cost-benefit analysis to convert future costs or benefits into present values.23 However, the current analysis did not predict the expense of either treatment beyond 12 weeks. Hence, the costs of each treatment were not discounted.
Results
Frequency of dressing changes. The frequency of dressing changes in the HCSE group decreased from 2.1 per week at baseline to 1.1 at week 12 (see Table 2). The number of dressing changes in the placebo group increased from 2.41 per week at baseline to 2.48 at week 12. The difference between study groups in the number of dressing changes required at week 12 was statistically significant (t-test, P = 0.009).
Staffing costs. The weekly staffing costs associated with the treatment of VLU consistently decreased in the HCSE group from $144 at baseline to $77 at week 12 (see Table 3). However, staffing costs in the placebo group fluctuated, decreasing in the first 8 weeks from $167 at baseline to $131 at week 8, then escalating to $172 at week 12. The difference in staffing costs between groups at baseline was not significant (t-test, P = 0.159). A cost saving of $95 per patient receiving HCSE was statistically significant at week 12 (t-test, P = 0.004).
Dressing costs. The weekly cost of dressing materials in the HCSE group consistently declined over the 12 weeks, from AUD $25.85 at baseline to AUD $15.34 at week 12 (see Table 4). However, the placebo group demonstrated a much smaller reduction in dressing costs over time, decreasing from $28.07 at baseline to $25.69 at 12 weeks. Despite an average weekly cost saving of $10 for each patient in the HCSE group at week 12, the difference in the cost of dressing materials between groups was not statistically significant at this time point (t-test, P = 0.342) or at any other week of assessment.
Total wound management costs. The total weekly cost of managing VLU in the HCSE group (including staffing, dressing, and intervention costs) decreased from $175 at baseline to $97.20 at week 12 (see Table 5). Total costs in the placebo group also declined in the first 8 weeks, from $195.39 at baseline to $156.43 at week 8, but increased thereafter to $198.15 at week 12. While total wound care costs at baseline were slightly higher in the placebo group, the difference between groups was not statistically significant (t-test, P = 0.238). The difference between groups at week 12 was statistically significant (t-test, P = 0.011), with an average saving of approximately $101 per patient in the HCSE group. These wound management costs and savings also were insensitive to variations in staffing costs and dressing costs within the range of ± 50%, remaining statistically significant at week 12 in all sensitivity checks.
Discussion
Venous leg ulcer management consumes a large number of resources, including dressing materials, staff, and infrastructure.5 Even though these resources are estimated to cost between AUD $554 million and $655 million annually, deliberation has been scant in Australia regarding the expense of treating VLU on an individual basis. However, recent overseas studies have shed some light on costs. A randomized controlled trial33 in the UK, for instance, found that treating patients with VLU for 24 weeks using two-layer and four-layer compression bandaging cost an estimated AUD $2,319 (USD $1,375) and AUD $2,218 (USD $1,315) per patient, respectively.33 In contrast, a cost-effectiveness modeling study34 found that treating venous ulcers at home for 12 weeks with a hydrocolloid dressing and high compression bandaging cost AUD $1,588 (USD $810) in France and AUD $2,240 (USD $1,143) in Europe for each wound healed. Managing VLU for 12 weeks in the current study cost AUD $1,740.52 per individual treated in the HCSE group and AUD $2,159.88 per individual in the placebo group. Regardless of the different types of primary and secondary dressings applied, the total costs in the current study were consistent with those previously reported by Meaume and Gemmen.34 The discrepancy between these findings and those reported in the UK study,33 however, may be explained by differences in the costs of dressing materials and the frequency of nursing visits.
From a healthcare provider perspective, HCSE was a more cost-effective supplement to the conventional treatment of venous ulceration than placebo. In particular, the total cost (in Australian dollars) of treating VLU with placebo and conventional care for 12 weeks was $2,159.88 per patient, compared to $1,740.52 per patient receiving HCSE and conventional care — a 24% cost saving. Moreover, these savings were insensitive to wide variations in the base-case estimates of staffing and dressing costs and, therefore, were robust to plausible explanations.32 Furthermore, because no significant cost differences were noted between groups at baseline, it is unlikely that these savings were the result of heterogenous costs at week zero. Most importantly, as reported elsewhere,17 patient outcomes in the HCSE group were no worse and in some cases were significantly better under HCSE treatment when compared to placebo, particularly in relation to wound slough and visit frequency. Therefore, using HCSE as a treatment of venous ulceration may lessen healthcare expenses while providing equal or greater benefit to patients.
According to data from a systematic review of 13 randomized controlled trials,35 HCSE can produce mild and transient gastrointestinal symptoms in 0.9% to 3% of patients. Although HCSE is available in the US from a number of manufacturers and standardized to contain at least 20 mg aescin per dose, the quality of the product in the US has not been ascertained. As such, and in light of current budgetary constraints, some organizations may be cautious about purchasing new and unconventional interventions. Despite these reservations, integrating HCSE into the management of VLU is likely to save the healthcare provider up to AUD $101 (USD $78) per patient per week, with an average saving of AUD $95 (USD $74) in organizational costs, and AUD $10 (USD $7.80) in dressing expenditure. Given the greater rate of wound healing brought about by HCSE,17 supplementing conventional treatment with HCSE also may avoid further costs to the organization by reducing demand on scarce healthcare resources. These savings, conservatively estimated, suggest that the treatment of venous ulceration with HCSE is likely to improve the efficiency of district nursing services.
The most expensive component of wound management is the cost of nursing care, a claim supported not only by the current study, but also by Meaume and Gemmen.34 Therefore, any intervention that reduces the need for nursing visits is likely to produce substantial cost savings for the healthcare provider and allow organizations to better accommodate staff shortages. Since all CBA expenses were multiplied by the frequency of nursing visits, any variations in the rate of dressing changes subsequently affected the total cost of wound care. The significant decline in total wound management costs, particularly staffing costs, realized with HCSE treatment are directly attributable to a reduction in the number of nursing visits, which may have been the result of wound healing outcome differences because no significant difference in visit frequency was evident between groups when healed ulcers were excluded from the analysis (t-test, P = 0.108).17 Alternatively, the need for fewer dressing changes in the HCSE group may have been influenced by greater reductions in wound slough.17 Therefore, improvements in patient outcomes associated with HCSE treatment are most likely responsible for the observed cost reductions.
The observed increase in dressing change frequency in the placebo group cannot be explained with certainty. Between-group differences at baseline were likely to be contributing factors because both groups were homogenous with regard to wound topography, wound history, and dressing use. The greater costs at baseline for the placebo group were directly attributable to the small increase in dressing change frequency; however, this small between-group difference was not statistically significant and most likely due to chance.
In addition, the difference between groups in the level of wound exudate (a factor guiding the frequency of dressing changes) was not statistically significant at week 4, 8, or 12. The authors can only assume that the withdrawal of participants with healed wounds and associated lower levels of exudate may have increased the overall values in each group. Nevertheless, this does not alter the study outcomes.
Limitations
This analysis represents the first Australian study to compare the costs of two alternate treatments of venous ulceration. Aside from highlighting the costs of managing VLU in the community, the current study also identified an intervention that may potentially decrease wound management expenditure. The precision and robustness of these findings were enhanced by utilizing a CBA design and by microcosting a range of inputs.
Sensitivity analyses also confirmed that the validity of the CBA was not threatened by the cost assumptions of the study.32 Even so, the current study was not without limitations.
First, the small sample in the current study (n = 54) limited the extent to which the CBA findings could be generalized. The 12-week period prohibits generalizations regarding the long-term benefit of using HCSE in the treatment of VLU. Although the study illustrated the short-term costs of HCSE treatment, future research is needed to examine the long-term benefit of the extract by 1) utilizing extended intervention periods, 2) calculating cost savings over the fiscal year, and 3) costing benefits.
Due to the difficulty in quantifying indirect costs such as pain, impaired mobility, and lost productivity,4 these expenses were not incorporated into the current CBA. This means the cost savings estimated in this study are almost certainly underestimated from a social perspective. The development of valid and appropriate instrumentation capable of measuring the indirect costs of wound care interventions in monetary terms is needed to facilitate a more comprehensive examination of the impact of such treatments.
Despite the small sample, short-term duration, and the absence of indirect expenses, the study provides the first step toward examining the economic impact of HCSE treatment. Future studies that comprehensively explore the long-term cost-effectiveness of treating venous ulceration with HCSE, for both the client and the healthcare system, to determine whether use of the extract is beneficial to society as a whole are warranted.36
Conclusion
Cost-benefit analysis “…provides an estimate of the value of resources used up by each programme compared to the value of resources the programme might save or create”.24 This study, comparing the cost of treating VLU with either HCSE or placebo, found using HCSE to be more cost-effective in the management of venous ulceration than placebo. In accordance with evidence-based practice, these findings provide economic justification for the decision to use HCSE in clinical practice. Moreover, the CBA suggests that the health and well-being of consumers and the efficiency of service provider operations may be enhanced through the use of HCSE by decreasing the frequency of nursing visits and allowing more clients to be treated at an equivalent cost.5
