- Email: [email protected]
Cost-Effectiveness of a Nonpharmacological Intervention in Pediatric Burn Care
VALUE IN HEALTH 18 (2015) 631–637
Available online at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/jval
Cost-Effectiveness of a Nonpharmacological Intervention in Pediatric Burn Care Nadia J. Brown, BAppSC(OT), MAppSc(OT), PhD1,*, Michael David, MEd, MEpi, MSc, PhD2,3, Leila Cuttle, BSc(Hons), PhD1,4, Roy M. Kimble, MD (Res), MBChB, FRCS, FRACS1, Sylvia Rodger, BOccThy, MEdSt, PhD5, Hideki Higashi, BA, MPH, MSc(HlthEcon), PhD6 1
Centre for Children’s Burns and Trauma Research, Queensland Children’s Medical Research Institute, The University of Queensland, Royal Children’s Hospital, Brisbane, Australia; 2Queensland Children’s Medical Research Institute, The University of Queensland, Royal Children’s Hospital, Brisbane, Australia; 3School of Population Health, The University of Queensland, Brisbane, Australia; 4Tissue Repair and Regeneration Program, Institute of Health and Biomedical Innovation, Queensland University of Technology, Brisbane, Australia; 5 Division of Occupational Therapy, School of Health and Rehabilitation Sciences, The University of Queensland, Brisbane, Australia; 6 Institute for Health Metrics and Evaluation, University of Washington, Seattle, WA
AB STR A CT
Objective: To report the cost-effectiveness of a tailored handheld computerized procedural preparation and distraction intervention (Ditto) used during pediatric burn wound care in comparison to standard practice. Methods: An economic evaluation was performed alongside a randomized controlled trial of 75 children aged 4 to 13 years who presented with a burn to the Royal Children’s Hospital, Brisbane, Australia. Participants were randomized to either the Ditto intervention (n ¼ 35) or standard practice (n ¼ 40) to measure the effect of the intervention on days taken for burns to re-epithelialize. Direct medical, direct nonmedical, and indirect cost data during burn re-epithelialization were extracted from the randomized controlled trial data and combined with scar management cost data obtained retrospectively from medical charts. Nonparametric bootstrapping was used to estimate statistical uncertainty in cost and effect differences and cost-effectiveness ratios. Results: On average, the Ditto intervention reduced the time to re-epithelialize by 3 days at AU$194 less cost for each patient compared with standard practice. The
incremental cost-effectiveness plane showed that 78% of the simulated results were within the more effective and less costly quadrant and 22% were in the more effective and more costly quadrant, suggesting a 78% probability that the Ditto intervention dominates standard practice (i.e., cost-saving). At a willingness-to-pay threshold of AU$120, there is a 95% probability that the Ditto intervention is cost-effective (or cost-saving) against standard care. Conclusions: This economic evaluation showed the Ditto intervention to be highly cost-effective against standard practice at a minimal cost for the significant benefits gained, supporting the implementation of the Ditto intervention during burn wound care. Keywords: burns, child, cost-effectiveness, economic evaluation, nonpharmacological intervention, randomized controlled trial, reepithelialization.
Introduction
practice being more effective than another. Changes in practice need to be quantified in monetary terms with an economic evaluation to determine the feasibility of implementing new cutting-edge treatments and interventions [3] and to justify setup costs to hospital administrators. Psychosocial costs to the patient include the pain, anxiety, and stress from the trauma of the burn injury itself and the associated wound care procedures and may encompass additional psychosocial issues resulting from the physical disfigurement of hypertrophic scarring. Severity of injury does not predict psychological costs to the patient [4]. No quality-of-life measure has been specifically developed and
It is estimated that more than 500,000 children worldwide are hospitalized with a burn injury every year [1]. Burn care is undeniably expensive and arguably has one of the highest financial costs in pediatric health care, yet limited health economic research has been conducted in the burns arena. External financial pressures on health care budgets create much deliberation and contentious debates over balancing quality health care versus less costly service delivery [2]. There is a great need for health research to extend beyond providing evidence of one
& 2015 Published by Elsevier Inc. on behalf of International Society for Pharmacoeconomics and Outcomes Research (ISPOR).
Conflict of interest: One of the supervisors of the trial, Roy M. Kimble, holds options with Diversionary Therapy Technologies (DTT); however, he did not stand to lose or gain financially or personally from the results during the clinical trial period and the time of submission. The principal researcher has no financial interest in the Ditto device or DTT. * Address correspondence to: Nadia J. Brown, Centre for Children’s Burns and Trauma Research, Queensland Children’s Medical Research Institute, Level 4, Foundation Building, Royal Children’s Hospital, Brisbane, QLD 4029, Australia. E-mail: [email protected]. 1098-3015$36.00 – see front matter & 2015 Published by Elsevier Inc. on behalf of International Society for Pharmacoeconomics and Outcomes Research (ISPOR). http://dx.doi.org/10.1016/j.jval.2015.04.011
632
VALUE IN HEALTH 18 (2015) 631–637
validated for the pediatric burn population [3,5]. Financial accountability that extends beyond acute wound care and scar management and includes the complete long-term psychosocial costs to the patient makes the “true” costs of burns unquantifiable. The most comprehensive cost-effectiveness analysis identified in the burn literature was a prospective study over a 5-year period conducted in the Burn Centre of Valencia, Spain [6]. This study incorporated incremental cost-utility ratio analysis to determine the cost and quality of life of burn patients. This cost analysis largely depended on Diagnostic Related Groups, used silver sulfadiazine in the treatment of burn wounds, and was carried out for primarily an adult inpatient population with a mean age of 40 years and mean total body surface area (TBSA) burns of 18.2%. General Diagnostic Related Group and Healthcare Resource Group cost coding provide a generic snapshot of costs and are not sensitive measures for picking up differences across individuals [7]. Two studies were identified in the burn literature that performed a cost-effectiveness evaluation on minor burns encompassing smaller TBSA (mean r10%) and primarily partialthickness depth [8,9]. Both studies compared burn dressings, which are not standard practice in burn centers in developed countries. Advances in the technology of burn dressings in recent years have changed standard practice from twice-daily applications of silver sulfadiazine and similar ointments or creams to improved silver-impregnated dressings such as Acticoat [10]. This change in practice has seen a shift from inpatient to outpatient management of the small partial-thickness burn injury, in which dressings are changed every 3 to 7 days. No economic evaluation has been performed in pediatric burn outpatients that uses advanced silver-impregnated dressings only, or that uses validated measures of re-epithelialization and wound depth. This study aimed to establish the cost-effectiveness of a computerized procedural preparation and distraction intervention Ditto, which was compared with standard practice, in a randomized controlled trial (RCT) measuring burn reepithelialization [11]. Use of virtual reality and computerized nonpharmacological interventions plays a significant role in dampening pain perception when used as an adjunct to pharmacological intervention during burn care and rehabilitation [12–21]. Several studies report on the effect of stress in delaying wound healing [22–29], and cellular links between pain and delayed wound healing are beginning to be explored [30]. The risk of hypertrophic scar formation is known to increase when wound re-epithelialization extends beyond 14 days [31,32]. Accelerating re-epithelialization is the prime objective of burn care, and has significant cost benefits to the health system and the patient.
Methods Setting and Participants Data were collected from the Stuart Pegg Paediatric Burns Centre at the Royal Children’s Hospital, Brisbane, which is a tertiary pediatric burn referral center servicing approximately 800 new patients per year. Data collection occurred in two parts: 1) RCT [11,33] data collected from August 2011 to August 2012 and 2) retrospective review of medical charts from August 2011 to September 2013 to obtain scar management data 12 months after re-epithelialization (Fig. 1). Children were included if they were aged 4 years to 13 years and had an acute burn injury of TBSA less than 15%. Children were ineligible if they were non-English speaking, had a diagnosed condition or illness in addition to a burns injury, or had a history of suspected child abuse. On the first dressing removal, further study eligibility was determined
(because of the inability to predict patient and wound management needs when wounds are covered on consent) including the following: erythema only, skin graft required, use of Entonox. The standard medical treatment received by patients was not altered by study participation. The Queensland Children’s Health Services (Royal Children’s Hospital) Human Research Ethics Committee and The University of Queensland Ethics Committee approved this RCT and it was registered with the Australian New Zealand Clinical Trials Registry (ACTRN12611000913976).
Interventions Administrative staff external to the study used a portable computerized random number generator to randomly assign participants to one of two groups: 1) to receive the Ditto intervention [34], consisting of a computerized multimodal device delivering the procedural preparation in the waiting room, and then the distraction intervention during wound care procedures; or 2) to receive standard preparation and standard distraction (such as videos, books, toys, television, and/or parental soothing).
Data Collection Part 1. Prospective data collection: RCT Recruitment and consent of participants occurred at the first change of dressing, where burn wounds were scanned with the Moor LDI2-BI2 laser Doppler imager (Moor Instruments Limited, Devon, UK) to measure burn depth, enabling accurate wound comparison across participants. Additional clinical characteristics collected included the following: mode of burn, TBSA of burn, site of burn, first aid administered, and skin tone. Visitrak grids (Smith þ Nephew, London, UK) were used at every dressing change to determine the percentage of wound re-epithelialization, the primary outcome measure of the RCT. Visitrak was the chosen measure of re-epithelialization for this study (over blinded reviewers determining the percentage of reepithelialization from photos taken at every dressing change [11]) due to Visitrak’s valid and reliable wound measurement technique [35,36] using mathematical calculations rather than clinical judgment. Clinical cost data collected at every dressing change until burn re-epithelialization included the following: dressing type (primarily Acticoat or Acticoat 7, with or without Mepitels) determined by consultant clinical judgment, and quantity of dressings applied; pharmacological pain relief administered (narcotic [oxycodone opioid, 0.1–0.2 mg/kg]; narcotic combined [oxycodone and either paracetamol or ibuprofen; or codeine and paracetamol]; or non-narcotic analgesia [paracetamol and/or ibuprofen]); and nursing and consultant time taken per patient. Caregivers completed a general demographic questionnaire (including pertinent cost-related questions regarding transport, parking, days off work due to attending burn center appointments, employment benefits received, highest education level, employment status, family income per annum, and postcodes to establish socioeconomic status [using Socio-Economic Indexes for Areas] [37]).
Effect Estimation The number of days from injury until more than 95% reepithelialization was used as a measure of the intervention effect. Minimizing the number of days to re-epithelialize reduces the likelihood of hypertrophic scarring [31,32,38] and therefore may reduce the cost to the health system and the patient.
633
PART TWO Retrospective Data Collection : Scar Management
PART ONE Prospective Data Collection : RCT
VALUE IN HEALTH 18 (2015) 631–637
*
Change of Dressing 1
Consented & Randomized (n = 117)
* Standard group (n = 62)
(n = 55)
** Standard group (n = 40)
group (n = 35)
*** Standard group (n = 19)
group (n = 9)
Dressing Changes
Received scar management (n = 11)
12 Months Post Re-epithelialization
Received scar management (n = 6)
Participants consented and randomized in the burn center waiting room prior to dressing removal
** occurring concurrently; early discharge to local regional hospital ***
Assessed for scar management (re-epithelialization >14 days)
Fig. 1
Burn Re-epithelialization Costs
Part 2. Retrospective data collection: Scar management costs
All data during the re-epithelialization phase of healing were collected prospectively at every dressing change from the health care provider and participant perspectives. Direct health care costs included the following: administered narcotic medication; size, type, and quantity of dressings; nursing time to remove and apply dressings; and consultant review time. Nursing wages were based on 2012 Queensland Health Clinical Nurse Grade 6 with a standard (for Queensland Health employees) addition of 25.85% on-costs [39]. Consultant wages were based on 2012 Queensland Health Medical Officer Level 2 (Staff Specialist Level 26) with 25.85% on-costs added [40] (see Appendix A in Supplemental Materials found at: http://dx.doi.org/10.1016/j.jval.2015.04.011). Direct non–health care costs included travel and hospital parking costs. Families were asked about their mode of transport and time spent traveling, and distances traveled to and from appointments by each patient were calculated via postcodes. Travel costs were calculated on the basis of Royal Automobile Club of Queensland estimates of running a medium-size car to be AU$0.76/km [41]. Postcodes were also used to determine public transport zones to establish travel costs. Indirect non-health care costs of caregiver(s) days off work were documented at every dressing change, and used together with their annual family income bracket (recorded on the participant demographic questionnaire) to establish time costs. Annual family income was estimated by calculating the arithmetic mean of the lower and upper limits of reported income bracket, of which there were four. In addition, income data from the Australian Bureau of Statistics household income distribution for Queensland in 2011 to 2012 [42] were used to establish overall minimum and maximum limits, such that means of the lowest and highest brackets could be calculated.
At 1 year after re-epithelialization of the last participant, a retrospective review of charts was conducted to establish whether scar management intervention was received. As a standard guideline, at the 12-month scar management review, therapists are able to determine whether scar management is having an effect and whether further scar management is warranted. The costs associated with each individual patient’s scar management intervention that occurred beyond 12 months were based on estimations by the treating occupational therapist. If patient charts reported the prescription of standard and/or customized pressure garments, patient order form records were obtained to establish exact costs. Documented use of silicone moulds alongside pressure garments was included in direct health care costs on the basis of product cost and occupational therapist’s estimation of average moulds made per tub. Tubigrip product costs were used and amounts given were estimated on the basis of the age of the child and the TBSA of burn. Documentation of prescribed silicone products including Kelo-cote cream and Mepiform sheets, together with standard prescribing practices within the burn center, was used to calculate amounts and costs of these products. The individual patient’s hospital record number was used to track the exact cost for steroid injections received. Occupational therapy scar management appointments are a standard 30 minutes. Scar management staff costs were calculated on the basis of documented therapy appointments (using 2012 Queensland Health, Health Professional Level 4 wage rate and 25.85% on-costs added [43]; Appendix A). Consultant reviews following scar management were based on an estimated average of 15-minute appointments (and wages were based on Queensland Health Medical Officer Level Two,
634
VALUE IN HEALTH 18 (2015) 631–637
MO2/L26, with 25.85% on-costs added [40]). Direct non–health care costs included travel to and from documented scar management appointments and parking costs (extrapolated from RCT data of parent-reported travel). Indirect non–health care costs of parental wages lost were estimated on the basis of data collected from parental reports of time spent during the previous hospital visits for burn dressing changes.
Statistical Analysis Analyses were carried out with a complete case data set for the primary outcome of re-epithelialization. When children had more than one burn, days to re-epithelialization were based on the deepest burn (determined by the laser Doppler imager depth variable). Thus, a new day to re-epithelialize variable was generated to identify health care costs per patient, not per burn. Missing data from cost variables were inputted using mean imputation. Demographic and clinical characteristics across groups were compared with chi-square tests or t tests for normally distributed data and Mann-Whitney U tests when variables were not normally distributed. Incremental costeffectiveness ratios (ICERs) were calculated from the ratio of mean cost differences between the two groups (the Ditto intervention and standard practice) divided by the difference in effect of days taken for burns to re-epithelialize. Nonparametric bootstrapping was performed using 10,000 iterations, which achieved convergence of the standard error and a stable estimate of the variance of the ICER. This statistical method uses large numbers of simulated sample data to calculate the uncertainty around the sampling distribution of the cost and effect estimates [44]. A costeffectiveness plane and cost-effectiveness acceptability curve were used to display the results. Sensitivity analysis was performed to test the robustness of the results. One-way sensitivity analysis was conducted on all probabilities and ratios by varying one variable at a time; these included removing the steroid injection variable because only one patient underwent this treatment as well as removing the Ditto variable and testing an alternate Ditto Lite variable, the newest version of the Ditto that came on the market after the RCT had commenced. A 3% discount rate was applied to all costs that extended beyond a period of 1 year [6] to calculate costs in 2012 Australian dollars. Summary statistics of data not normally distributed were reported in median and interquartile ranges, rather than as means and SDs. The statistical analyses were conducted using the Stata statistical software package (StataCorp, College Station, Texas) [45] and Excel 2010 (Microsoft Pty. Limited, North Ryde, NSW, Australia).
Results One hundred seventeen children presenting to the Burn Centre met the eligibility criteria and consent for participation was obtained. At the second stage of eligibility screening (establishing patient and wound management needs), 42 participants were found to be ineligible. Five patients were lost to follow-up, and three patients were discontinued because of the inability to collect data on multiple participants at the same time. A total of 75 children were included in the analysis (Fig. 1), 40 in the standard group and 35 in the Ditto group. Baseline demographic and clinical characteristics were comparable across groups (Table 1). The sample was typical of our outpatient burn population, with most of the patients being male (60% standard group, 57% Ditto group), presenting with a partial-thickness burn, and a higher proportion of scalds as the mechanism of injury (55% standard group, 51% Ditto group). Only two variables were missing from the primary outcome variable, days to re-
epithelialize. On average, the cost for a typical pediatric burn outpatient presenting with a superficial partial-thickness burn who receives standard care (Table 2) is approximately AU$1000. Refer to Appendix A for unit costs.
Clinical Outcome Children who received the Ditto procedural preparation and distraction intervention during burn wound care procedures reepithelialized on average 3 days faster (P ¼ 0.033) than children in the standard practice group (Table 2). This reduction in reepithelialization time translated to a smaller number of children requiring scar management in the Ditto group. In the standard practice group, 19 (47.5%) children re-epithelialized in 14 or more days and were therefore assessed for scar management, of which 11 (27.5%) went on to receive the scar management intervention. Nine (25.7%) children in the Ditto group re-epithelialized in 14 or more days and were assessed for scar management. Six (17.1%) of these children went on to receive the scar management intervention. Three of these participants continued to receive the scar management intervention beyond 1 year after reepithelialization.
Cost-Effectiveness Analysis On average, direct health care costs, direct non–health care costs, and overall costs per child were lower in the Ditto group (Table 2). The standard group had, on average, lower indirect non–health care costs (Table 2). Parents of children in the standard group reported fewer days off work taken to attend appointments, with the number of work days missed by fathers being significantly less (standard group mean ¼ 0.16 days, Ditto group mean ¼ 1.2 days, P o 0.001) in the standard group than in the Ditto group. The mean total cost of direct health care per patient was AU $475 ⫾ A$1073 in the standard group, compared with AU$185 ⫾ $200 in the Ditto group (Table 2). Itemizing the direct health care costs, the highest percentage of cost was for customized pressure garments for the standard group and burn dressings for the Ditto group (see Appendix A). Cost differences are explained in the reduction in days for burns to re-epithelialize, which translated to fewer patients in the Ditto group requiring the scar management of pressure garments, silicone products, and steroid injections. Comparing the Ditto intervention against standard care on a cost-effectiveness plane showed that 77.6% of the cost/effect pairs were within the southeast quadrant (Fig. 2). This is a strong indication that the Ditto intervention is more effective and costs less than standard care (i.e., cost-saving). The percentage of incremental cost/effect pairs located in the northeast quadrant, indicating that the Ditto intervention is more effective and more costly, was 21.6%. The point estimate gave us a base-case incremental cost of –AU$194 per child, with a mean incremental effect of 3 days reduction to re-epithelialize. The mean ICER was dominant (95% uncertainty interval ¼ dominant to AU$207) for every 1-day reduction in days to re-epithelialize when children had access to the Ditto intervention, over standard practice (Table 3). The cost-effectiveness acceptability curve shows that the probability of the Ditto intervention being cost-effective (or cost-saving) would be 95% if the willingness-to-pay threshold were AU$120 per 1-day reduction in re-epithelialization (Fig. 3). Sensitivity analysis confirmed that the results were robust. The influence of the outlying high cost of steroid injections, received by only one patient in the standard care group as a scar management intervention, was tested by removing the cost from the analysis. Comparable results were obtained in this sensitivity analysis (see Appendix B in Supplemental Materials found at: http://dx.doi.org/10.1016/j.jval.2015.04.011). In addition, the cost
635
VALUE IN HEALTH 18 (2015) 631–637
Table 1 – Demographic and clinical characteristics. Characteristic Age (y), mean SD Sex, n (%) Male Female Ethnicity, n (%) Caucasian Non-Caucasian Family income (A$), n (%) o26,000 26,000–$67,500 67,500–$115,000 4115,000 SEIFA, n (%) High Medium Low Mechanism, n (%) Scald Contact Flame Friction Chemical Depth (PU), mean ⫾ SD Mean PU of burn area Deepest PU in burn area
Standard group (n ¼ 40)
Ditto group (n ¼ 35)
P
7.8 ⫾ 4.79
8.1 ⫾ 4.77
0.683*
24 (60) 16 (40)
20 (57) 15 (43)
0.802†
32 (80) 8 (20)
32 (91) 3 (9)
0.393†
9 6 13 10
5 8 12 8
0.718†
(24) (16) (34) (26)
(15) (24) (36) (24)
4 (10) 14 (35) 22 (55)
4 (11.5) 13 (37) 18 (51.5)
0.949†
22 12 1 4 1
18 12 3 2 0
0.601†
(55) (30) (2.5) (10) (2.5)
1107.02 ⫾ 299.30 598.09 ⫾ 182.32
(51) (34) (9) (6) (0)
1113.79 ⫾ 291.87 605.92 ⫾ 120.81
0.418* 0.850*
PU, perfusion units of blood flow (laser Doppler images measuring burn depth); SEIFA, Socio-Economic Indexes for Areas. * t test. † chi-square.
of the newest version of the intervention device, the “Ditto Lite,” was analyzed and yielded very similar results to the original results (Appendix B). The cost-effectiveness plane showed that 78% of the cost/effect pairs were within the southeast quadrant and 21% in the northeast quadrant.
Discussion This is the first comprehensive economic evaluation that examined the cost-effectiveness of a nonpharmacological intervention
targeting a pediatric burn outpatient population. The Ditto device intervention resulted in a substantial gain, hastening burn wound re-epithelialization [33], confirming the results of a preliminary study [19]. Despite the initial outlay cost for the Ditto device, when considered against the effects, intervention was found to be more effective and less costly than standard care. The benefits greatly outweigh the costs, with 77% of the cost/effect pairs falling in the more effective and less costly southeastern quadrant. The speed of burn re-epithelialization is paramount for the patient [46] and the health system. A reduction of 3 days could
Table 2 – Costs and outcomes of providing the Ditto intervention and standard care to participants during burn re-epithelialization and scar management. Costs Total direct health care cost Direct health care cost per child, mean ⫾ SD Total direct non–health care cost Direct non–health care cost per child, mean ⫾ SD Total indirect non–health care cost Indirect non–health care cost per child, mean ⫾ SD Total cost for therapy group Total cost per child, mean ⫾ SD Outcome Days to re-epithelialize Note. Values are in Australian dollars (A$). CI, confidence interval. * Mann-Whitney test.
Standard group costs
Ditto group costs
P
19,003.38 475.08 ⫾ 1073.45 11,856.77 296.42 ⫾ 291.30 10,753.44 268.84 ⫾ 369.84 41,613.59 1040.34 ⫾ 1410.32
6,483.64 185.25 ⫾ 199.96 8,287.92 236.80 ⫾ 193.96 14,830.88 423.74 ⫾ 551.56 29,602.44 845.78 ⫾ 661.09
Mean (CI)
Mean (CI)
P
14.5 (12.19–16.81)
11.5 (10.26–12.82)
0.094*
0.531* 0.454* 0.196* 0.694*
636
VALUE IN HEALTH 18 (2015) 631–637
Fig. 2 – AU$, Australian Dollars.
Fig. 3 – AU$, Australian Dollars.
mean a burn re-epithelializes within 10 to 14 days of injury, with decreased risk of hypertrophic scar development [31,32]. This may reduce the burden of injury for the patient in terms of psychological and psychosocial implications commonly accompanying burn scarring [4] and for the health system, with regard to costs associated with scar management and rehabilitation. The mean direct health care cost per patient for the Ditto group was AU$185, compared with AU$475 in the standard practice group (Table 2). Limited comparisons can be made with other studies due to the minimal economic evaluations that have been conducted in the area of burn injuries. Of the few studies investigating the cost of burns, the foci have been on inpatient stays and more severe burn injuries requiring grafting and often multiple surgeries [47–51]. These studies are either retrospective reviews or cost analyses that relied on codes of Diagnostic Related Group. Analysis of costs based on hospital charges often grossly overestimate costs [47]. Hospitalizations in Australia resulting from burn injuries, in comparison to the population, were reported to equate to 47.9 cases per 100,000 population per year (based on data collected during the period 1999 to 2004) [52]. In 2000, the hospital costs of burns were estimated to be between AU$38.7 million and A$40.2 million per year [52]. The shift in practice of treating superficial to deep partial-thickness burn injuries predominately on an outpatient basis, together with standard recording of burn data when patients are admitted for longer than 24 hours only [53], highlights that collected burn data represent only a fraction of all pediatric burn injuries treated at Australian and New Zealand burn centers. The burden of burn injuries to the health system and patients is undeniably significant, and this study is the first to report on burn outpatient costs. The strengths of this study include the individualized cost data collected on each patient, the valid measures used in the
assessment of re-epithelialization and burn depth, and the statistical analyses conducted to account for the uncertainty of the ICER of a nonpharmacological intervention in pediatric burn outpatients. A limitation of this study is the small sample size; unequal increments of annual family income brackets used in the parental questionnaire to collect indirect non-health care cost data; and the fact that parental days off work to attend appointments did not consider shift work, full-time mothers, and parents on annual leave. The accuracy of indirect non-health care costs may be improved in future studies with narrower ranges and equal distribution brackets of annual family income. Second, the scar management data were retrospectively collected. Order forms of specialized custom-made pressure garments were obtained, enabling exact costs to be established, and standard 30-minute scar management appointments enabled occupational therapy staff costs to be calculated. The accuracy of costing for silicone products, however, relied on accurate documentation, and product amounts were estimated on the basis of the size of the burn and occupational therapy protocols. Burn care is resource intensive and requires specialist care [3]. With the challenge of health care budget constraints and the complexity of burn care, justifying the inclusion of cutting-edge treatments and interventions within a specialist burn center requires comprehensive economic evaluations to be undertaken. The inclusion of wound care management, scar management, psychosocial, and quality-of-life measures in future economic evaluations is warranted to establish quality and cost-effective practices. Source of financial support: The randomized controlled trial was supported by a grant given to the Royal Children’s Hospital, Brisbane, by Diversionary Therapy Technologies (DTT). Despite this financial support, DTT had no role in the study design, data collection, and analysis of this project, nor will it have any involvement in the publication of results. The principal researcher received a scholarship from the Centre for Children’s Burns and Trauma Research to complete this economic evaluation.
Table 3 – Incremental cost-effectiveness results. Incremental outcomes Incremental cost Incremental effect ICER
Mean
95% UI
194 2.95 Dominant
733 to 243 0.57 to 5.54 Dominant to 207
Note. Values are in Australian dollars (AU$). ICER, incremental cost-effectiveness ratio; UI, uncertainty interval.
Supplemental materials Supplemental material accompanying this article can be found in the online version as a hyperlink at http://dx.doi.org/10.1016/j. jval.2015.04.011 or, if a hard copy of article, at www.valueinhealth journal.com/issues (select volume, issue, and article).
VALUE IN HEALTH 18 (2015) 631–637
R EF E R EN CE S
[1] Burd A, Yuen C. A global study of hospitalized paediatric burn patients. Burns 2005;31:432–8. [2] Mandal A. Quality and cost-effectiveness - effects in burn care. Burns 2007;33:414–7. [3] Stavrou D, Weissman O, Winkler E, et al. Managing the relationship between quality and cost-effective burn care. Burns 2011;37:367–76. [4] Maskell J, Newcombe P, Martin G. Psychosocial functioning differences in pediatric burn survivors compared with healthy norms. J Burn Care Res 2013;34:465–76. [5] Eyles P, Browne G, Byrne C, et al. Methodological problems in studies of burn survivors and their psychosocial prognosis. Burns Incl Therm Inj 1984;10:427–33. [6] Sanchez JL, Pereperez SB, Bastida JL, et al. Cost-utility analysis applied to the treatment of burn patients in a specialized center. Arch Surg 2007;142:50–7; discussion 57. [7] Griffiths HR, Thornton KL, Clements CM, et al. The cost of a hot drink scald. Burns 2006;32:372–4. [8] Silverstein P, Heimbach D, Meites H, et al. An open, parallel, randomized, comparative, multicenter study to evaluate the costeffectiveness, performance, tolerance, and safety of a silver-containing soft silicone foam dressing (intervention) vs silver sulfadiazine cream. J Burn Care Res 2011;32:617–26. [9] Carayanni V, Tsati E, Spyropoulou G, et al. Comparing oil based ointment versus standard practice for the treatment of moderate burns in Greece: a trial based cost effectiveness evaluation. BMC Complement Altern Med 2011;11:122. [10] Cuttle L, Naidu S, Mill J, et al. A retrospective cohort study of Acticoat™ versus Silvazine™ in a paediatric population. Burns 2007;33:701–7. [11] Brown NJ, Rodger S, Ware RS, et al. Efficacy of a children’s procedural preparation and distraction device on healing in acute burn wound care procedures: study protocol for a randomized controlled trial. Trials 2012;13:238. [12] Hoffman HG, Richards TL, Van Oostrom T, et al. The analgesic effects of opioids and immersive virtual reality distraction: evidence from subjective and functional brain imaging assessments. Anesth Analg 2007;105:1776–83. [13] Hoffman HG, Richards TL, Coda B, et al. Modulation of thermal painrelated brain activity with virtual reality: evidence from fMRI. Neuroreport 2004;15:1245. [14] Hoffman HG, Patterson DR, Soltani M, et al. Virtual reality pain control during physical therapy range of motion exercises for a patient with multiple blunt force trauma injuries. Cyberpsychol Behav 2009;12:47–9. [15] Hoffman HG, Patterson DR, Seibel E, et al. Virtual reality pain control during burn wound debridement in the hydrotank. Clin J Pain 2008;24:299–304. [16] Hoffman HG, Chambers GT, Meyer WJ III, et al. Virtual reality as an adjunctive non-pharmacologic analgesic for acute burn pain during medical procedures. Ann Behav Med 2011;41:183–91. [17] Hoffman HG, Patterson DR, Carrougher GJ, et al. Effectiveness of virtual reality-based pain control with multiple treatments. Clin J Pain 2001;17:229–35. [18] Mott J, Bucolo S, Cuttle L, et al. The efficacy of an augmented virtual reality system to alleviate pain in children undergoing burns dressing changes: a randomised controlled trial. Burns 2008;34:803–8. [19] Miller K, Rodger S. Kipping B, et al. A novel technology approach to pain management in children with burns: a prospective randomized controlled trial. Burns 2011;37:395–405. [20] Miller K, Rodger S, Bucolo S, et al. Multi-modal distraction: using technology to combat pain in young children with burn injuries. Burns 2010;36:647–58. [21] Miller K, Kipping B, Rodger S, et al. Attention‐based interventions for the management of pain and distress in young children (3‐12 years) with burn injuries. Cochrane Database Syst Rev 2010(12). [22] Muizzuddin N, Matsui MS, Marenus KD, et al. Impact of stress of marital dissolution on skin barrier recovery: tape stripping and measurement of trans-epidermal water loss (TEWL). Skin Res Technol 2003;9:34–8. [23] Robles TF. Stress, social support, and delayed skin barrier recovery. Psychosom Med 2007;69:807–15. [24] Garg A, Chren MM, Sands LP, et al. Psychological stress perturbs epidermal permeability barrier homeostasis: implications for the pathogenesis of stress-associated skin disorders. Arch Dermatol 2001;137:53–9. [25] Ebrecht M, Hextall J, Kirtley LG, et al. Perceived stress and cortisol levels predict speed of wound healing in healthy male adults. Psychoneuroendocrinology 2004;29:798–809. [26] Kiecolt-Glaser JK, Marucha PT, Malarkey WB, et al. Slowing of wound healing by psychological stress. Lancet 1995;346:1194–96.
637
[27] Marucha PT, Kiecolt-Glaser JK, Favagehi M. Mucosal wound healing is impaired by examination stress. Psychosom Med 1998;60:362–5. [28] Broadbent E, Petrie KJ, Alley PG, et al. Psychological stress impairs early wound repair following surgery. Psychosom Med 2003;65: 865–9. [29] Broadbent E, Kahokehr A, Booth RJ, et al. A brief relaxation intervention reduces stress and improves surgical wound healing response: a randomised trial. Brain Behav Immun 2012;26:212–7. [30] Widgerow AD, Kalaria S. Pain mediators and wound healing— establishing the connection. Burns 2012;38:951–9. [31] Deitch EA, Wheelahan TM, Rose MP, et al. Hypertrophic burn scars: analysis of variables. J Trauma 1983;23:895. [32] Cubison T, Pape SA, Parkhouse N. Evidence for the link between healing time and the development of hypertrophic scars (HTS) in paediatric burns due to scald injury. Burns 2006;32:992–9. [33] Brown NJ, Kimble RM, Rodger S, et al. Play and heal: randomized controlled trial of Ditto™ intervention efficacy on improving re-epithelialization in pediatric burns. Burns 2014;40: 204–13. [34] Miller K, Bucolo S, Patterson E, et al. The emergence of multi-modal distraction as a paediatric pain management tool. Stud Health Technol Inform 2008;132:287–92. [35] Sugama J, Matsui Y, Sanada H, et al. A study of the efficiency and convenience of an advanced portable Wound Measurement System (VISITRAK). J Clin Nurs 2007;16:1265–9. [36] Haghpanah S, Bogie K, Wang X, et al. Reliability of electronic versus manual wound measurement techniques. Arch Phys Med Rehabil 2006;87:1396–402. [37] Australian Bureau of Statistics. 2033.0.55.001 - Census of Population and Housing: Socio-Economic Indexes for Areas (SEIFA), Australia, 2011. Available from: http://www.abs.gov.au/AUSSTATS/[email protected]/ DetailsPage/2033.0.55.0012011. [Accessed September 18, 2011]. [38] Gangemi EN, Gregori D, Berchialla P, et al. Epidemiology and risk factors for pathologic scarring after burn wounds. Arch Facial Plast Surg 2008;10:93–102. [39] Queensland Government. Nursing Wage Rates: as at 1 April 2013. Available from: http://www.health.qld.gov.au/hrpolicies/wage_rates/ nursing.asp. [Accessed October 10, 2013]. [40] Queensland Government. District Health Services: Senior Medical Officers’ and Resident Medical Officers’ Award - State 2012. Available from: http://www.health.qld.gov.au/hrpolicies/wage_rates/medical. asp. [Accessed October 10, 2013]. [41] Royal Automobile Club of Queensland. Vehicle Running Costs 2013. Available from: http://www.racq.com.au/motoring/cars/car_economy/ vehicle_running_costs_2013. [Accessed October 10, 2013]. [42] Australian Bureau of Statistics. 6523.0 - Household Income and Income Distribution, Australia, 2011-12. Available from: http://www.abs.gov.au/ AUSSTATS/[email protected]/DetailsPage/6523.02011-12. [Accessed September 23, 2013]. [43] Queensland Government. Queensland Health Wage Rates. Australia Health Practitioners’ Stream. Queensland Government, 2013. Available from: http://www.health.qld.gov.au/hrpolicies/wage_rates/. [Accessed October 10, 2013]. [44] Briggs AH, Wonderling DE, Mooney CZ. Pulling cost-effectiveness analysis up by its bootstraps: a non-parametric approach to confidence interval estimation. Health Econ 1997;6:327–40. [45] StataCorp. Stata Statistical Software: Release 12. College Station, TX: StataCorp LP, 2011. [46] Brown NJ, Kimble RM, Gramotnev G, et al. Predictors of reepithelialization in pediatric burn. Burns 2014;40:751–8. [47] Klein MB, Hollingworth W, Rivara FP, et al. Hospital costs associated with pediatric burn injury. J Burn Care Res 2008;29:632–7. [48] Kai-Yang L, Shi-Hui Z, Hong-Tai T, et al. The direct hospitalisation costs of paediatric scalds: 2-year results of a prospective case series. Burns 2009;35:738–45. [49] Carey K, Kazis LE, Lee AF, et al. Measuring the cost of care for children with acute burn injury. J Trauma Acute Care Surg 2012;73: S229–33. [50] Mirastschijski U, Sander J-T, Weyand B, et al. Rehabilitation of burn patients: an underestimated socio-economic burden. Burns 2013;39:262–8. [51] Hemington-Gorse SJ, Potokar TS, Drew PJ, et al. Burn care costing: the Welsh experience. Burns 2009;35:378–82. [52] Australian Government. Australian Institute of Health and Welfare. Burns and Scalds, October 2006. Available from: http://www.nisu. flinders.edu.au/pubs/reports/2006/injcat92.pdf. [Accessed September 4, 2013]. [53] Gabbe B, Picton N, Loh I, Watterson D. Bi-National Burns Registry Annual Report 1st July 2011 – 30th June 2012. In: Department of Epidemiology and Preventive Medicine, ed. Melbourne, Australia: Monash University, 2012.
Available online at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/jval
Cost-Effectiveness of a Nonpharmacological Intervention in Pediatric Burn Care Nadia J. Brown, BAppSC(OT), MAppSc(OT), PhD1,*, Michael David, MEd, MEpi, MSc, PhD2,3, Leila Cuttle, BSc(Hons), PhD1,4, Roy M. Kimble, MD (Res), MBChB, FRCS, FRACS1, Sylvia Rodger, BOccThy, MEdSt, PhD5, Hideki Higashi, BA, MPH, MSc(HlthEcon), PhD6 1
Centre for Children’s Burns and Trauma Research, Queensland Children’s Medical Research Institute, The University of Queensland, Royal Children’s Hospital, Brisbane, Australia; 2Queensland Children’s Medical Research Institute, The University of Queensland, Royal Children’s Hospital, Brisbane, Australia; 3School of Population Health, The University of Queensland, Brisbane, Australia; 4Tissue Repair and Regeneration Program, Institute of Health and Biomedical Innovation, Queensland University of Technology, Brisbane, Australia; 5 Division of Occupational Therapy, School of Health and Rehabilitation Sciences, The University of Queensland, Brisbane, Australia; 6 Institute for Health Metrics and Evaluation, University of Washington, Seattle, WA
AB STR A CT
Objective: To report the cost-effectiveness of a tailored handheld computerized procedural preparation and distraction intervention (Ditto) used during pediatric burn wound care in comparison to standard practice. Methods: An economic evaluation was performed alongside a randomized controlled trial of 75 children aged 4 to 13 years who presented with a burn to the Royal Children’s Hospital, Brisbane, Australia. Participants were randomized to either the Ditto intervention (n ¼ 35) or standard practice (n ¼ 40) to measure the effect of the intervention on days taken for burns to re-epithelialize. Direct medical, direct nonmedical, and indirect cost data during burn re-epithelialization were extracted from the randomized controlled trial data and combined with scar management cost data obtained retrospectively from medical charts. Nonparametric bootstrapping was used to estimate statistical uncertainty in cost and effect differences and cost-effectiveness ratios. Results: On average, the Ditto intervention reduced the time to re-epithelialize by 3 days at AU$194 less cost for each patient compared with standard practice. The
incremental cost-effectiveness plane showed that 78% of the simulated results were within the more effective and less costly quadrant and 22% were in the more effective and more costly quadrant, suggesting a 78% probability that the Ditto intervention dominates standard practice (i.e., cost-saving). At a willingness-to-pay threshold of AU$120, there is a 95% probability that the Ditto intervention is cost-effective (or cost-saving) against standard care. Conclusions: This economic evaluation showed the Ditto intervention to be highly cost-effective against standard practice at a minimal cost for the significant benefits gained, supporting the implementation of the Ditto intervention during burn wound care. Keywords: burns, child, cost-effectiveness, economic evaluation, nonpharmacological intervention, randomized controlled trial, reepithelialization.
Introduction
practice being more effective than another. Changes in practice need to be quantified in monetary terms with an economic evaluation to determine the feasibility of implementing new cutting-edge treatments and interventions [3] and to justify setup costs to hospital administrators. Psychosocial costs to the patient include the pain, anxiety, and stress from the trauma of the burn injury itself and the associated wound care procedures and may encompass additional psychosocial issues resulting from the physical disfigurement of hypertrophic scarring. Severity of injury does not predict psychological costs to the patient [4]. No quality-of-life measure has been specifically developed and
It is estimated that more than 500,000 children worldwide are hospitalized with a burn injury every year [1]. Burn care is undeniably expensive and arguably has one of the highest financial costs in pediatric health care, yet limited health economic research has been conducted in the burns arena. External financial pressures on health care budgets create much deliberation and contentious debates over balancing quality health care versus less costly service delivery [2]. There is a great need for health research to extend beyond providing evidence of one
& 2015 Published by Elsevier Inc. on behalf of International Society for Pharmacoeconomics and Outcomes Research (ISPOR).
Conflict of interest: One of the supervisors of the trial, Roy M. Kimble, holds options with Diversionary Therapy Technologies (DTT); however, he did not stand to lose or gain financially or personally from the results during the clinical trial period and the time of submission. The principal researcher has no financial interest in the Ditto device or DTT. * Address correspondence to: Nadia J. Brown, Centre for Children’s Burns and Trauma Research, Queensland Children’s Medical Research Institute, Level 4, Foundation Building, Royal Children’s Hospital, Brisbane, QLD 4029, Australia. E-mail: [email protected]. 1098-3015$36.00 – see front matter & 2015 Published by Elsevier Inc. on behalf of International Society for Pharmacoeconomics and Outcomes Research (ISPOR). http://dx.doi.org/10.1016/j.jval.2015.04.011
632
VALUE IN HEALTH 18 (2015) 631–637
validated for the pediatric burn population [3,5]. Financial accountability that extends beyond acute wound care and scar management and includes the complete long-term psychosocial costs to the patient makes the “true” costs of burns unquantifiable. The most comprehensive cost-effectiveness analysis identified in the burn literature was a prospective study over a 5-year period conducted in the Burn Centre of Valencia, Spain [6]. This study incorporated incremental cost-utility ratio analysis to determine the cost and quality of life of burn patients. This cost analysis largely depended on Diagnostic Related Groups, used silver sulfadiazine in the treatment of burn wounds, and was carried out for primarily an adult inpatient population with a mean age of 40 years and mean total body surface area (TBSA) burns of 18.2%. General Diagnostic Related Group and Healthcare Resource Group cost coding provide a generic snapshot of costs and are not sensitive measures for picking up differences across individuals [7]. Two studies were identified in the burn literature that performed a cost-effectiveness evaluation on minor burns encompassing smaller TBSA (mean r10%) and primarily partialthickness depth [8,9]. Both studies compared burn dressings, which are not standard practice in burn centers in developed countries. Advances in the technology of burn dressings in recent years have changed standard practice from twice-daily applications of silver sulfadiazine and similar ointments or creams to improved silver-impregnated dressings such as Acticoat [10]. This change in practice has seen a shift from inpatient to outpatient management of the small partial-thickness burn injury, in which dressings are changed every 3 to 7 days. No economic evaluation has been performed in pediatric burn outpatients that uses advanced silver-impregnated dressings only, or that uses validated measures of re-epithelialization and wound depth. This study aimed to establish the cost-effectiveness of a computerized procedural preparation and distraction intervention Ditto, which was compared with standard practice, in a randomized controlled trial (RCT) measuring burn reepithelialization [11]. Use of virtual reality and computerized nonpharmacological interventions plays a significant role in dampening pain perception when used as an adjunct to pharmacological intervention during burn care and rehabilitation [12–21]. Several studies report on the effect of stress in delaying wound healing [22–29], and cellular links between pain and delayed wound healing are beginning to be explored [30]. The risk of hypertrophic scar formation is known to increase when wound re-epithelialization extends beyond 14 days [31,32]. Accelerating re-epithelialization is the prime objective of burn care, and has significant cost benefits to the health system and the patient.
Methods Setting and Participants Data were collected from the Stuart Pegg Paediatric Burns Centre at the Royal Children’s Hospital, Brisbane, which is a tertiary pediatric burn referral center servicing approximately 800 new patients per year. Data collection occurred in two parts: 1) RCT [11,33] data collected from August 2011 to August 2012 and 2) retrospective review of medical charts from August 2011 to September 2013 to obtain scar management data 12 months after re-epithelialization (Fig. 1). Children were included if they were aged 4 years to 13 years and had an acute burn injury of TBSA less than 15%. Children were ineligible if they were non-English speaking, had a diagnosed condition or illness in addition to a burns injury, or had a history of suspected child abuse. On the first dressing removal, further study eligibility was determined
(because of the inability to predict patient and wound management needs when wounds are covered on consent) including the following: erythema only, skin graft required, use of Entonox. The standard medical treatment received by patients was not altered by study participation. The Queensland Children’s Health Services (Royal Children’s Hospital) Human Research Ethics Committee and The University of Queensland Ethics Committee approved this RCT and it was registered with the Australian New Zealand Clinical Trials Registry (ACTRN12611000913976).
Interventions Administrative staff external to the study used a portable computerized random number generator to randomly assign participants to one of two groups: 1) to receive the Ditto intervention [34], consisting of a computerized multimodal device delivering the procedural preparation in the waiting room, and then the distraction intervention during wound care procedures; or 2) to receive standard preparation and standard distraction (such as videos, books, toys, television, and/or parental soothing).
Data Collection Part 1. Prospective data collection: RCT Recruitment and consent of participants occurred at the first change of dressing, where burn wounds were scanned with the Moor LDI2-BI2 laser Doppler imager (Moor Instruments Limited, Devon, UK) to measure burn depth, enabling accurate wound comparison across participants. Additional clinical characteristics collected included the following: mode of burn, TBSA of burn, site of burn, first aid administered, and skin tone. Visitrak grids (Smith þ Nephew, London, UK) were used at every dressing change to determine the percentage of wound re-epithelialization, the primary outcome measure of the RCT. Visitrak was the chosen measure of re-epithelialization for this study (over blinded reviewers determining the percentage of reepithelialization from photos taken at every dressing change [11]) due to Visitrak’s valid and reliable wound measurement technique [35,36] using mathematical calculations rather than clinical judgment. Clinical cost data collected at every dressing change until burn re-epithelialization included the following: dressing type (primarily Acticoat or Acticoat 7, with or without Mepitels) determined by consultant clinical judgment, and quantity of dressings applied; pharmacological pain relief administered (narcotic [oxycodone opioid, 0.1–0.2 mg/kg]; narcotic combined [oxycodone and either paracetamol or ibuprofen; or codeine and paracetamol]; or non-narcotic analgesia [paracetamol and/or ibuprofen]); and nursing and consultant time taken per patient. Caregivers completed a general demographic questionnaire (including pertinent cost-related questions regarding transport, parking, days off work due to attending burn center appointments, employment benefits received, highest education level, employment status, family income per annum, and postcodes to establish socioeconomic status [using Socio-Economic Indexes for Areas] [37]).
Effect Estimation The number of days from injury until more than 95% reepithelialization was used as a measure of the intervention effect. Minimizing the number of days to re-epithelialize reduces the likelihood of hypertrophic scarring [31,32,38] and therefore may reduce the cost to the health system and the patient.
633
PART TWO Retrospective Data Collection : Scar Management
PART ONE Prospective Data Collection : RCT
VALUE IN HEALTH 18 (2015) 631–637
*
Change of Dressing 1
Consented & Randomized (n = 117)
* Standard group (n = 62)
(n = 55)
** Standard group (n = 40)
group (n = 35)
*** Standard group (n = 19)
group (n = 9)
Dressing Changes
Received scar management (n = 11)
12 Months Post Re-epithelialization
Received scar management (n = 6)
Participants consented and randomized in the burn center waiting room prior to dressing removal
** occurring concurrently; early discharge to local regional hospital ***
Assessed for scar management (re-epithelialization >14 days)
Fig. 1
Burn Re-epithelialization Costs
Part 2. Retrospective data collection: Scar management costs
All data during the re-epithelialization phase of healing were collected prospectively at every dressing change from the health care provider and participant perspectives. Direct health care costs included the following: administered narcotic medication; size, type, and quantity of dressings; nursing time to remove and apply dressings; and consultant review time. Nursing wages were based on 2012 Queensland Health Clinical Nurse Grade 6 with a standard (for Queensland Health employees) addition of 25.85% on-costs [39]. Consultant wages were based on 2012 Queensland Health Medical Officer Level 2 (Staff Specialist Level 26) with 25.85% on-costs added [40] (see Appendix A in Supplemental Materials found at: http://dx.doi.org/10.1016/j.jval.2015.04.011). Direct non–health care costs included travel and hospital parking costs. Families were asked about their mode of transport and time spent traveling, and distances traveled to and from appointments by each patient were calculated via postcodes. Travel costs were calculated on the basis of Royal Automobile Club of Queensland estimates of running a medium-size car to be AU$0.76/km [41]. Postcodes were also used to determine public transport zones to establish travel costs. Indirect non-health care costs of caregiver(s) days off work were documented at every dressing change, and used together with their annual family income bracket (recorded on the participant demographic questionnaire) to establish time costs. Annual family income was estimated by calculating the arithmetic mean of the lower and upper limits of reported income bracket, of which there were four. In addition, income data from the Australian Bureau of Statistics household income distribution for Queensland in 2011 to 2012 [42] were used to establish overall minimum and maximum limits, such that means of the lowest and highest brackets could be calculated.
At 1 year after re-epithelialization of the last participant, a retrospective review of charts was conducted to establish whether scar management intervention was received. As a standard guideline, at the 12-month scar management review, therapists are able to determine whether scar management is having an effect and whether further scar management is warranted. The costs associated with each individual patient’s scar management intervention that occurred beyond 12 months were based on estimations by the treating occupational therapist. If patient charts reported the prescription of standard and/or customized pressure garments, patient order form records were obtained to establish exact costs. Documented use of silicone moulds alongside pressure garments was included in direct health care costs on the basis of product cost and occupational therapist’s estimation of average moulds made per tub. Tubigrip product costs were used and amounts given were estimated on the basis of the age of the child and the TBSA of burn. Documentation of prescribed silicone products including Kelo-cote cream and Mepiform sheets, together with standard prescribing practices within the burn center, was used to calculate amounts and costs of these products. The individual patient’s hospital record number was used to track the exact cost for steroid injections received. Occupational therapy scar management appointments are a standard 30 minutes. Scar management staff costs were calculated on the basis of documented therapy appointments (using 2012 Queensland Health, Health Professional Level 4 wage rate and 25.85% on-costs added [43]; Appendix A). Consultant reviews following scar management were based on an estimated average of 15-minute appointments (and wages were based on Queensland Health Medical Officer Level Two,
634
VALUE IN HEALTH 18 (2015) 631–637
MO2/L26, with 25.85% on-costs added [40]). Direct non–health care costs included travel to and from documented scar management appointments and parking costs (extrapolated from RCT data of parent-reported travel). Indirect non–health care costs of parental wages lost were estimated on the basis of data collected from parental reports of time spent during the previous hospital visits for burn dressing changes.
Statistical Analysis Analyses were carried out with a complete case data set for the primary outcome of re-epithelialization. When children had more than one burn, days to re-epithelialization were based on the deepest burn (determined by the laser Doppler imager depth variable). Thus, a new day to re-epithelialize variable was generated to identify health care costs per patient, not per burn. Missing data from cost variables were inputted using mean imputation. Demographic and clinical characteristics across groups were compared with chi-square tests or t tests for normally distributed data and Mann-Whitney U tests when variables were not normally distributed. Incremental costeffectiveness ratios (ICERs) were calculated from the ratio of mean cost differences between the two groups (the Ditto intervention and standard practice) divided by the difference in effect of days taken for burns to re-epithelialize. Nonparametric bootstrapping was performed using 10,000 iterations, which achieved convergence of the standard error and a stable estimate of the variance of the ICER. This statistical method uses large numbers of simulated sample data to calculate the uncertainty around the sampling distribution of the cost and effect estimates [44]. A costeffectiveness plane and cost-effectiveness acceptability curve were used to display the results. Sensitivity analysis was performed to test the robustness of the results. One-way sensitivity analysis was conducted on all probabilities and ratios by varying one variable at a time; these included removing the steroid injection variable because only one patient underwent this treatment as well as removing the Ditto variable and testing an alternate Ditto Lite variable, the newest version of the Ditto that came on the market after the RCT had commenced. A 3% discount rate was applied to all costs that extended beyond a period of 1 year [6] to calculate costs in 2012 Australian dollars. Summary statistics of data not normally distributed were reported in median and interquartile ranges, rather than as means and SDs. The statistical analyses were conducted using the Stata statistical software package (StataCorp, College Station, Texas) [45] and Excel 2010 (Microsoft Pty. Limited, North Ryde, NSW, Australia).
Results One hundred seventeen children presenting to the Burn Centre met the eligibility criteria and consent for participation was obtained. At the second stage of eligibility screening (establishing patient and wound management needs), 42 participants were found to be ineligible. Five patients were lost to follow-up, and three patients were discontinued because of the inability to collect data on multiple participants at the same time. A total of 75 children were included in the analysis (Fig. 1), 40 in the standard group and 35 in the Ditto group. Baseline demographic and clinical characteristics were comparable across groups (Table 1). The sample was typical of our outpatient burn population, with most of the patients being male (60% standard group, 57% Ditto group), presenting with a partial-thickness burn, and a higher proportion of scalds as the mechanism of injury (55% standard group, 51% Ditto group). Only two variables were missing from the primary outcome variable, days to re-
epithelialize. On average, the cost for a typical pediatric burn outpatient presenting with a superficial partial-thickness burn who receives standard care (Table 2) is approximately AU$1000. Refer to Appendix A for unit costs.
Clinical Outcome Children who received the Ditto procedural preparation and distraction intervention during burn wound care procedures reepithelialized on average 3 days faster (P ¼ 0.033) than children in the standard practice group (Table 2). This reduction in reepithelialization time translated to a smaller number of children requiring scar management in the Ditto group. In the standard practice group, 19 (47.5%) children re-epithelialized in 14 or more days and were therefore assessed for scar management, of which 11 (27.5%) went on to receive the scar management intervention. Nine (25.7%) children in the Ditto group re-epithelialized in 14 or more days and were assessed for scar management. Six (17.1%) of these children went on to receive the scar management intervention. Three of these participants continued to receive the scar management intervention beyond 1 year after reepithelialization.
Cost-Effectiveness Analysis On average, direct health care costs, direct non–health care costs, and overall costs per child were lower in the Ditto group (Table 2). The standard group had, on average, lower indirect non–health care costs (Table 2). Parents of children in the standard group reported fewer days off work taken to attend appointments, with the number of work days missed by fathers being significantly less (standard group mean ¼ 0.16 days, Ditto group mean ¼ 1.2 days, P o 0.001) in the standard group than in the Ditto group. The mean total cost of direct health care per patient was AU $475 ⫾ A$1073 in the standard group, compared with AU$185 ⫾ $200 in the Ditto group (Table 2). Itemizing the direct health care costs, the highest percentage of cost was for customized pressure garments for the standard group and burn dressings for the Ditto group (see Appendix A). Cost differences are explained in the reduction in days for burns to re-epithelialize, which translated to fewer patients in the Ditto group requiring the scar management of pressure garments, silicone products, and steroid injections. Comparing the Ditto intervention against standard care on a cost-effectiveness plane showed that 77.6% of the cost/effect pairs were within the southeast quadrant (Fig. 2). This is a strong indication that the Ditto intervention is more effective and costs less than standard care (i.e., cost-saving). The percentage of incremental cost/effect pairs located in the northeast quadrant, indicating that the Ditto intervention is more effective and more costly, was 21.6%. The point estimate gave us a base-case incremental cost of –AU$194 per child, with a mean incremental effect of 3 days reduction to re-epithelialize. The mean ICER was dominant (95% uncertainty interval ¼ dominant to AU$207) for every 1-day reduction in days to re-epithelialize when children had access to the Ditto intervention, over standard practice (Table 3). The cost-effectiveness acceptability curve shows that the probability of the Ditto intervention being cost-effective (or cost-saving) would be 95% if the willingness-to-pay threshold were AU$120 per 1-day reduction in re-epithelialization (Fig. 3). Sensitivity analysis confirmed that the results were robust. The influence of the outlying high cost of steroid injections, received by only one patient in the standard care group as a scar management intervention, was tested by removing the cost from the analysis. Comparable results were obtained in this sensitivity analysis (see Appendix B in Supplemental Materials found at: http://dx.doi.org/10.1016/j.jval.2015.04.011). In addition, the cost
635
VALUE IN HEALTH 18 (2015) 631–637
Table 1 – Demographic and clinical characteristics. Characteristic Age (y), mean SD Sex, n (%) Male Female Ethnicity, n (%) Caucasian Non-Caucasian Family income (A$), n (%) o26,000 26,000–$67,500 67,500–$115,000 4115,000 SEIFA, n (%) High Medium Low Mechanism, n (%) Scald Contact Flame Friction Chemical Depth (PU), mean ⫾ SD Mean PU of burn area Deepest PU in burn area
Standard group (n ¼ 40)
Ditto group (n ¼ 35)
P
7.8 ⫾ 4.79
8.1 ⫾ 4.77
0.683*
24 (60) 16 (40)
20 (57) 15 (43)
0.802†
32 (80) 8 (20)
32 (91) 3 (9)
0.393†
9 6 13 10
5 8 12 8
0.718†
(24) (16) (34) (26)
(15) (24) (36) (24)
4 (10) 14 (35) 22 (55)
4 (11.5) 13 (37) 18 (51.5)
0.949†
22 12 1 4 1
18 12 3 2 0
0.601†
(55) (30) (2.5) (10) (2.5)
1107.02 ⫾ 299.30 598.09 ⫾ 182.32
(51) (34) (9) (6) (0)
1113.79 ⫾ 291.87 605.92 ⫾ 120.81
0.418* 0.850*
PU, perfusion units of blood flow (laser Doppler images measuring burn depth); SEIFA, Socio-Economic Indexes for Areas. * t test. † chi-square.
of the newest version of the intervention device, the “Ditto Lite,” was analyzed and yielded very similar results to the original results (Appendix B). The cost-effectiveness plane showed that 78% of the cost/effect pairs were within the southeast quadrant and 21% in the northeast quadrant.
Discussion This is the first comprehensive economic evaluation that examined the cost-effectiveness of a nonpharmacological intervention
targeting a pediatric burn outpatient population. The Ditto device intervention resulted in a substantial gain, hastening burn wound re-epithelialization [33], confirming the results of a preliminary study [19]. Despite the initial outlay cost for the Ditto device, when considered against the effects, intervention was found to be more effective and less costly than standard care. The benefits greatly outweigh the costs, with 77% of the cost/effect pairs falling in the more effective and less costly southeastern quadrant. The speed of burn re-epithelialization is paramount for the patient [46] and the health system. A reduction of 3 days could
Table 2 – Costs and outcomes of providing the Ditto intervention and standard care to participants during burn re-epithelialization and scar management. Costs Total direct health care cost Direct health care cost per child, mean ⫾ SD Total direct non–health care cost Direct non–health care cost per child, mean ⫾ SD Total indirect non–health care cost Indirect non–health care cost per child, mean ⫾ SD Total cost for therapy group Total cost per child, mean ⫾ SD Outcome Days to re-epithelialize Note. Values are in Australian dollars (A$). CI, confidence interval. * Mann-Whitney test.
Standard group costs
Ditto group costs
P
19,003.38 475.08 ⫾ 1073.45 11,856.77 296.42 ⫾ 291.30 10,753.44 268.84 ⫾ 369.84 41,613.59 1040.34 ⫾ 1410.32
6,483.64 185.25 ⫾ 199.96 8,287.92 236.80 ⫾ 193.96 14,830.88 423.74 ⫾ 551.56 29,602.44 845.78 ⫾ 661.09
Mean (CI)
Mean (CI)
P
14.5 (12.19–16.81)
11.5 (10.26–12.82)
0.094*
0.531* 0.454* 0.196* 0.694*
636
VALUE IN HEALTH 18 (2015) 631–637
Fig. 2 – AU$, Australian Dollars.
Fig. 3 – AU$, Australian Dollars.
mean a burn re-epithelializes within 10 to 14 days of injury, with decreased risk of hypertrophic scar development [31,32]. This may reduce the burden of injury for the patient in terms of psychological and psychosocial implications commonly accompanying burn scarring [4] and for the health system, with regard to costs associated with scar management and rehabilitation. The mean direct health care cost per patient for the Ditto group was AU$185, compared with AU$475 in the standard practice group (Table 2). Limited comparisons can be made with other studies due to the minimal economic evaluations that have been conducted in the area of burn injuries. Of the few studies investigating the cost of burns, the foci have been on inpatient stays and more severe burn injuries requiring grafting and often multiple surgeries [47–51]. These studies are either retrospective reviews or cost analyses that relied on codes of Diagnostic Related Group. Analysis of costs based on hospital charges often grossly overestimate costs [47]. Hospitalizations in Australia resulting from burn injuries, in comparison to the population, were reported to equate to 47.9 cases per 100,000 population per year (based on data collected during the period 1999 to 2004) [52]. In 2000, the hospital costs of burns were estimated to be between AU$38.7 million and A$40.2 million per year [52]. The shift in practice of treating superficial to deep partial-thickness burn injuries predominately on an outpatient basis, together with standard recording of burn data when patients are admitted for longer than 24 hours only [53], highlights that collected burn data represent only a fraction of all pediatric burn injuries treated at Australian and New Zealand burn centers. The burden of burn injuries to the health system and patients is undeniably significant, and this study is the first to report on burn outpatient costs. The strengths of this study include the individualized cost data collected on each patient, the valid measures used in the
assessment of re-epithelialization and burn depth, and the statistical analyses conducted to account for the uncertainty of the ICER of a nonpharmacological intervention in pediatric burn outpatients. A limitation of this study is the small sample size; unequal increments of annual family income brackets used in the parental questionnaire to collect indirect non-health care cost data; and the fact that parental days off work to attend appointments did not consider shift work, full-time mothers, and parents on annual leave. The accuracy of indirect non-health care costs may be improved in future studies with narrower ranges and equal distribution brackets of annual family income. Second, the scar management data were retrospectively collected. Order forms of specialized custom-made pressure garments were obtained, enabling exact costs to be established, and standard 30-minute scar management appointments enabled occupational therapy staff costs to be calculated. The accuracy of costing for silicone products, however, relied on accurate documentation, and product amounts were estimated on the basis of the size of the burn and occupational therapy protocols. Burn care is resource intensive and requires specialist care [3]. With the challenge of health care budget constraints and the complexity of burn care, justifying the inclusion of cutting-edge treatments and interventions within a specialist burn center requires comprehensive economic evaluations to be undertaken. The inclusion of wound care management, scar management, psychosocial, and quality-of-life measures in future economic evaluations is warranted to establish quality and cost-effective practices. Source of financial support: The randomized controlled trial was supported by a grant given to the Royal Children’s Hospital, Brisbane, by Diversionary Therapy Technologies (DTT). Despite this financial support, DTT had no role in the study design, data collection, and analysis of this project, nor will it have any involvement in the publication of results. The principal researcher received a scholarship from the Centre for Children’s Burns and Trauma Research to complete this economic evaluation.
Table 3 – Incremental cost-effectiveness results. Incremental outcomes Incremental cost Incremental effect ICER
Mean
95% UI
194 2.95 Dominant
733 to 243 0.57 to 5.54 Dominant to 207
Note. Values are in Australian dollars (AU$). ICER, incremental cost-effectiveness ratio; UI, uncertainty interval.
Supplemental materials Supplemental material accompanying this article can be found in the online version as a hyperlink at http://dx.doi.org/10.1016/j. jval.2015.04.011 or, if a hard copy of article, at www.valueinhealth journal.com/issues (select volume, issue, and article).
VALUE IN HEALTH 18 (2015) 631–637
R EF E R EN CE S
[1] Burd A, Yuen C. A global study of hospitalized paediatric burn patients. Burns 2005;31:432–8. [2] Mandal A. Quality and cost-effectiveness - effects in burn care. Burns 2007;33:414–7. [3] Stavrou D, Weissman O, Winkler E, et al. Managing the relationship between quality and cost-effective burn care. Burns 2011;37:367–76. [4] Maskell J, Newcombe P, Martin G. Psychosocial functioning differences in pediatric burn survivors compared with healthy norms. J Burn Care Res 2013;34:465–76. [5] Eyles P, Browne G, Byrne C, et al. Methodological problems in studies of burn survivors and their psychosocial prognosis. Burns Incl Therm Inj 1984;10:427–33. [6] Sanchez JL, Pereperez SB, Bastida JL, et al. Cost-utility analysis applied to the treatment of burn patients in a specialized center. Arch Surg 2007;142:50–7; discussion 57. [7] Griffiths HR, Thornton KL, Clements CM, et al. The cost of a hot drink scald. Burns 2006;32:372–4. [8] Silverstein P, Heimbach D, Meites H, et al. An open, parallel, randomized, comparative, multicenter study to evaluate the costeffectiveness, performance, tolerance, and safety of a silver-containing soft silicone foam dressing (intervention) vs silver sulfadiazine cream. J Burn Care Res 2011;32:617–26. [9] Carayanni V, Tsati E, Spyropoulou G, et al. Comparing oil based ointment versus standard practice for the treatment of moderate burns in Greece: a trial based cost effectiveness evaluation. BMC Complement Altern Med 2011;11:122. [10] Cuttle L, Naidu S, Mill J, et al. A retrospective cohort study of Acticoat™ versus Silvazine™ in a paediatric population. Burns 2007;33:701–7. [11] Brown NJ, Rodger S, Ware RS, et al. Efficacy of a children’s procedural preparation and distraction device on healing in acute burn wound care procedures: study protocol for a randomized controlled trial. Trials 2012;13:238. [12] Hoffman HG, Richards TL, Van Oostrom T, et al. The analgesic effects of opioids and immersive virtual reality distraction: evidence from subjective and functional brain imaging assessments. Anesth Analg 2007;105:1776–83. [13] Hoffman HG, Richards TL, Coda B, et al. Modulation of thermal painrelated brain activity with virtual reality: evidence from fMRI. Neuroreport 2004;15:1245. [14] Hoffman HG, Patterson DR, Soltani M, et al. Virtual reality pain control during physical therapy range of motion exercises for a patient with multiple blunt force trauma injuries. Cyberpsychol Behav 2009;12:47–9. [15] Hoffman HG, Patterson DR, Seibel E, et al. Virtual reality pain control during burn wound debridement in the hydrotank. Clin J Pain 2008;24:299–304. [16] Hoffman HG, Chambers GT, Meyer WJ III, et al. Virtual reality as an adjunctive non-pharmacologic analgesic for acute burn pain during medical procedures. Ann Behav Med 2011;41:183–91. [17] Hoffman HG, Patterson DR, Carrougher GJ, et al. Effectiveness of virtual reality-based pain control with multiple treatments. Clin J Pain 2001;17:229–35. [18] Mott J, Bucolo S, Cuttle L, et al. The efficacy of an augmented virtual reality system to alleviate pain in children undergoing burns dressing changes: a randomised controlled trial. Burns 2008;34:803–8. [19] Miller K, Rodger S. Kipping B, et al. A novel technology approach to pain management in children with burns: a prospective randomized controlled trial. Burns 2011;37:395–405. [20] Miller K, Rodger S, Bucolo S, et al. Multi-modal distraction: using technology to combat pain in young children with burn injuries. Burns 2010;36:647–58. [21] Miller K, Kipping B, Rodger S, et al. Attention‐based interventions for the management of pain and distress in young children (3‐12 years) with burn injuries. Cochrane Database Syst Rev 2010(12). [22] Muizzuddin N, Matsui MS, Marenus KD, et al. Impact of stress of marital dissolution on skin barrier recovery: tape stripping and measurement of trans-epidermal water loss (TEWL). Skin Res Technol 2003;9:34–8. [23] Robles TF. Stress, social support, and delayed skin barrier recovery. Psychosom Med 2007;69:807–15. [24] Garg A, Chren MM, Sands LP, et al. Psychological stress perturbs epidermal permeability barrier homeostasis: implications for the pathogenesis of stress-associated skin disorders. Arch Dermatol 2001;137:53–9. [25] Ebrecht M, Hextall J, Kirtley LG, et al. Perceived stress and cortisol levels predict speed of wound healing in healthy male adults. Psychoneuroendocrinology 2004;29:798–809. [26] Kiecolt-Glaser JK, Marucha PT, Malarkey WB, et al. Slowing of wound healing by psychological stress. Lancet 1995;346:1194–96.
637
[27] Marucha PT, Kiecolt-Glaser JK, Favagehi M. Mucosal wound healing is impaired by examination stress. Psychosom Med 1998;60:362–5. [28] Broadbent E, Petrie KJ, Alley PG, et al. Psychological stress impairs early wound repair following surgery. Psychosom Med 2003;65: 865–9. [29] Broadbent E, Kahokehr A, Booth RJ, et al. A brief relaxation intervention reduces stress and improves surgical wound healing response: a randomised trial. Brain Behav Immun 2012;26:212–7. [30] Widgerow AD, Kalaria S. Pain mediators and wound healing— establishing the connection. Burns 2012;38:951–9. [31] Deitch EA, Wheelahan TM, Rose MP, et al. Hypertrophic burn scars: analysis of variables. J Trauma 1983;23:895. [32] Cubison T, Pape SA, Parkhouse N. Evidence for the link between healing time and the development of hypertrophic scars (HTS) in paediatric burns due to scald injury. Burns 2006;32:992–9. [33] Brown NJ, Kimble RM, Rodger S, et al. Play and heal: randomized controlled trial of Ditto™ intervention efficacy on improving re-epithelialization in pediatric burns. Burns 2014;40: 204–13. [34] Miller K, Bucolo S, Patterson E, et al. The emergence of multi-modal distraction as a paediatric pain management tool. Stud Health Technol Inform 2008;132:287–92. [35] Sugama J, Matsui Y, Sanada H, et al. A study of the efficiency and convenience of an advanced portable Wound Measurement System (VISITRAK). J Clin Nurs 2007;16:1265–9. [36] Haghpanah S, Bogie K, Wang X, et al. Reliability of electronic versus manual wound measurement techniques. Arch Phys Med Rehabil 2006;87:1396–402. [37] Australian Bureau of Statistics. 2033.0.55.001 - Census of Population and Housing: Socio-Economic Indexes for Areas (SEIFA), Australia, 2011. Available from: http://www.abs.gov.au/AUSSTATS/[email protected]/ DetailsPage/2033.0.55.0012011. [Accessed September 18, 2011]. [38] Gangemi EN, Gregori D, Berchialla P, et al. Epidemiology and risk factors for pathologic scarring after burn wounds. Arch Facial Plast Surg 2008;10:93–102. [39] Queensland Government. Nursing Wage Rates: as at 1 April 2013. Available from: http://www.health.qld.gov.au/hrpolicies/wage_rates/ nursing.asp. [Accessed October 10, 2013]. [40] Queensland Government. District Health Services: Senior Medical Officers’ and Resident Medical Officers’ Award - State 2012. Available from: http://www.health.qld.gov.au/hrpolicies/wage_rates/medical. asp. [Accessed October 10, 2013]. [41] Royal Automobile Club of Queensland. Vehicle Running Costs 2013. Available from: http://www.racq.com.au/motoring/cars/car_economy/ vehicle_running_costs_2013. [Accessed October 10, 2013]. [42] Australian Bureau of Statistics. 6523.0 - Household Income and Income Distribution, Australia, 2011-12. Available from: http://www.abs.gov.au/ AUSSTATS/[email protected]/DetailsPage/6523.02011-12. [Accessed September 23, 2013]. [43] Queensland Government. Queensland Health Wage Rates. Australia Health Practitioners’ Stream. Queensland Government, 2013. Available from: http://www.health.qld.gov.au/hrpolicies/wage_rates/. [Accessed October 10, 2013]. [44] Briggs AH, Wonderling DE, Mooney CZ. Pulling cost-effectiveness analysis up by its bootstraps: a non-parametric approach to confidence interval estimation. Health Econ 1997;6:327–40. [45] StataCorp. Stata Statistical Software: Release 12. College Station, TX: StataCorp LP, 2011. [46] Brown NJ, Kimble RM, Gramotnev G, et al. Predictors of reepithelialization in pediatric burn. Burns 2014;40:751–8. [47] Klein MB, Hollingworth W, Rivara FP, et al. Hospital costs associated with pediatric burn injury. J Burn Care Res 2008;29:632–7. [48] Kai-Yang L, Shi-Hui Z, Hong-Tai T, et al. The direct hospitalisation costs of paediatric scalds: 2-year results of a prospective case series. Burns 2009;35:738–45. [49] Carey K, Kazis LE, Lee AF, et al. Measuring the cost of care for children with acute burn injury. J Trauma Acute Care Surg 2012;73: S229–33. [50] Mirastschijski U, Sander J-T, Weyand B, et al. Rehabilitation of burn patients: an underestimated socio-economic burden. Burns 2013;39:262–8. [51] Hemington-Gorse SJ, Potokar TS, Drew PJ, et al. Burn care costing: the Welsh experience. Burns 2009;35:378–82. [52] Australian Government. Australian Institute of Health and Welfare. Burns and Scalds, October 2006. Available from: http://www.nisu. flinders.edu.au/pubs/reports/2006/injcat92.pdf. [Accessed September 4, 2013]. [53] Gabbe B, Picton N, Loh I, Watterson D. Bi-National Burns Registry Annual Report 1st July 2011 – 30th June 2012. In: Department of Epidemiology and Preventive Medicine, ed. Melbourne, Australia: Monash University, 2012.