If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Corresponding author: Lindley B. Wall, MD, MSc, Washington University Orthopedics BJC Institute of Health, 5th Floor, 660 South Euclid Avenue, St Louis, MO 63110.
The purpose of this study was to compare the cost-effectiveness of surgical release to botulinum toxin injections in the treatment of upper-extremity (UE) cerebral palsy (CP).
Methods
A Markov transition-state model was developed to assess the direct and indirect costs as well as accumulated quality-adjusted life-years associated with surgery (surgery group) and continuous botulinum toxin injections (botulinum group) for the treatment of UE CP in children aged 7 to 12 years. Direct medical costs were obtained from institutional billing departments. The number of parental missed workdays associated with each treatment was estimated and previously published regressions were used to calculate indirect costs associated with missed work. Total costs, cost-effectiveness, and incremental cost-effectiveness ratios were calculated. Incremental cost-effectiveness ratios and willingness to pay thresholds were used to make decisions regarding society’s willingness to pay for the incremental cost of each treatment given the incremental benefit.
Results
The surgery group demonstrated lower direct, indirect, and total costs compared with the botulinum group. Direct costs were $29,250.50 for the surgery group and $50,596.00 for the botulinum group. Indirect costs were $9,467.46 for the surgery group and $44,428.60 for the botulinum group. Total costs were $38,717.96 for the surgery group and $95,024.60 for the botulinum group, a difference of $56,306.64. The incremental cost-effectiveness ratio was –$42,019.88, indicating that surgery is a less costly and more effective treatment and that botulinum injections fall outside the societal willingness to pay threshold. Excluding indirect costs associated with parental missed work during home occupational therapy did not have a significant impact on the model.
Conclusions
Surgery is associated with lower direct, indirect, and total costs, as well as a greater number of accumulated quality-adjusted life-years. Surgery provides a greater benefit at a lower cost, which suggests that botulinum injections should be used sparingly in this population. Treatment with surgery could represent savings of $5.6 to $11.3 billion annually in the United States.
Dr. Graham interviews Dr. Lindley Wall, senior author of the lead article in the May 2021 issue of the Journal of Hand Surgery, "Cost Comparison of Botulinum Toxin Injections Versus Surgical Treatment in Pediatric Patients With Cerebral Palsy: A Markov Model".
Cerebral palsy (CP) is an irreversible, nonprogressive, developmental condition caused by a disturbance in the growth and maturation of the fetal or infant brain. Globally, about 2 to 4 out of every 1,000 live births is affected by CP.
Adverse musculoskeletal adaptations often result from long-term malpositioning, muscle spasticity, and increased baseline muscle tone if left untreated. To improve cosmesis, restore function, promote more normal musculoskeletal maturation, and decrease the risk for adverse musculoskeletal adaptations, multiple treatment options have been developed, including the use of an orthosis or casting and occupational therapy, surgical intervention (tendon lengthening, releases, and/or transfers), and medical management, including botulinum toxin A injections. The efficacy of each of the 3 treatment regimens for UE CP has been studied, but evidence from a randomized controlled trial comparing these treatments in measures of functional use of the UE, quality of life, and satisfaction demonstrated notable advantages for surgical intervention.
Despite the apparent clinical advantages in favor of surgery, it is unclear whether the increased up-front costs of this intervention warrant its use. Therefore, the objective of the study was to determine the direct costs, indirect costs, and accumulated quality-adjusted life years (QALYs) associated with surgical intervention and botulinum toxin injections for UE CP. This assessment allows a comparison of the societal cost-effectiveness of these interventions and can potentially identify an optimal treatment strategy to minimize medical and societal costs while maximizing patient benefit.
Materials and Methods
Model overview
We obtained institutional review board exemption before we initiated this study. A Markov transition-state model was developed to assess the lifetime costs and accumulated QALYs associated with surgery and continuous botulinum toxin injections for the treatment of chronic UE CP. Total costs included an assessment of both direct and indirect societal costs. The model calculated costs and QALYs based on 1-year cycles. As is standard in assessments of cost and QALY, these parameters were discounted at an annual rate of 3%.
Modeled cases were transitioned into either the surgical or botulinum toxin treatment pathway and accumulated the associated direct and indirect costs related to their respective treatment. Patients in the surgery group underwent one surgical intervention during the model period, whereas patients in the botulinum toxin group received botulinum toxin injections 2 times annually for the duration of the model. Although patients often undergo injections every 4 months,
we selected this conservative number to minimize the risk for bias in the model. Cases were also exposed to potential indirect societal costs associated with treatment, as well as QALY values corresponding to the predicted efficacy of their respective treatments. The patient exited the model at age 12 years.
Direct treatment costs
Direct treatment costs were defined as medical costs associated with the surgery group and botulinum group. Surgical costs included the surgeon’s professional fees associated with pronator tendon release, wrist flexor tendon transfer, thenar muscle release, and thumb extensor tendon transfer, as well as an estimated 2.5 hours of anesthesia and a hospital facility cost (Table 1). Surgical costs were obtained through the authors’ institutional billing department using Current Procedural Terminology codes. Anesthesia and hospital facility costs were obtained from the billing department. Costs obtained from the 2 institutions participating in this study were averaged to arrive at the final costs.
Table 1Direct Costs Associated With Surgery, Botulinum Toxin Injections, and PT
Costs displayed here represent individual costs, and not total costs associated with treatment during the model. Botulinum injection dosing is based on patient weight. Typically, a dose of 0.5 to 1.0 U/kg is used in the adductor pollicis and 1.0 to 2.0 U/kg is used in both the pronator teres and flexor carpi ulnaris.7 However, typically there is a maximum dose of 50 U/injection site to prevent potential systemic toxicity.7 Average weight as a function of age was averaged for males and females based on data from the World Health Organization and Centers for Disease Control and Prevention,10 and this weight was used to calculate the dose and injection cost for children from ages 7 to 12 years.
Variable
Cost
Pretreatment evaluation
$293.00
First posttreatment visit
$492.00
First posttreatment orthosis
$68.00
Posttreatment maintenance (per visit)
$400.00
Surgery
Pronator tendon release
$2,243.00
Wrist flexor tendon transfer
$3,184.50
Thenar release
$3,186.00
Thumb extensor tendon transfer
$3,784.00
Anesthesia 2.5 h
$1,040.00
Hospital facility
$11,760.00
Botulinum toxin
Physician cost (per injection)
$415.00
Age 7 (22.7 kg)
$1,061.72
Age 8 (25.7 kg)
$1,204.69
Age 9 (28.4 kg)
$1,328.91
Age 10 (32.0 kg)
$1,497.66
Age 11 (36.3 kg)
$1,644.53
∗ Costs displayed here represent individual costs, and not total costs associated with treatment during the model. Botulinum injection dosing is based on patient weight. Typically, a dose of 0.5 to 1.0 U/kg is used in the adductor pollicis and 1.0 to 2.0 U/kg is used in both the pronator teres and flexor carpi ulnaris.
Average weight as a function of age was averaged for males and females based on data from the World Health Organization and Centers for Disease Control and Prevention,
and specialists at the institutions involved in the current study were consulted to determine current practice protocols for the treatment of UE CP. Costs associated with the botulinum group included the cost of botulinum toxin injections in the adductor pollicis, pronator teres, and flexor carpi ulnaris, as well as a physician charge. The cost of botulinum injections, $12.50/unit, was obtained from our institution’s billing department. Table 1 lists the methodology for botulinum injection dosing and costs.
Direct rehabilitation costs
For both the surgery and botulinum groups, outpatient rehabilitation costs included the cost of a pretreatment evaluation, a first posttreatment visit, a first posttreatment orthosis evaluation or fitting, and 8 posttreatment maintenance visits. Patients in the surgery group completed this cycle of occupational therapy (OT), often by certified UE physical therapists, once during the postoperative year. Similarly, patients in the botulinum group underwent this entire treatment cycle during the first treatment. Because patients underwent ongoing OT with each injection, they accrued the direct costs of posttreatment maintenance OT with each subsequent botulinum injection but did not repeatedly incur the costs of pretreatment evaluation, a first posttreatment visit, or a first posttreatment orthosis evaluation or fitting.
Indirect costs
An assessment of indirect cost is crucial to understanding the societal impact of various treatments. The parents of the pediatric patients in our model were likely to be relatively young, which placed them in high-earning income brackets. As a result, indirect costs associated with the high-intensity care of a child with CP were likely to be large. Prior studies
to estimate indirect costs associated with missed work. A similar methodology was used to estimate indirect costs in this study, as detailed in Figure 1. In the surgery group, we estimated that a parent would miss a full work week (40 hours) during the child’s week of surgery. In the botulinum group, we assumed that a parent would miss 2 hours of work for each injection appointment (1 hour of visiting and 1 hour for travel). In both groups, we estimated that a parent would miss 2 hours of work per outpatient OT visit (1 hour of visiting and 1 hour for travel).
Figure 1Calculating the cost of missed workdays. Using previously described methodology,
a baseline reference income and a baseline reference number of missed workdays were established. Because the predicted income varies as a function of age, this required the selection of a baseline parental age for our model. We assumed that parents of children between ages 7 and 12 years would be between ages 30 and 50 years, and averaged the income from individuals in this age range to form a baseline estimate of income. Assuming a 40-hour work week and a 240-day work year, consistent with prior models,
we then used this reference to calculate the earnings associated with each workday, and hence the cost of a missed workday. After estimating the number of missed workdays associated with treatment with surgery and botulinum injections, the indirect costs associated with this missed work could be estimated using the equation shown in Figure 1. Reference wages were updated to 2019 US dollars using the Consumer Price Index and then used to estimate parental earnings and losses.
we assumed that treatment with surgery and botulinum injections were also accompanied by home OT conducted by the parent or caregiver. Because home OT in this model is conducted by the parent or caregiver, not a visiting therapist, there is a strong likelihood that the parent must miss work or other obligations to participate. Therefore, we assumed that each hour of home OT would be associated with 1 hour of missed parental work. Home OT consisted of 1 hour of therapy, 5 days/wk. We assumed that patients in the surgery group would undergo 5 days/wk of home OT for the full year after surgery, and that patients in the botulinum group would undergo 5 days/wk of home OT for the duration of the model.
Because indirect costs associated with home OT have a large impact on the overall costs in our model, and because these costs are likely to vary greatly based on patient and family characteristics, a secondary analysis was performed excluding these costs, to avoid potential bias.
Quality-adjusted life-years
Quality-adjusted life-years provide a measure to quantify the clinical impact of the treatments being assessed and provide a tool to measure the quantity and quality of years lived by the members of each study group.
Currently, no available literature directly compares QALY measures between surgery and botulinum injections. However, data exist comparing these 2 treatments using Pediatric Quality of Life Inverntory TM (PedsQL)-parent scores.
Therefore, existing PedsQL-parent scores were mapped into GMFCS scores, which were then converted to QALY measures, as demonstrated in Figure 2.
Figure 2Mapping PedsQL-parent movement scores into assessments of QALYs. Because the PedsQL scores for botulinum injections mapped imperfectly into the GMFCS score and fell between the ranges for GMFCS scores of II and III, we averaged the QALY values associated with these 2 GMFCS scores to arrive at the final QALY value for botulinum injections. Although Peds-QL scores were closer to the values associated with a GMFCS score of III, we elected to average the QALY values associated with GMFCS scores of II and III to avoid biasing this assessment. For surgery, the posttreatment Peds-QL parent movement score of 78.9 mapped into a final QALY value of 0.888. For botulinum injections, the Peds-QL score of 56.1 mapped into a final QALY value of 0.620. Although the QALY score was lower in the botulinum group, we assumed this to be a fair estimate because prior Cochrane reviews demonstrated no improvement in quality of life from baseline in patients treated with botulinum.
Cost-effectiveness was calculated as the total 5-year costs divided by the number of accumulated QALYs. Incremental cost-effectiveness ratios (ICERs) were also calculated to convert the model into a binary decision-making tool capable of determining society’s willingness to pay (WTP) for treatment. We compared ICER values with the accepted WTP threshold of $50,000
to determine whether these treatments represented reasonable alternatives from a societal perspective.
Sensitivity analysis
A one-way sensitivity analysis was performed to determine the stability of our model to variations in direct and indirect costs, as well as QALYs. Values were varied by ±25% and ±75% to detect alterations in ICER-based treatment decisions. This allowed us to detect changes that had an impact on the final ICER-based decision output of the model, and was in accordance with previous studies assessing ICER sensitivity.
When indirect costs associated with home OT were included in our primary analysis, the surgery group demonstrated a lower direct cost, indirect cost, and total cost compared with the botulinum group. Direct costs over the 5-year model for the surgery and injection groups are demonstrated in Tables 2 and 3, respectively. Direct costs in the surgery group were $29,250.50/patient compared with $50,596.00 in the botulinum group. When direct costs were broken down into treatment and OT costs, our analysis indicated that OT costs were the major driver of these large differences in direct cost. The direct cost of treatment (surgery and associated fees) in the surgery group was $25,197.50 compared with $13,475.00 (injections and associated fees) in the botulinum group. However, whereas the cost of outpatient OT was $4,053.00 in the surgery group, it was $32,921.00 in the botulinum group.
The cost of posttreatment orthoses reflects the cost of 2 orthosis visits. The cost of posttreatment maintenance reflects the cost of 80 total maintenance visits (2 injections/y for 5 years, each followed by 8 maintenance PT visits). Physician costs reflect the total costs associated with 10 botulinum toxin injections. Injection costs reflect the price of 2 annual injections in the adductor pollicis 0.75 U/kg, pronator teres 1.5 U/kg, and flexor carpi ulnaris 1.5 U/kg. The base-case patient in our model underwent 2 rounds/y of injections.
PT
Cost
Pretreatment evaluation
$293.00
First posttreatment visit
$492.00
Posttreatment orthoses
$136.00
Posttreatment maintenance
$32,000.00
PT total
$32,921.00
Botulinum toxin
Physician cost
$4,150.00
Age 7 (22.7 kg)
$2,123.44
Age 8 (25.7 kg)
$2,409.38
Age 9 (28.4 kg)
$2,657.81
Age 10 (32.0 kg)
$2,995.31
Age 11 (36.3 kg)
$3,289.06
Injection total
$17,625.00
Total direct costs
$50,546.00
∗ The cost of posttreatment orthoses reflects the cost of 2 orthosis visits. The cost of posttreatment maintenance reflects the cost of 80 total maintenance visits (2 injections/y for 5 years, each followed by 8 maintenance PT visits). Physician costs reflect the total costs associated with 10 botulinum toxin injections. Injection costs reflect the price of 2 annual injections in the adductor pollicis 0.75 U/kg, pronator teres 1.5 U/kg, and flexor carpi ulnaris 1.5 U/kg. The base-case patient in our model underwent 2 rounds/y of injections.
Similar relationships were seen with respect to indirect costs. Indirect costs in the surgery group were $9,467.46 versus $44,428.60 in the botulinum group. These differences were largely driven by lost wages associated with outpatient and home physical therapy (PT), as summarized in Table 4. Total costs, the sum of indirect and direct costs, were $38,717.96 in the surgery group and $95,024.60 in the botulinum group, a difference of $56,306.64 (Table 5).
A negative difference in cost or cost-effectiveness represents a cost advantage for surgery. A positive difference in QALY represents an advantage for surgery. The first value shows the ICER analysis including indirect costs of home PT, whereas the value in parentheses shows the ICER analysis excluding these costs.
Cost Variable
Surgery
Botulinum
Difference
Direct cost
$29,250.50
$50,596.00
–$21,345.50
Indirect cost
$9,467.46 ($1,525.94)
$44,428.60 ($5,242.81)
–$34,961.14 (–$3,716.88)
Total cost
$ 38,717.96 ($30,776.44)
$95,024.60 ($55,788.81)
–$56,306.64 ($30,776.44)
QALY
4.44
3.10
+1.34
Cost-effectiveness
$8,720.26 ($6,931.63)
$30,653.10 ($17,996.39)
–$21,932.84 (–$11,064.76)
∗ A negative difference in cost or cost-effectiveness represents a cost advantage for surgery. A positive difference in QALY represents an advantage for surgery. The first value shows the ICER analysis including indirect costs of home PT, whereas the value in parentheses shows the ICER analysis excluding these costs.
When the indirect costs associated with home OT were excluded in the secondary analysis, the indirect costs associated with surgery and botulinum injections decreased to $2,670.39 and $9,174.43, respectively. Total costs decreased to $31,920.89 and $59,720.93, respectively, still in favor of surgery.
Cost-effectiveness and ICER
In the primary analysis, cost-effectiveness was maximized by performing surgery. The cost-effectiveness of treatment from age 7 to 12 years was $8,720.26 in the surgery group and $30,653.10 in the botulinum group. The difference in cost-effectiveness between groups was $21,932.84 in favor of surgery. Moreover, ICER was –$42,019, indicating that surgery provided a greater benefit at a lower cost.
Removing indirect costs associated with home OT in the secondary analysis improved the cost-effectiveness of both surgery and botulinum injections but did not have an appreciable impact on the relationship between the 2. The difference in cost-effectiveness between groups remained high at $23,075.42 in favor of surgery. In addition, ICER indicated persistent dominance for surgery, at –$20,746.29 (Table 5).
Sensitivity analysis
The results of the one-way sensitivity analysis demonstrated that our model was highly robust to changes of ±25% and ±75% for nearly all parameters assessed (Table 6). According to our ICER-based decision model, the only change that altered the preferred strategy in the primary analysis was a 75% decrease in the QALY associated with surgery. In the secondary analysis, which excluded the indirect costs of home PT, the preferred strategy was sensitive to a 75% decrease in botulinum injection frequency, surgery QALY, and outpatient OT frequency, as well as a 75% increase in botulinum injection QALY.
The first value of each column shows the ICER analysis including indirect costs of home PT, whereas the value in parentheses shows the ICER analysis excluding these costs. The baseline ICER value in our model was –$41,982.56. The ICER-based decisions were sensitive only to a 75% decrease in the QALY value (0.888 to 0.222) associated with surgery and a 75% increase in the QALY value (0.620 to a maximum of 1.0) associated with botulinum injections when the indirect costs of home PT were included in our primary analysis. The ICER changed from –$41,982.56 to $28,269.66 when the surgery QALY was decreased by 75%, indicating that both surgery and botulinum injections would be acceptable alternatives (below a WTP of $50,000), with injections offering a greater benefit at a higher cost. The ICER changed from –$41,982.56 to $100,458.28 when the botulinum injection QALY was increased by 75%, indicating that botulinum toxin was no longer dominated by surgery, but that it remained an unacceptable alternative from a societal perspective because it exceeded the WTP threshold. When the indirect costs of home PT were excluded from the model, it was sensitive to 75% decreases in botulinum injection frequency (2 to 0.5 injections/y), surgery QALY (0.888 to 0.222), and outpatient PT frequency (8 to 2 visits/y), as well as 75% increases in botulinum injection QALY (0.65 to 1.0). Decreases in botulinum injection frequency were also associated with a decrease in outpatient PT frequency, which further decreased costs. Each of these changes adjusted the ICER value from a negative value, indicating dominance for surgery, to a positive value less than $50,000, indicating that both surgery and botulinum injections would be reasonable alternatives according to the WTP threshold.
∗ The first value of each column shows the ICER analysis including indirect costs of home PT, whereas the value in parentheses shows the ICER analysis excluding these costs. The baseline ICER value in our model was –$41,982.56. The ICER-based decisions were sensitive only to a 75% decrease in the QALY value (0.888 to 0.222) associated with surgery and a 75% increase in the QALY value (0.620 to a maximum of 1.0) associated with botulinum injections when the indirect costs of home PT were included in our primary analysis. The ICER changed from –$41,982.56 to $28,269.66 when the surgery QALY was decreased by 75%, indicating that both surgery and botulinum injections would be acceptable alternatives (below a WTP of $50,000), with injections offering a greater benefit at a higher cost. The ICER changed from –$41,982.56 to $100,458.28 when the botulinum injection QALY was increased by 75%, indicating that botulinum toxin was no longer dominated by surgery, but that it remained an unacceptable alternative from a societal perspective because it exceeded the WTP threshold. When the indirect costs of home PT were excluded from the model, it was sensitive to 75% decreases in botulinum injection frequency (2 to 0.5 injections/y), surgery QALY (0.888 to 0.222), and outpatient PT frequency (8 to 2 visits/y), as well as 75% increases in botulinum injection QALY (0.65 to 1.0). Decreases in botulinum injection frequency were also associated with a decrease in outpatient PT frequency, which further decreased costs. Each of these changes adjusted the ICER value from a negative value, indicating dominance for surgery, to a positive value less than $50,000, indicating that both surgery and botulinum injections would be reasonable alternatives according to the WTP threshold.
Value of botulinum toxin injections preceding a comprehensive rehabilitation period for children with spastic cerebral palsy: a cost-effectiveness study.
to our knowledge, there are no data comparing the cost-effectiveness of surgical treatment and botulinum toxin injections for UE CP in the pediatric population. In light of data demonstrating enhanced efficacy for surgery compared with botulinum injections,
it is important to understand whether the clinical advantages of surgery are sufficient to outweigh its increased initial direct costs. The goal of the current study was to perform an economic assessment of the cost-effectiveness of surgery and botulinum injections in the pediatric population during the first 5 years after initiating treatment for UE CP.
The results of our analysis demonstrate strong economic evidence in favor of surgery compared with continuous botulinum injections. Surgery was associated with lower direct, indirect, and total costs, as well as a greater number of accumulated QALYs. Although the direct cost of surgery alone was greater than the direct cost of botulinum injections alone, direct costs were pushed heavily in favor of surgery when the costs of ongoing therapy were incorporated into the model. Given the increased need for long-term facility-based and in-home therapy associated with ongoing botulinum injection treatment, botulinum injections were also associated with far higher indirect costs. As a result of this large difference in total costs, as well as advantages in accumulated QALYs for surgery, our analysis demonstrated a large difference in cost-effectiveness in favor of surgery. Most important, ICER values were negative, indicating that surgery provided a greater benefit at a lower cost. From an economic perspective, this indicates that botulinum injections are not a reasonable alternative to surgery in this population. Surprisingly, these relationships remained even when the indirect costs associated with home OT were eliminated from the model in the secondary analysis. This is an important point because it indicates that one of the most costly, variable, and difficult-to-estimate inputs in the model, which has the potential to introduce a large amount of variability into the model inputs, does not have the ability to influence the eventual model output.
As demonstrated by our sensitivity analysis, these findings appear to be robust to moderate alterations in model inputs. Although the model in both the primary and secondary analyses was sensitive to dramatic decreases and increases in surgery and botulinum injection QALY, respectively, these largely represent unrealistic alterations to the model, because children with UE CP are unlikely to experience such extreme values of QALY. However, the model in the secondary analysis suggests that in the absence of indirect costs associated with home PT, decreasing the frequency of botulinum injections and outpatient OT may bring botulinum injections within the WTP threshold.
In contrast to the high volume of studies comparing botulinum injections with traditional therapy for the treatment of CP, there are minimal data comparing botulinum injections with surgical treatment. To our knowledge, the only available randomized, controlled trial comparing these interventions demonstrated clear advantages for surgery in terms of the functional use of the UE, quality of life, and satisfaction.
In conjunction with the economic findings of the current study, this is strong evidence for the preferential use of surgical management of UE CP in the pediatric population. Combined with the clear advantages in terms of convenience associated with a one-time surgical treatment, as well as the potential to minimize suboptimal outcomes associated with noncompliance, these cost and efficacy advantages provide an even stronger case for the preferential use of surgical treatment.
The results of the current study have important national implications for the health care system. With an incidence of 2 to 4 out of every 1,000 live births,1 roughly 800,000 to 1,600,000 children are born with CP each year in the United States. If 1 out of 8 of these children are born with UE CP, 100,000 to 200,000 patients each year fit the criteria for the model presented here. With an average savings of $56,306.64/patient undergoing a 5-year treatment cycle with surgery compared with botulinum injections, this represents a potential yearly savings of $5.6 to $11.3 billion.
By assessing costs from a direct payer perspective as well as an indirect societal perspective, this model provides an effective assessment of cost-effectiveness for all stakeholders involved in the decision-making process. The resilience of the model in our sensitivity analysis indicates that these results are likely to remain externally valid in diverse patient populations and hospital systems with variable outcomes and costs. However, there are limitations of this model that are worth considering. First, this study required the estimation of QALY values by mapping Peds-QL scores into QALY scores. Although this may not perfectly capture the QALY outcomes associated with these treatments, in this assumption and others, we erred on the side of selecting more conservative measures for botulinum injection treatment to minimize bias. In addition, owing to the limited availability of data assessing QALY in patients treated for UE CP, GMFCS scores, which are traditionally used to assess lower-extremity CP, were used to estimate QALYs. Second, the lack of data describing the indirect costs associated with these treatment modalities necessitated our estimating these costs using available regression data. Therefore, the accuracy of these assessments is only as reliable as the accuracy of these regressions. Third, this study included indirect costs based on presumed missed work resulting from parental performance of home OT sessions. Although we believe these costs are difficult to estimate reliably, the fact that the exclusion of these costs did not have a considerable impact on the preferred strategy suggested by our model indicates that regardless of the accuracy of our indirect cost measurements, surgical treatment remains the overwhelmingly more cost-effective option. Fourth, given the lack of standardization in injection protocols for UE CP, we did not include the costs of highly variable treatment practices such as patient sedation during injections. This may have led to underestimations of the direct costs of injections, but it likely provides a more accurate estimate of conservative injection protocols. Fifth, patients with CP tend to have a lower socioeconomic status. Therefore, it is possible that our model overestimates parental income and thus overestimates indirect costs associated with missed work. As mentioned, however, because our sensitivity analysis demonstrated that the model was insensitive to changes in parental income, this is unlikely to be a major driver of change. Finally, owing to the paucity of literature describing complications and revisions associated with surgery for UE CP, it was not possible to estimate the associated costs of these factors in our model. This is an important shortcoming of this model. Importantly, however, the insensitivity of the model to a near-doubling of the cost of surgery indicates that even if a large number of patients were to require repeat surgery, it would not have an impact on the preferred strategy indicated by the model. Future studies should focus on investigating the sensitive parameters of this model (botulinum toxin injection frequency and OT frequency) to determine whether they represent actionable strategies for further optimizing the cost-effectiveness of treatment for UE CP.
Value of botulinum toxin injections preceding a comprehensive rehabilitation period for children with spastic cerebral palsy: a cost-effectiveness study.
00045-9&id=gr1.jpg){kind=link}
00045-9&id=gr2.jpg){kind=link}