World J Surg DOI 10.1007/s00268-014-2530-2
Cost-Effectiveness of the Ponseti Method for Treatment of Clubfoot in Pakistan Hamidah Hussain • Aziza Moiz Burfat • Lubna Samad • Fayez Jawed • Muhammad Amin Chinoy Mansoor Ali Khan
•
Ó Socie´te´ Internationale de Chirurgie 2014
Abstract Background Clubfoot is disabling, with an incidence of 0.9/1,000 live births to 7/1,000 live births. It affects mobility, productivity, and quality of life. Patients are treated surgically or non-surgically using the Ponseti method. We estimated the cost per patient treated with both methods and the cost-effectiveness of these methods in Pakistan. Methods Parents of patients treated, either surgically or with the Ponseti method, at the Indus Hospital’s free program for clubfoot were interviewed between February and May 2012. We measured the direct and indirect household expenditures for pre-diagnosis, incomplete treatment, and current treatment until the first brace for Ponseti method and the first corrective surgery for surgically treated patients. Hospital expenditure was measured by existing accounts. Results Average per-patient cost was $349 for the Ponseti method and $810 for patients treated surgically. Of these, the Indus hospital costs were $170 the for Ponseti method and $452 for surgically treated patients. The direct household expenditure was $154 and $314 for the Ponseti and surgical methods, respectively. The majority of the H. Hussain (&) Interactive Research and Development (IRD), Suite 508, Shahrah-e-Faisal, Ibrahim Trade Towers, Karachi, Pakistan e-mail: [email protected] A. M. Burfat L. Samad F. Jawed Indus Hospital Research Center, 4th Floor, Indus Hospital, Korangi Crossing, Karachi, Pakistan e-mail: [email protected] M. A. Chinoy M. A. Khan Department of Orthopaedics & Trauma, Indus Hospital, Korangi Crossing, Karachi, Pakistan
costs were incurred pre-diagnosis and after inadequate treatment, with the largest proportion spent on transportation, material, and fee for service. The Ponseti method is shown to be the dominant method of treatment, with an incremental cost-effectiveness ratio of $1,225. Conclusions The Ponseti method is clearly the treatment of choice in resource-constrained settings like Pakistan. Household costs for clubfoot treatment are substantial, even in programs offering free diagnostics and treatments and may be a barrier to service utilization for the poorest patients.
Introduction Clubfoot or congenital talipes equinovarus (CTEV) is a disabling condition with an incidence of 0.9/1,000 live births to 7/1,000 live births in different populations [1–3]. The deformity affects an individual’s mobility, threatens that person’s potential productivity, and reduces his or her standard of living [4]. In Pakistan, surgical intervention has been the preferred treatment for clubfoot for many years for children of all age groups. Owing to limited financial resources, surgical treatment is not always affordable for the patient and this leads to a large burden of untreated or neglected clubfoot. In October 2011, the Ponseti method for treatment of clubfoot—now the standard of care for treatment for uncomplicated cases in infants [5]—was introduced at the Indus Hospital in Karachi. This method does not require extensive surgical intervention or anesthesia. Instead, it uses a series of gentle manipulations to correct the deformity with serial casting and bracing to maintain the correction [6, 7]. Older children continued to be treated surgically.
123
World J Surg
Studies have clearly demonstrated the clinical effectiveness of the Ponseti method with an initial correction rate of 90 % in idiopathic feet and a faster decrease in the Pirani score [8–11]. In comparison with the Ponseti method, treatment by clubfoot surgery is associated with poorer long-term outcomes. Complications such as stiffness and weakness of the foot, as well as pain, have been reported, ultimately leading to secondary procedures [5, 12–17]. However, limited data are available on costeffectiveness. A study from New Zealand concluded that the Ponseti method was less resource intensive for the health system with lower cost and morbidity because the patient spent less time in the hospital [18]. The Ponseti method of treatment is appealing in a resource-constrained setting like Pakistan, where most health care expenses are paid out of pocket and the cost of surgical treatment is prohibitive for most families. The objective of the present study was to estimate the cost per patient treated with both the surgical approach and the Ponseti method from a societal perspective, and to calculate the cost-effectiveness of these methods in Pakistan.
Materials and methods This cross-sectional study was conducted at the Indus Hospital in Karachi, which serves a catchment population of 2.4 million, as well as referrals from other parts of Pakistan. The Indus Hospital is a tertiary care facility where all inpatient services are free of charge to patients. The Pehla Qadam (First Step) Clinic established at the Indus Hospital in October 2011 is a collaborative effort between the Department of Orthopaedics and the Indus Hospital Research Center. Initially, enrolment was restricted to children of age 1 year and below at the time of presentation. Children older than 1 year were treated with surgical treatment, which was accepted as the standard practice at that time. In addition to clinical management, the program team worked to create awareness for early identification and referral in the catchment population, as well as providing close patient follow-up and data management. Two groups of patients were enrolled: group A: 56 consecutive patients were enrolled from the Pehla Qadam Clinic and treated using the Ponseti method from October 2011 till May 2012; group B: 48 patients who underwent surgical correction at the Indus Hospital from July 2007 till May 2012. Orthopedic surgeons and residents provided medical care to both groups. A trained two-member team conducted all interviews at the patient’s home using a standardized cost-estimate tool from February 2012 till May 2012 after informed consent was obtained from parents. Costs were collected from a societal perspective. Data were collected on demographics; income; and direct out-
123
of-pocket expenditures for consultations, diagnostic tests, medicines, hospitalization, meals, and transport. For the Ponseti method, costs were collected from the onset of symptoms, serial casting, and tenotomy until the first brace was applied; for the patients treated with the surgical method, costs were collected from the onset of symptoms, through casting and initial corrective surgery at Indus Hospital, until discharge. The questionnaire was divided into three sections. The first section dealt with costs incurred by families in seeking medical care before the start of the treatment, such as consultation, treatment, travel, and meals (the pre-diagnostic period). The second section covered costs for the period during which patients were receiving treatment but the care was left incomplete because of lack of family resources or for any other reason (incomplete treatment). The last section focused on costs for the current treatment, either serial casting or surgery at the Indus Hospital. Indirect costs from lost earnings were estimated for one family member accompanying the child for outpatient visits, including travel and waiting time; hospital days were estimated at 8 h of work loss per day based on the minimum wage in Pakistan. Health system costs for the current treatment were calculated by reviewing existing accounts for all activities related to treatment. Indus Hospital maintains all medical and treatment expenditure data electronically, with detailed documentation of resources utilized on a case-by-case basis. Resources were quantified and included personnel salaries, capital costs, utilities, overhead, maintenance, and supplies. In addition to costs for clinical care, costs for the awareness campaign, close patient follow-up, and reimbursement for transport were also included as part of the total expenditure. Disability weights, and therefore disability-adjusted life years (DALYs) for clubfoot are not available and were not estimated by the most recent Global Burden of Disease (GBD) study. We used a conservative estimate of successful treatment outcome from the literature (70 % treatment success for surgery and 90 % treatment success for Ponseti) to calculate the incremental cost-effective ratio (ICER) [19, 20]. Costs were collected in Pakistan Rupees (PKR) and converted to US dollars using the average conversion rate for the fiscal year 2011–2012 (1 USD = 90 PKR). The data were double-entered in Microsoft Access (Microsoft Corporation Seattle, WA) and analyzed with Stata (Stata Corp LP. College Station, TX, USA). The review board of Interactive Research and Development (IRD) approved the study.
Results Of the 104 patients enrolled, 56 patients were treated with the Ponseti method (group A) and 48 patients were treated
World J Surg Table 1 Demographic and economic characteristics of patients Group A (Ponseti method) N = 56
Group B (surgery) N = 48
Male
45 (80 %)
28 (58 %)
Female
11 (20 %)
20 (42 %)
Mean age Unilateral
4 months (SD) 37 (66 %)
4 years (SD) 21 (44 %)
Bilateral
19 (34 %)
27 (56 %)
Average number of people in household
4.8
4.8
Average household income/month
PKR 10, 952 USD 122
PKR 10, 985 USD 122
PKR Pakistan rupees, USD U.S. dollars
surgically (group B). The majority of patients were male, with a mean age of 4 months versus 4 years in groups A and B, respectively. Both groups had similar socioeconomic characteristics (Table 1). The direct household expenditure per patient during all treatment periods was $154 for group A as compared to $314 for patients in group B. Of the direct expenditure, over 50 % of costs were paid for treatment that ultimately was not successful. During each treatment phase, travel cost was the largest direct expenditure, ranging from 22 to 60 % (Table 2). Besides the cost to patients, Indus Hospital incurred an additional $170 and $452 for group A and B patients, respectively. For patients treated with the Ponseti protocol, an average of six casts were required, followed by tenotomy in the outpatient setting before the brace was applied. Of the $170 spent by the hospital on the Ponseti treatment, $124 was for clinical care and $45 was for program costs. The program costs included partial reimbursement to patients toward transportation costs (Table 3). For surgical patients, generally one admission was required with an average hospital stay of 5 days. Patients treated surgically had to purchase their own locally made brace costing between $10 and $20, as opposed to patients treated with the Ponseti method, where standardized braces were provided to the patients free of cost as part of the treatment plan (Table 4). The costs were twice as high per patient treated surgically compared to those treated with the Ponseti method for both the patient and the hospital (Table 1). The Ponseti method is shown to be a dominant method of treatment with the ICER of $1,225 (Table 5)
Table 2 Direct and indirect patient costs during all phases of treatment Expense categories
Prediagnostic period Mean number of outpatient visits Mean number of hours for all visits Direct costs Consultation fees Casting Laboratory/radiology/ medication Transport Food Additional costs Indirect cost from lost income Subtotal cost Incomplete treatment Mean number of outpatient visits/patient Mean number of hours for all visits Direct costs Consultation fees Casting Day care procedure/brace Laboratory/radiology/ medication Transport Food Indirect cost from lost income Subtotal cost Current treatment Mean number of outpatient visits Mean number of admissions Mean number of hours for all visits Direct costs Consultation fees Laboratory/radiology/ medication Cast and brace post surgery Transport Food Additional costs Indirect cost from lost income Subtotal cost
Ponseti mean cost per patient (USD)
Surgical mean cost per patient (USD)
3
5
14
25
7 1 6
5 10 6
17 4 0 5
51 10 2 10
39
94
7
13
18
38
10 35 11 2
10 45 29 17
21 6 7
45 17 14
92
176
7
6
0
1
35
70
0 0
0 3
0
17
30 3 1 13
25 13 10 26
48
94
Discussion The results of the present investigation show that the average per patient cost was $349 for the Ponseti method
and $810 for patients treated surgically. The significant difference in the cost for the two methods is primarily based on no hospitalization for the Ponseti method. The
123
World J Surg Table 3 Costs incurred by Indus Hospital per patient treated with the Ponseti method Expense categories
USD
Clinical costs Mean number of outpatient clinic visits (8)
35
Casts
16
Tenotomy
7
Follow-up outpatient clinics
9
Locally made brace Pediatric clinic referral for % of patient Subtotal clinical cost/patient
56 2 124
Program cost Program personal salary Transport money provided to patients
6 10
Communication campaign
10
Program operational cost
19
Subtotal program cost/patient Total cost
45 170
Table 4 Cost incurred by Indus Hospital per patient treated surgically Expense categories
USD
Outpatient costs Mean number of outpatient clinic visits (2)
7
Preoperative clinic and tests
10
Post-surgery follow-up
11
Subtotal outpatient cost/patient
28
Inpatient cost Bed charge per admission Laboratory and pharmacy Surgeon fee
85 18 157
Anesthesia
63
Operating room charges
63
Operating room consumables
39
Subtotal inpatient cost/patient
424
Total cost
452
hospital costs incurred in our study are lower than those reported by other studies [18, 21, 22]. Possible reasons could be that the Indus Hospital is a not-for-profit entity and the costs detailed above are the actual expenditures incurred by the hospital; tenotomies were conducted in an outpatient clinic setting, bringing down the procedurerelated costs significantly; and we did not estimate any costs associated with complications, adherence to brace or long-term follow-up of patients in either group. Household costs for clubfoot are substantial, even when programs such as Pehla Qadam offer free diagnostics and treatment, and these may be a barrier to service utilization
123
for the poorest patients. Over 50 % of all direct expenditures borne by patients were for treatment that did not result in correction of the defect. Large proportions of these costs were for travel, fee for service, and materials. As these costs exceeded the capacity of the household, treatment was left incomplete, thus not achieving optimum results. Other countries have documented similar barriers to treatment [23, 24]. In the absence of any social health insurance system in Pakistan, awareness campaigns and education for early identification and treatment of clubfoot will save resources for both families and the health system. We estimated the rate of treatment success at 70 % with surgery and 90 % with the Ponseti method. Assuming an equal treatment success rate of 90 % for both programs, the Ponseti method still cost less. Patients treated with the Ponseti method spent almost half for incomplete treatment. Assuming the cost of inadequate treatment is the same for both groups, the cost per patient for the Ponseti group increases to $433. Even then, the Ponseti method is less expensive than the surgical option. The high cost of treatment of clubfoot is many times over the estimated $22 per capita expenditure on health in Pakistan. Based on the crude birth rate, the birth cohort for Pakistan is an estimated 5 million children. The estimated incidence of clubfoot in Pakistan is 1.5/1,000 children. This means that there are approximately 7,500 children born with clubfoot in Pakistan every year. If all children with clubfoot are treated with the Ponseti method, the cost will be $2,617,500 as compared to $6,075,000 if children are treated surgically—a considerable savings. Our study clearly shows that in a setting like Pakistan it is cost effective to treat patients with the Ponseti method rather then using surgical intervention. In addition, long-term follow-up studies have shown better functional results, a lower relapse rate, and less morbidity over time, including pain, osteoarthritis, and flatfoot deformity for children treated with the Ponseti method [25]. Management of these postoperative complications will place a high financial burden on the health system. In addition, there will be a comparable burden on society with loss of productivity if these 7,500 children with clubfoot are treated surgically and have difficulty at work as a result of postoperative difficulties after correction of clubfoot. Although cost is a very important factor, there are many other elements that need to be taken into consideration for policy development of clubfoot care in Pakistan. The success of clubfoot treatment is associated with early identification and referral to an appropriate health care facility [26]. However, clubfoot is widely misunderstood in Pakistani populations with a lack of awareness about the disorder and the treatment method [27]. A strong community mobilization arm can therefore be instrumental in
World J Surg Table 5 Cost saved per successfully treated patient Alternative strategy
Cost, USD
Incremental cost, USD
Effect (% of patients treated successfully)
Incremental effect (%)
ICER, USD
Surgical
39,163
–
70
–
–
Ponseti
19,556
19,607
90
20
1,225
ICER incremental cost-effectiveness ratio
the success of a clubfoot care program. The prevalence of clubfoot is not high enough to justify development of facilities solely dedicated to the treatment of clubfoot and training of individuals as clubfoot treatment specialists. The answer may lie in the integration of clubfoot clinics within existing health systems, using facilities now in use, and training the orthopedic faculties in the more effective form of treatment—the Ponseti method. There are some limitations to this study. The two groups of children enrolled presented to the hospital for treatment and therefore they constitute a convenient sample. The groups differed in age, gender, and unilateral versus bilateral involvement. At Indus Hospital at the time, the Ponseti method was instituted only for children younger than 1 year of age. We believe that the costs estimated in the study underestimate of the actual costs as no health system costs were available from the unsuccessful treatment phase. In addition, costs of complications and longterm follow-up could not be assessed. At the time of the study, the Pehla Qadam program was fairly young; in over 100 patients treated with the Ponseti method only 1 patient did not respond to serial casting and had to undergo corrective surgery. This may well be an underestimate in the long term. Lastly, in the absence of DALYs, probability of successful treatment was used as the effectiveness measure. For the future, there is a need to measure the disability weights and calculate DALYs for clubfoot. Despite the limitations, this study provides scientifically valid estimates of the cost that clubfoot treatment represents in Pakistan. Current societal norms and methods leave a number of children untreated until they are beyond the age at which the Ponseti method might be expected to be effective; therefore these children are subject to the added costs, financial and otherwise, of surgical intervention. Efforts should be made to introduce the Ponseti method throughout the country, as it is clearly the method of choice for treatment of clubfoot in early childhood in resource-constrained settings like Pakistan, as well as nonresource-constrained settings such as the United States. Acknowledgments The authors are grateful to the Finance Department of Indus Hospital for their contribution in articulating the expenditure incurred during treatment of patients enrolled in the study. This work was supported by the Amer Haider Charitable Foundation.
Conflict of interest
None.
References 1. Beals RK (1978) Club foot in the Maori: a genetic study of 50 kindreds. N Z Med J 88:144–146 2. Mittal RL, Sekhon AS, Singh G et al (1993) The prevalence of congenital orthopaedic anomalies in a rural community. Int Orthop 17:11–12 3. Shiels WE 2nd, Coley BD, Kean J et al (2007) Focused dynamic sonographic examination of the congenital clubfoot. Pediatr Radiol 37:1118–1124 4. WHO (2011) World report on disability. World Health Organization, Malta 5. Khan SA, Kumar A (2010) Ponseti’s manipulation in neglected clubfoot in children more than 7 years of age: a prospective evaluation of 25 feet with long-term follow-up. J Pediatr Orthop B 19:385–389 6. Ponseti IV (1992) Treatment of congenital club foot. J Bone Joint Surg 74:448–454 7. Ponseti IV (1998) Correction of the talar neck angle in congenital clubfoot with sequential manipulation and casting. Iowa Orthop J 18:74–75 8. Cosma D, Vasilescu D, Valeanu M (2007) Comparative results of the conservative treatment in clubfoot by two different protocols. J Pediatr Orthop B 16:317–321 9. Cooper DM, Dietz FR (1995) Treatment of idiopathic clubfoot. A 30-year follow-up note. J Bone Joint Surg 77:1477–1489 10. Jowett CR, Morcuende JA, Ramachandran M (2011) Management of congenital talipes equinovarus using the Ponseti method: a systematic review. J Bone Joint Surg 93:1160–1164 11. Gray K, Pacey V, Gibbons P et al (2012) Interventions for congenital talipes equinovarus (clubfoot). Cochrane Database Syst Rev 4:CD008602 12. Morcuende JA, Egbert M, Ponseti IV (2003) The effect of the internet in the treatment of congenital idiopathic clubfoot. Iowa Orthop J 23:83–86 13. Staheli L, Morcuende JA, Pirani S, et al (2009) Clubfoot: Ponseti management. Global Help http://www.global-help.org/. Accessed 3 Dec 2013 14. Herzenberg JE, Radler C, Bor N (2002) Ponseti versus traditional methods of casting for idiopathic clubfoot. J Pediatr Orthop 22:517–521 15. Radler C (2013) The Ponseti method for the treatment of congenital club foot: review of the current literature and treatment recommendations. Int Orthop 37:1747–1753 16. van Gelder JH, van Ruiten AG, Visser JD et al (2010) Long-term results of the posteromedial release in the treatment of idiopathic clubfoot. J Pediatr Orthop 30:700–704 17. Graf A, Hassani S, Krzak J et al (2010) Long-term outcome evaluation in young adults following clubfoot surgical release. J Pediatr Orthop 30:379–385 18. Halanski MA, Huang JC, Walsh SJ et al (2009) Resource utilization in clubfoot management. Clin Orthop Relat Res 467:1171–1179
123
World J Surg 19. Dobbs MB, Nunley R, Schoenecker PL (2006) Long-term followup of patients with clubfeet treated with extensive soft-tissue release. J Bone Joint Surg 88:986–996 20. Richards BS, Faulks S, Rathjen KE et al (2008) A comparison of two nonoperative methods of idiopathic clubfoot correction: the Ponseti method and the French functional (physiotherapy) method. J Bone Joint Surg 90:2313–2321 21. Ferreira LF, Nogueira MP, Pereira JC et al (2011) Comparison of hospital costs and duration of treatment with two different clubfoot protocols. Iowa Orthop J 31:49–51 22. Haft GF, Walker CG, Crawford HA (2007) Early clubfoot recurrence after use of the Ponseti method in a New Zealand population. J Bone Joint Surg 89:487–493 23. Lu N, Zhao L, Du Q et al (2010) From cutting to casting: impact and initial barriers to the Ponseti method of clubfoot treatment in China. Iowa Orthop J 30:1–6
123
24. Pirani S, Naddumba E, Mathias R et al (2009) Towards effective Ponseti clubfoot care: the uganda sustainable clubfoot care project. Clin Orthop Relat Res 467:1154–1163 25. Ippolito E, Farsetti P, Caterini R et al (2003) Long-term comparative results in patients with congenital clubfoot treated with two different protocols. J Bone Joint Surg 85-A:1286–1294 26. Verma A, Mehtani A, Sural S et al (2012) Management of idiopathic clubfoot in toddlers by Ponseti’s method. J Pediatr Orthop B 21:79–84 27. Burfat A, Mohammed S, Siddiqi O et al (2013) Understanding the knowledge and perceptions about clubfoot in Karachi, Pakistan: a qualitative exploration. Iowa Orthop 33:149–152
Cost-Effectiveness of the Ponseti Method for Treatment of Clubfoot in Pakistan Hamidah Hussain • Aziza Moiz Burfat • Lubna Samad • Fayez Jawed • Muhammad Amin Chinoy Mansoor Ali Khan
•
Ó Socie´te´ Internationale de Chirurgie 2014
Abstract Background Clubfoot is disabling, with an incidence of 0.9/1,000 live births to 7/1,000 live births. It affects mobility, productivity, and quality of life. Patients are treated surgically or non-surgically using the Ponseti method. We estimated the cost per patient treated with both methods and the cost-effectiveness of these methods in Pakistan. Methods Parents of patients treated, either surgically or with the Ponseti method, at the Indus Hospital’s free program for clubfoot were interviewed between February and May 2012. We measured the direct and indirect household expenditures for pre-diagnosis, incomplete treatment, and current treatment until the first brace for Ponseti method and the first corrective surgery for surgically treated patients. Hospital expenditure was measured by existing accounts. Results Average per-patient cost was $349 for the Ponseti method and $810 for patients treated surgically. Of these, the Indus hospital costs were $170 the for Ponseti method and $452 for surgically treated patients. The direct household expenditure was $154 and $314 for the Ponseti and surgical methods, respectively. The majority of the H. Hussain (&) Interactive Research and Development (IRD), Suite 508, Shahrah-e-Faisal, Ibrahim Trade Towers, Karachi, Pakistan e-mail: [email protected] A. M. Burfat L. Samad F. Jawed Indus Hospital Research Center, 4th Floor, Indus Hospital, Korangi Crossing, Karachi, Pakistan e-mail: [email protected] M. A. Chinoy M. A. Khan Department of Orthopaedics & Trauma, Indus Hospital, Korangi Crossing, Karachi, Pakistan
costs were incurred pre-diagnosis and after inadequate treatment, with the largest proportion spent on transportation, material, and fee for service. The Ponseti method is shown to be the dominant method of treatment, with an incremental cost-effectiveness ratio of $1,225. Conclusions The Ponseti method is clearly the treatment of choice in resource-constrained settings like Pakistan. Household costs for clubfoot treatment are substantial, even in programs offering free diagnostics and treatments and may be a barrier to service utilization for the poorest patients.
Introduction Clubfoot or congenital talipes equinovarus (CTEV) is a disabling condition with an incidence of 0.9/1,000 live births to 7/1,000 live births in different populations [1–3]. The deformity affects an individual’s mobility, threatens that person’s potential productivity, and reduces his or her standard of living [4]. In Pakistan, surgical intervention has been the preferred treatment for clubfoot for many years for children of all age groups. Owing to limited financial resources, surgical treatment is not always affordable for the patient and this leads to a large burden of untreated or neglected clubfoot. In October 2011, the Ponseti method for treatment of clubfoot—now the standard of care for treatment for uncomplicated cases in infants [5]—was introduced at the Indus Hospital in Karachi. This method does not require extensive surgical intervention or anesthesia. Instead, it uses a series of gentle manipulations to correct the deformity with serial casting and bracing to maintain the correction [6, 7]. Older children continued to be treated surgically.
123
World J Surg
Studies have clearly demonstrated the clinical effectiveness of the Ponseti method with an initial correction rate of 90 % in idiopathic feet and a faster decrease in the Pirani score [8–11]. In comparison with the Ponseti method, treatment by clubfoot surgery is associated with poorer long-term outcomes. Complications such as stiffness and weakness of the foot, as well as pain, have been reported, ultimately leading to secondary procedures [5, 12–17]. However, limited data are available on costeffectiveness. A study from New Zealand concluded that the Ponseti method was less resource intensive for the health system with lower cost and morbidity because the patient spent less time in the hospital [18]. The Ponseti method of treatment is appealing in a resource-constrained setting like Pakistan, where most health care expenses are paid out of pocket and the cost of surgical treatment is prohibitive for most families. The objective of the present study was to estimate the cost per patient treated with both the surgical approach and the Ponseti method from a societal perspective, and to calculate the cost-effectiveness of these methods in Pakistan.
Materials and methods This cross-sectional study was conducted at the Indus Hospital in Karachi, which serves a catchment population of 2.4 million, as well as referrals from other parts of Pakistan. The Indus Hospital is a tertiary care facility where all inpatient services are free of charge to patients. The Pehla Qadam (First Step) Clinic established at the Indus Hospital in October 2011 is a collaborative effort between the Department of Orthopaedics and the Indus Hospital Research Center. Initially, enrolment was restricted to children of age 1 year and below at the time of presentation. Children older than 1 year were treated with surgical treatment, which was accepted as the standard practice at that time. In addition to clinical management, the program team worked to create awareness for early identification and referral in the catchment population, as well as providing close patient follow-up and data management. Two groups of patients were enrolled: group A: 56 consecutive patients were enrolled from the Pehla Qadam Clinic and treated using the Ponseti method from October 2011 till May 2012; group B: 48 patients who underwent surgical correction at the Indus Hospital from July 2007 till May 2012. Orthopedic surgeons and residents provided medical care to both groups. A trained two-member team conducted all interviews at the patient’s home using a standardized cost-estimate tool from February 2012 till May 2012 after informed consent was obtained from parents. Costs were collected from a societal perspective. Data were collected on demographics; income; and direct out-
123
of-pocket expenditures for consultations, diagnostic tests, medicines, hospitalization, meals, and transport. For the Ponseti method, costs were collected from the onset of symptoms, serial casting, and tenotomy until the first brace was applied; for the patients treated with the surgical method, costs were collected from the onset of symptoms, through casting and initial corrective surgery at Indus Hospital, until discharge. The questionnaire was divided into three sections. The first section dealt with costs incurred by families in seeking medical care before the start of the treatment, such as consultation, treatment, travel, and meals (the pre-diagnostic period). The second section covered costs for the period during which patients were receiving treatment but the care was left incomplete because of lack of family resources or for any other reason (incomplete treatment). The last section focused on costs for the current treatment, either serial casting or surgery at the Indus Hospital. Indirect costs from lost earnings were estimated for one family member accompanying the child for outpatient visits, including travel and waiting time; hospital days were estimated at 8 h of work loss per day based on the minimum wage in Pakistan. Health system costs for the current treatment were calculated by reviewing existing accounts for all activities related to treatment. Indus Hospital maintains all medical and treatment expenditure data electronically, with detailed documentation of resources utilized on a case-by-case basis. Resources were quantified and included personnel salaries, capital costs, utilities, overhead, maintenance, and supplies. In addition to costs for clinical care, costs for the awareness campaign, close patient follow-up, and reimbursement for transport were also included as part of the total expenditure. Disability weights, and therefore disability-adjusted life years (DALYs) for clubfoot are not available and were not estimated by the most recent Global Burden of Disease (GBD) study. We used a conservative estimate of successful treatment outcome from the literature (70 % treatment success for surgery and 90 % treatment success for Ponseti) to calculate the incremental cost-effective ratio (ICER) [19, 20]. Costs were collected in Pakistan Rupees (PKR) and converted to US dollars using the average conversion rate for the fiscal year 2011–2012 (1 USD = 90 PKR). The data were double-entered in Microsoft Access (Microsoft Corporation Seattle, WA) and analyzed with Stata (Stata Corp LP. College Station, TX, USA). The review board of Interactive Research and Development (IRD) approved the study.
Results Of the 104 patients enrolled, 56 patients were treated with the Ponseti method (group A) and 48 patients were treated
World J Surg Table 1 Demographic and economic characteristics of patients Group A (Ponseti method) N = 56
Group B (surgery) N = 48
Male
45 (80 %)
28 (58 %)
Female
11 (20 %)
20 (42 %)
Mean age Unilateral
4 months (SD) 37 (66 %)
4 years (SD) 21 (44 %)
Bilateral
19 (34 %)
27 (56 %)
Average number of people in household
4.8
4.8
Average household income/month
PKR 10, 952 USD 122
PKR 10, 985 USD 122
PKR Pakistan rupees, USD U.S. dollars
surgically (group B). The majority of patients were male, with a mean age of 4 months versus 4 years in groups A and B, respectively. Both groups had similar socioeconomic characteristics (Table 1). The direct household expenditure per patient during all treatment periods was $154 for group A as compared to $314 for patients in group B. Of the direct expenditure, over 50 % of costs were paid for treatment that ultimately was not successful. During each treatment phase, travel cost was the largest direct expenditure, ranging from 22 to 60 % (Table 2). Besides the cost to patients, Indus Hospital incurred an additional $170 and $452 for group A and B patients, respectively. For patients treated with the Ponseti protocol, an average of six casts were required, followed by tenotomy in the outpatient setting before the brace was applied. Of the $170 spent by the hospital on the Ponseti treatment, $124 was for clinical care and $45 was for program costs. The program costs included partial reimbursement to patients toward transportation costs (Table 3). For surgical patients, generally one admission was required with an average hospital stay of 5 days. Patients treated surgically had to purchase their own locally made brace costing between $10 and $20, as opposed to patients treated with the Ponseti method, where standardized braces were provided to the patients free of cost as part of the treatment plan (Table 4). The costs were twice as high per patient treated surgically compared to those treated with the Ponseti method for both the patient and the hospital (Table 1). The Ponseti method is shown to be a dominant method of treatment with the ICER of $1,225 (Table 5)
Table 2 Direct and indirect patient costs during all phases of treatment Expense categories
Prediagnostic period Mean number of outpatient visits Mean number of hours for all visits Direct costs Consultation fees Casting Laboratory/radiology/ medication Transport Food Additional costs Indirect cost from lost income Subtotal cost Incomplete treatment Mean number of outpatient visits/patient Mean number of hours for all visits Direct costs Consultation fees Casting Day care procedure/brace Laboratory/radiology/ medication Transport Food Indirect cost from lost income Subtotal cost Current treatment Mean number of outpatient visits Mean number of admissions Mean number of hours for all visits Direct costs Consultation fees Laboratory/radiology/ medication Cast and brace post surgery Transport Food Additional costs Indirect cost from lost income Subtotal cost
Ponseti mean cost per patient (USD)
Surgical mean cost per patient (USD)
3
5
14
25
7 1 6
5 10 6
17 4 0 5
51 10 2 10
39
94
7
13
18
38
10 35 11 2
10 45 29 17
21 6 7
45 17 14
92
176
7
6
0
1
35
70
0 0
0 3
0
17
30 3 1 13
25 13 10 26
48
94
Discussion The results of the present investigation show that the average per patient cost was $349 for the Ponseti method
and $810 for patients treated surgically. The significant difference in the cost for the two methods is primarily based on no hospitalization for the Ponseti method. The
123
World J Surg Table 3 Costs incurred by Indus Hospital per patient treated with the Ponseti method Expense categories
USD
Clinical costs Mean number of outpatient clinic visits (8)
35
Casts
16
Tenotomy
7
Follow-up outpatient clinics
9
Locally made brace Pediatric clinic referral for % of patient Subtotal clinical cost/patient
56 2 124
Program cost Program personal salary Transport money provided to patients
6 10
Communication campaign
10
Program operational cost
19
Subtotal program cost/patient Total cost
45 170
Table 4 Cost incurred by Indus Hospital per patient treated surgically Expense categories
USD
Outpatient costs Mean number of outpatient clinic visits (2)
7
Preoperative clinic and tests
10
Post-surgery follow-up
11
Subtotal outpatient cost/patient
28
Inpatient cost Bed charge per admission Laboratory and pharmacy Surgeon fee
85 18 157
Anesthesia
63
Operating room charges
63
Operating room consumables
39
Subtotal inpatient cost/patient
424
Total cost
452
hospital costs incurred in our study are lower than those reported by other studies [18, 21, 22]. Possible reasons could be that the Indus Hospital is a not-for-profit entity and the costs detailed above are the actual expenditures incurred by the hospital; tenotomies were conducted in an outpatient clinic setting, bringing down the procedurerelated costs significantly; and we did not estimate any costs associated with complications, adherence to brace or long-term follow-up of patients in either group. Household costs for clubfoot are substantial, even when programs such as Pehla Qadam offer free diagnostics and treatment, and these may be a barrier to service utilization
123
for the poorest patients. Over 50 % of all direct expenditures borne by patients were for treatment that did not result in correction of the defect. Large proportions of these costs were for travel, fee for service, and materials. As these costs exceeded the capacity of the household, treatment was left incomplete, thus not achieving optimum results. Other countries have documented similar barriers to treatment [23, 24]. In the absence of any social health insurance system in Pakistan, awareness campaigns and education for early identification and treatment of clubfoot will save resources for both families and the health system. We estimated the rate of treatment success at 70 % with surgery and 90 % with the Ponseti method. Assuming an equal treatment success rate of 90 % for both programs, the Ponseti method still cost less. Patients treated with the Ponseti method spent almost half for incomplete treatment. Assuming the cost of inadequate treatment is the same for both groups, the cost per patient for the Ponseti group increases to $433. Even then, the Ponseti method is less expensive than the surgical option. The high cost of treatment of clubfoot is many times over the estimated $22 per capita expenditure on health in Pakistan. Based on the crude birth rate, the birth cohort for Pakistan is an estimated 5 million children. The estimated incidence of clubfoot in Pakistan is 1.5/1,000 children. This means that there are approximately 7,500 children born with clubfoot in Pakistan every year. If all children with clubfoot are treated with the Ponseti method, the cost will be $2,617,500 as compared to $6,075,000 if children are treated surgically—a considerable savings. Our study clearly shows that in a setting like Pakistan it is cost effective to treat patients with the Ponseti method rather then using surgical intervention. In addition, long-term follow-up studies have shown better functional results, a lower relapse rate, and less morbidity over time, including pain, osteoarthritis, and flatfoot deformity for children treated with the Ponseti method [25]. Management of these postoperative complications will place a high financial burden on the health system. In addition, there will be a comparable burden on society with loss of productivity if these 7,500 children with clubfoot are treated surgically and have difficulty at work as a result of postoperative difficulties after correction of clubfoot. Although cost is a very important factor, there are many other elements that need to be taken into consideration for policy development of clubfoot care in Pakistan. The success of clubfoot treatment is associated with early identification and referral to an appropriate health care facility [26]. However, clubfoot is widely misunderstood in Pakistani populations with a lack of awareness about the disorder and the treatment method [27]. A strong community mobilization arm can therefore be instrumental in
World J Surg Table 5 Cost saved per successfully treated patient Alternative strategy
Cost, USD
Incremental cost, USD
Effect (% of patients treated successfully)
Incremental effect (%)
ICER, USD
Surgical
39,163
–
70
–
–
Ponseti
19,556
19,607
90
20
1,225
ICER incremental cost-effectiveness ratio
the success of a clubfoot care program. The prevalence of clubfoot is not high enough to justify development of facilities solely dedicated to the treatment of clubfoot and training of individuals as clubfoot treatment specialists. The answer may lie in the integration of clubfoot clinics within existing health systems, using facilities now in use, and training the orthopedic faculties in the more effective form of treatment—the Ponseti method. There are some limitations to this study. The two groups of children enrolled presented to the hospital for treatment and therefore they constitute a convenient sample. The groups differed in age, gender, and unilateral versus bilateral involvement. At Indus Hospital at the time, the Ponseti method was instituted only for children younger than 1 year of age. We believe that the costs estimated in the study underestimate of the actual costs as no health system costs were available from the unsuccessful treatment phase. In addition, costs of complications and longterm follow-up could not be assessed. At the time of the study, the Pehla Qadam program was fairly young; in over 100 patients treated with the Ponseti method only 1 patient did not respond to serial casting and had to undergo corrective surgery. This may well be an underestimate in the long term. Lastly, in the absence of DALYs, probability of successful treatment was used as the effectiveness measure. For the future, there is a need to measure the disability weights and calculate DALYs for clubfoot. Despite the limitations, this study provides scientifically valid estimates of the cost that clubfoot treatment represents in Pakistan. Current societal norms and methods leave a number of children untreated until they are beyond the age at which the Ponseti method might be expected to be effective; therefore these children are subject to the added costs, financial and otherwise, of surgical intervention. Efforts should be made to introduce the Ponseti method throughout the country, as it is clearly the method of choice for treatment of clubfoot in early childhood in resource-constrained settings like Pakistan, as well as nonresource-constrained settings such as the United States. Acknowledgments The authors are grateful to the Finance Department of Indus Hospital for their contribution in articulating the expenditure incurred during treatment of patients enrolled in the study. This work was supported by the Amer Haider Charitable Foundation.
Conflict of interest
None.
References 1. Beals RK (1978) Club foot in the Maori: a genetic study of 50 kindreds. N Z Med J 88:144–146 2. Mittal RL, Sekhon AS, Singh G et al (1993) The prevalence of congenital orthopaedic anomalies in a rural community. Int Orthop 17:11–12 3. Shiels WE 2nd, Coley BD, Kean J et al (2007) Focused dynamic sonographic examination of the congenital clubfoot. Pediatr Radiol 37:1118–1124 4. WHO (2011) World report on disability. World Health Organization, Malta 5. Khan SA, Kumar A (2010) Ponseti’s manipulation in neglected clubfoot in children more than 7 years of age: a prospective evaluation of 25 feet with long-term follow-up. J Pediatr Orthop B 19:385–389 6. Ponseti IV (1992) Treatment of congenital club foot. J Bone Joint Surg 74:448–454 7. Ponseti IV (1998) Correction of the talar neck angle in congenital clubfoot with sequential manipulation and casting. Iowa Orthop J 18:74–75 8. Cosma D, Vasilescu D, Valeanu M (2007) Comparative results of the conservative treatment in clubfoot by two different protocols. J Pediatr Orthop B 16:317–321 9. Cooper DM, Dietz FR (1995) Treatment of idiopathic clubfoot. A 30-year follow-up note. J Bone Joint Surg 77:1477–1489 10. Jowett CR, Morcuende JA, Ramachandran M (2011) Management of congenital talipes equinovarus using the Ponseti method: a systematic review. J Bone Joint Surg 93:1160–1164 11. Gray K, Pacey V, Gibbons P et al (2012) Interventions for congenital talipes equinovarus (clubfoot). Cochrane Database Syst Rev 4:CD008602 12. Morcuende JA, Egbert M, Ponseti IV (2003) The effect of the internet in the treatment of congenital idiopathic clubfoot. Iowa Orthop J 23:83–86 13. Staheli L, Morcuende JA, Pirani S, et al (2009) Clubfoot: Ponseti management. Global Help http://www.global-help.org/. Accessed 3 Dec 2013 14. Herzenberg JE, Radler C, Bor N (2002) Ponseti versus traditional methods of casting for idiopathic clubfoot. J Pediatr Orthop 22:517–521 15. Radler C (2013) The Ponseti method for the treatment of congenital club foot: review of the current literature and treatment recommendations. Int Orthop 37:1747–1753 16. van Gelder JH, van Ruiten AG, Visser JD et al (2010) Long-term results of the posteromedial release in the treatment of idiopathic clubfoot. J Pediatr Orthop 30:700–704 17. Graf A, Hassani S, Krzak J et al (2010) Long-term outcome evaluation in young adults following clubfoot surgical release. J Pediatr Orthop 30:379–385 18. Halanski MA, Huang JC, Walsh SJ et al (2009) Resource utilization in clubfoot management. Clin Orthop Relat Res 467:1171–1179
123
World J Surg 19. Dobbs MB, Nunley R, Schoenecker PL (2006) Long-term followup of patients with clubfeet treated with extensive soft-tissue release. J Bone Joint Surg 88:986–996 20. Richards BS, Faulks S, Rathjen KE et al (2008) A comparison of two nonoperative methods of idiopathic clubfoot correction: the Ponseti method and the French functional (physiotherapy) method. J Bone Joint Surg 90:2313–2321 21. Ferreira LF, Nogueira MP, Pereira JC et al (2011) Comparison of hospital costs and duration of treatment with two different clubfoot protocols. Iowa Orthop J 31:49–51 22. Haft GF, Walker CG, Crawford HA (2007) Early clubfoot recurrence after use of the Ponseti method in a New Zealand population. J Bone Joint Surg 89:487–493 23. Lu N, Zhao L, Du Q et al (2010) From cutting to casting: impact and initial barriers to the Ponseti method of clubfoot treatment in China. Iowa Orthop J 30:1–6
123
24. Pirani S, Naddumba E, Mathias R et al (2009) Towards effective Ponseti clubfoot care: the uganda sustainable clubfoot care project. Clin Orthop Relat Res 467:1154–1163 25. Ippolito E, Farsetti P, Caterini R et al (2003) Long-term comparative results in patients with congenital clubfoot treated with two different protocols. J Bone Joint Surg 85-A:1286–1294 26. Verma A, Mehtani A, Sural S et al (2012) Management of idiopathic clubfoot in toddlers by Ponseti’s method. J Pediatr Orthop B 21:79–84 27. Burfat A, Mohammed S, Siddiqi O et al (2013) Understanding the knowledge and perceptions about clubfoot in Karachi, Pakistan: a qualitative exploration. Iowa Orthop 33:149–152
