ORIGINAL ARTICLE

The Bangladesh Clubfoot Project: The First 5000 Feet

Vikki A. Ford-Powell, BSc,* Simon Barker, MD,w Md Shariful I. Khan, BSc,z
Angela M. Evans, PhD,y8 and Fredrick R. Deitz, MDz

setting with appropriate logistical support. The use of local

Background: Bangladesh is one of the most populous countries physiotherapists and paramedics yielded good clinical outcomes
in the world at 160 million with 1/3 existing below the poverty in an environment with full access to clinical review and ongoing
line. With an annual birth rate of approximately 3.2 million, an training. A higher than expected number of atypical cases have
estimated incidence of 1:900 live births, the country has ap- been noted, requiring modified Ponseti treatment. Complica-
proximately 5000 new cases of idiopathic congenital talipes tions have been few at this early stage.
equinovarus per annum. The Bangladesh sustainable clubfoot Level of Evidence: Level 2—therapeutic study.
program, Walk for Life (WFL), was conceived to respond to
this unmet need. Key Words: clubfoot, Bangladesh, Ponseti, Pirani, developing
Methods: WFL started in 2009 and has rapidly grown to 35 nations, government, NGO, health and development
clinics. Overseas experts initially increased local capacity by (J Pediatr Orthop 2013;33:e40–e44)
training a team of national paramedical staff. Government
support enabled integration with the public hospital system and
enhanced sustainability. WFL has supplied materials, educa-
tional, administrative, and clinical support throughout. All
recruited cases underwent Ponseti casting. Demographic, Pirani
T he effective treatment of idiopathic congenital talipes
equinovarus (ICTEV), by means of the Ponseti
method, is well established to generate the best outcome
scores, cast, tenotomy, and bracing data have been prospectively
collected from all patients. Detailed review has been undertaken
with least complications.1,2 The Ponseti technique has
for 1040 patients after 12 months of treatment in 3 divisions of
also been shown to be resource appropriate for the de-
Bangladesh.
veloping world,3,4 yet access to this treatment for many
Results: Between 2009 and 2011, 6069 feet (3922 patients) were
remains limited and significant challenges have been en-
recruited to the project. Of these 1643 feet (1040 patients) have
countered in its introduction.5 Key issues encountered
completed a minimum of 1-year follow-up. The male:female
have included the accurate and early identification of
ratio was 2.7:1 with a mean age of 22 months at presentation
ICTEV, use and supervision of paramedical staff, access
(range, 0 to 36). Typical idiopathic congenital talipes equi-
to materials,6 and patient compliance with the treatment
novarus responded in a median of 5 casts (range, 1 to 25) with
protocol.7
76% undergoing tenotomy. Thirteen percent were atypical feet
Untreated ICTEV precludes normal function, pre-
requiring a median of 5 casts. The percentage of patients missing
disposes to further pathology, limits employment pros-
at the 12-month point was 12%. Two percent of patients ex-
pects, and is a significant barrier to societal acceptance.
perienced complications.
Within low-income environments, these issues are mag-
Conclusions: The Bangladesh clubfoot program demonstrates
nified as economic survival depends largely upon physical
that rapid case ascertainment is possible in a developing world
labor and community inclusion.
On the basis of a birth rate of 27 per 1000, a pop-
ulation of 162 million8 and an estimated ICTEV preva-
From the *Walk For Life Ponseti, LAMB Project, Dinajpur; zClinical lence of 1:900 live births, there is an expectation of 4851
Services, Walk For Life, Dhaka, Bangladesh; wDepartment Paedi-
atric Orthopaedic Surgery, Royal Aberdeen Childrens Hospital,
new cases of ICTEV in Bangladesh each year. The Ban-
Aberdeen, UK; yWalk For Life, AUT University, Auckland, New gladesh Clubfoot Project, Walk for Life (WFL), a regis-
Zealand; 8Walk For Life, University of South Australia, Adelaide, tered nongovernment organization (NGO) in Bangladesh,
SA, Australia; and zDepartment of Orthopaedic Surgery, University was established in 2009. The aim of this project is to
of Iowa, Iowa City, IA. provide access for the Bangladesh population to Ponseti
Project sponsorship and leadership from Colin Macfarlane and the
Glencoe Foundation. None of the authors received payment for this treatment for all children below the age of 4, within a
article. 60 km radius of home. The results of the first 2 years are
Md.S.I.K. is employed by The Glencoe Foundation. A.M.E. has a presented, including an in-depth review of clinics after the
consultancy with Australian Health Professional Regulatory Agency, first year and discussion of key factors that have facili-
a small project grant with Medicine Today journal and receives
payments for lecture, royalties and travel costs outside of the sub-
tated the program.
mitted work. The remaining authors declare no conflict of interest.
Reprints: Vikki A. Ford-Powell, BSc, Walk For Life Ponseti, LAMB METHODS
Project, Rajabashor, Parbatipur, Dinajpur 5250, Bangladesh.
E-mail: vikkifp@gmail.com. In 2009, a pilot project was carried out at LAMB
Copyright r 2013 by Lippincott Williams & Wilkins community health and development NGO situated in

e40 | www.pedorthopaedics.com J Pediatr Orthop  Volume 33, Number 4, June 2013

J Pediatr Orthop  Volume 33, Number 4, June 2013 The Bangladesh Clubfoot Project

Northwest Bangladesh. Any child below 3 years of age

with a diagnosis of ICTEV was accepted by WFL for
Ponseti treatment free of charge. After the success of the
pilot project, further clinics were established opportun-
istically in Government hospitals located in several urban
areas of Bangladesh. This was done by agreement with
local clinicians, using project physiotherapists who were
taught by visiting overseas Ponseti trainers. A memo-
randum of understanding (MOU) with the Government
of Bangladesh, committing to joint working on the proj-
ect aims, led to the expansion of the program nationwide
and helped to overcome resistance to the introduction of a
novel technique. In-country Steenbeek brace manufactur-
ing was organized using a Bangladeshi blacksmith and
cobbler, who reproduced the brace after the original spec-
ifications. Satellite clinics in rural locations were sub- FIGURE 1. Feet affected across first 14 clinics after 12 months.
sequently established to provide weekly clinics throughout
Bangladesh and located such that no patient needs to travel
further than 60 km to reach a clinic. A “hub and spoke” more limited data set that were not included in the de-
model of delivery has been developed with physiotherapy tailed review.
teams in 8 key urban “hubs,” providing clinics in 35 mixed Mean age at first cast was 22 months with a range of
urban and rural “spoke” clinics. 0 to 36. There were 6 outliers who ranged from 3 to 5.5
In all clinics, a comprehensive patient record has years who were accepted on a case-by-case basis for a trial
been prospectively gathered including demographic data, of Ponseti treatment. After 12 months, a detailed review
serial Pirani scores, tenotomy practice, and subsequent of 1040 patients, the total number treated in the first year,
progress with bracing. was carried out. This involved 12 “WFL” clinics in op-
Patient data are stored on hard copy and uploaded to eration, with 2 more in the process of opening. Of these,
an Excel (Microsoft, Dhaka, Bangladesh) database by 44% were unilateral clubfoot cases and 56% bilateral
project administrators on a monthly basis. WFL is respon- (Fig. 1) with a male:female ratio of 2.7:1 (Fig. 2). Forty-
sible for ethical practice; all data have been anonymized. five percent of the children were in braces, and the mean
In 3 centers (Dhaka, Khulna, and Rajshahi), the loss to follow-up was 12%, with a further 2% referred for
lead author (V.A.F-P.) has undertaken a standardized surgery. A median number of 5 casts were used per foot
review of practice. Using a mixture of quantitative rating (range, 1 to 25), and the national mean tenotomy rate
and qualitative reporting, a culturally appropriate non- was 76%.
validated outcome measurement tool was developed The national mean initial Pirani score was 4/6 with
based upon the widely accepted Pirani scoring system and a mean corrected score of 0.5/6. In total, 82% of children
Ponseti treatment protocols. The tool maps Pirani Scores progressed to boots and bar by the 12-month review.
for all patients assesses casting technique, explores access Isolating outliers (3% of children who required 11 to 25
to doctors and tenotomy facilities, reviews tenotomy casts), the rate of children progressing to boots and bar
protocol and rates, notes all complications, and, further, within 10 casts was 79%. From the remaining 21%, 2%
addresses clinic management issues, treatment com-
pliance, and parental satisfaction. The records of the 1040
patients who had completed 12 months were audited to
identify trends and review practice. Subsequently, an
audit was carried out involving an extended visit to each
clinical team to achieve a complete picture of treatment.
Feedback, including strengths and challenges, was given
to the clinical team, and issues were reported back to the
WFL management team to facilitate improvement.

RESULTS
Between 2009 and 2011, a total of 3922 patients
have entered the project. Of these, 3358 patients were
in WFL and LAMB clinics. A full data set is available
for 3249 patients. Partner clinics initiated by “Zero
Clubfoot” (Chittagong Lions Foundation and Pedrollo
Welfare Foundation) covering around 20% of the Ban- FIGURE 2. Male:female ratio across first 14 clinics after 12
gladesh population in the Chittagong division provided a months.

r 2013 Lippincott Williams & Wilkins www.pedorthopaedics.com | e41

Ford-Powell et al
Children progressing to
braces within 10 casts
Children progressing to
braces within 25 casts
Missing
Referred for surgery
incomplete data
FIGURE 3. Treatment outcomes at 12-month points (1040
patients).
were referred for surgery, 12% were missing from treat-
ment, and incomplete data prevented analysis of the re-
maining 7% (Fig. 3). The number of casts required for
correction was in proportion to the severity of the pre-
senting Pirani score of feet as demonstrated in Table 1.
Figure 4 demonstrates that the modal correction
pattern of the 1040 children in the detailed study group
was typical for Ponseti treatment. In this group, 3% (31)
of the children required between 11 and 25 casts, and of
these only 26%7 were atypical clubfoot cases.
Complications encountered included pressure sores,
skin infections, and casts broken at the knee and heel.
The range of complications varied between centers under
review from 9.5% in 1 center to 2% in another. Those
patients still in casts in the centers undergoing detailed
review were not included due to incomplete data collection.
Ponseti et al9 defined atypical or complex clubfoot
as clubfeet “in severe equinus and supination, short and
stubby, with the adducted metatarsals in plantar flexion
and deep creases across the sole of the foot and above the
heel.” This is a qualitative definition and the diagnosis in
the clinics was suggested by a deep crease across the entire
sole of the foot, a deep crease above the heel, and un-
usually severe cavus and equinus. On the basis of these
diagnostic criteria, atypical or complex clubfoot was
identified in 13% (135) of the patients seen. These feet
required a median of 5 casts for correction. Clinic audits
suggest that the percentage of atypical feet may be even
higher when strictly applying these diagnostic criteria.
Complications were more common in feet labeled atyp-
ical. The audit visits identified 16% (166) of children
treated fell into the category of neglected clubfoot being
TABLE 1. Severity of Pirani Score at Presentation and Number
of Casts Required to Correct
Pirani Severity Modal No. Casts Required Range
6 5 1-25
5 5 1-14
4 4 1-19
3 1 1-15
2 1 1-9
e42 | www.pedorthopaedics.com
J Pediatr Orthop  Volume 33, Number 4, June 2013
FIGURE 4. Pirani mapping across 14 clinics 12 months after
pilot.
above the age of 2 years and 1% (10) of these were above
5 years old. These feet were harder to correct, taking
longer, and demonstrated an increased incidence of re-
currence. Key strengths and challenges identified from the
clinic visits can be seen in Figure 5.
DISCUSSION
Project Organization
Although Owen et al5 have reported a larger series
of 7705 cases, these were drawn from 10 different nations,
with no single nation contributing >651 patients. The
Bangladeshi clubfoot project has therefore secured an
unprecedented rate of case ascertainment in its first 2
years. Several factors are recognized to have underpinned
this:
1. A project leader with business expertise has delivered
timelines that are seldom realized in a (developing
world) medical project setting.
2. Effective mass communication through multiple media
routes.
3. The partnership of resources between Government
hospitals and the project has been pivotal in expanding
the project nationwide.
4. Overseas volunteer trainers delivered early momentum
to clinics and has facilitated rapport across profes-
sional boundaries in the Bangladeshi health care
system.
5. Partnership working between project employed practi-
tioners (physiotherapists) and local government doc-
tors (orthopaedic surgeons).
6. A “hub and spoke” clinic model with practitioners
located in 8 hubs delivering clinics in 35 locations,
augmented in 3 areas by partnerships with other
NGOs to avoid duplication.
7. Central administrative support and logistics for
materials supply.
8. Employment of a Bengali Clinical Director and Country
Director for program and personnel management has
r 2013 Lippincott Williams & Wilkins
J Pediatr Orthop  Volume 33, Number 4, June 2013
Strengths
Case ascertainment
Functional assessments and outcomes
Time and case management
Consistent Pirani scoring
Competent Ponseti technique
High level of satisfactory outcomes
Spacious, well ventilated clinical
environments
Interdisciplinary team work
MoU with Government – national
ownership of programme
Robust data collection
FIGURE 5. Strengths and challenges identified from clinic visits.
increased the sustainability of the project and reduced
dependence on visiting experts.
Most significantly, in the author’s view, an NGO/
government equal partnership has produced a sustainable
system that has meaningful national ownership. Relation-
ships with all levels of management, from the local civil
servants in each hospital through to the Minister of Health,
have grounded the future success of the project. Without
this level of local investment in a project, the long-term
sustainability would be at risk. Many partnership projects
exist within Bangladesh, but few have secured an MOU
with the government, defining a partnership based on re-
lationships that produce an entirely equal responsibility for
the provision of care.
The majority of referrals have come from the medical
profession, indicating that the MOU with the government,
aiming to increase the credibility of the treatment, has
generated the necessary awareness amongst local clinicians.
The high proportion of parent-to-parent referral is taken to
be an encouraging surrogate measure of parental sat-
isfaction. Further clinics are scheduled to open that should
further enhance case ascertainment.
Project Outcomes
Overall, correction of clubfeet using the Ponseti
method through the WFL project was good, with over
80% wearing boots and bars at the 12-month review.
Loss to follow-up is concerning and raises the possibility
of selection bias in the cohort under review as those
missing may include unknown complications. Each clinic
was able to account for each patient treated while still
within the program; however, some loss was incurred
transferring data from regions to the central office, re-
flecting the problems of data collection within a resource-
limited setting. However, the database and mapping from
all the clinics provided an ideal opportunity to study
outcome measures, demographic implications, public
health implications, and disease trends.
Although all the clinics were led by a trained
therapist, they often relied on assistants, with limited or
no training, as the second practitioner. Nevertheless, the
r 2013 Lippincott Williams & Wilkins
The Bangladesh Clubfoot Project
Challenges
Overcrowded clinics
Maintaining quality under pressure
Consistent location of head of talus
Manipulation time decreased
Management of pressure sores
Access to timely tenotomise with
reduced tenotomy rates
Sterile techniques for tenotomy,
Recognising and dealing with atypical
feet
Ergonomic difficulties
Lack of hand washing facilities
correction rate remained consistently high, showing the
efficacy of the Ponseti method as staffed in this project.
Pirani score mapping nationally showed that staff fol-
lowed the Ponseti treatment protocol well with limited
complications. Despite the poor infrastructure and pov-
erty of the country, compliance to the lengthy treatment
was considered satisfactory with only 12% lost to follow-
up after 2 years.
Program issues differed from area to area depending
upon the number of trained staff, skill levels, and the
working environment. Some shared issues were common
throughout the program. Clinic overcrowding, decreased
manipulation time due to pressure of work, and poor
recognition of atypical feet were highlighted as the main
issues. Complications were noted in 2% of cases and re-
flected slipped casts or poor recognition of the talar ful-
crum. Often these could be attributed to misdiagnosis of
atypical feet or pressure of case numbers within the
clinics. The range of casts required for correction ranged
from 1 to 25. The reasons recorded for using over 10 casts
included: atypical clubfeet, an older “neglected” case, cast
breakage, and complications (eg, pressure sores requiring
time without casts).
Each clinic has been organized with 2 staffs in a
limited space, resulting in a recommended maximum of
20 patients per clinic. Demand has exceeded initial pro-
vision in most clinics with deleterious effect on clinical
practice such as rushed treatment resulting in poor tech-
nique, lengthened treatment, and pressure sores. Pirani
score mapping demonstrated that although the treatment
protocol was followed, the correction of the adductus
took longer than expected in many cases, suggesting
manipulation was not adequate before cast application.
This may reflect the limited time available for each pa-
tient, contributing to increased treatment time before te-
notomy after the cavus correction. WFL have responded
to the high demand by increasing the number of clinics.
A higher than expected incidence of atypical club-
foot was noted (13%). Across the clinics, early recognition
of atypical clubfeet was poor. Although the number of
reported atypical feet on the database was high, during the
www.pedorthopaedics.com | e43
Ford-Powell et al
audit visits many more atypical clubfeet were seen that
had not been diagnosed. Hence, for these infants the
necessarily modified Ponseti method had not been used,
resulting in increased complications (eg, slipped casts,
edema, and pressure sores) being seen when compared
with the typical group. This reflected the emphasis during
initial training for the new Ponseti practitioners to be
competent in the standard treatment process. Mis-
diagnosis of atypical feet can lead to complications during
the casting process, an increase in symptoms, and deteri-
oration in treatment outcome.9 Given the deficit in train-
ing and the apparent high incidence of atypical feet, WFL
has carried out a second wave of advanced training to
address this need. A further review is necessary to evaluate
the gain from this training.
Limitations of the current study include the un-
certain incidence of ICTEV in Bangladesh, limiting the
reliability of population statistics. This is a common
problem in developing countries. The data are weakened
by loss to follow-up (12%) and incomplete data sheets
(7%) raising the concern of selection bias in this cohort.
Further research is required to ascertain whether the early
success of the program is sustained and to evaluate lon-
ger-term results, including compliance with the boots and
bar, and recurrence rate of the deformity.
CONCLUSIONS
Previous use of the Ponseti method for the correc-
tion of ICTEV within resource-limited settings has been
successful, but data collection and sustainability has been
problematic.6,7 WFL have rapidly created a successful,
sustainable treatment for ICTEV patients that works
within the Bangladeshi cultural setting and maximizes
on already existing health structures. A combination of
business skills, clinical specialist input, ongoing funding,
and supervision has produced a quality of service that
equals those within developed nations. The results col-
lected from the first 2 years and the success of the MOU
with the government suggest long-term sustainability
and ongoing quality assurance ensures standards are
maintained. Further research is required to assess the long-
term outcomes from treatment and the long-term sus-
tainable nature of this model of project implementation.
e44 | www.pedorthopaedics.com
J Pediatr Orthop  Volume 33, Number 4, June 2013
ACKNOWLEDGMENTS
The authors thank Colin Macfarlane, founder of
Walk for Life, Partner clinics at LAMB Hospital,
Dinajpur; Zero Clubfoot project, Chittagong, CRP, Dhaka;
Government of Bangladesh with special thanks to Professor
A.F.M. Ruhal Haque, MP, FRCS (Edin) Hon’ble Minister
of health and Family Welfare; Dr Kh. Abdul Awal Rizvi,
Director of National institute of Trauma & Orthopedic
Rehabilitation (NITOR); all orthopaedic surgeons in
participating clinics; and volunteer overseas visiting. The
authors also thank Ponseti trainers especially Dr Paul
Wade, Podiatric Surgeon, Adelaide, SA, Australia;
Professor Shafique Pirani, Orthopedic Surgeon, Vancouver;
Denise Watson Physiotherapist, London, UK; Kate Lock,
Hospital Hdministrator, Dorset UK; Steve Mannion,
Orthopedic Surgeon, Blackpool, UK and “Walk for life”
team in Bangladesh, specifically Jahangir Alam (Country
Director) and Dr Abdur Rouf (Honorary Medical Advisor).
REFERENCES
1. Cooper DM, Dietz FR. Treatment of Idiopathic clubfoot. A thirty
year follow-up note. J Bone Joint Surg Am. 1995;77:1477–1489.
2. Dobbs M, Nunley R, Schoenecker P. Long-term follow-up of
patients with clubfeet treated with extensive soft-tissue release.
J Bone Joint Surg Am. 2006;88:986–996.
3. Pirani S, Carlson W. Factors affecting compliance in the treatment
of congenital clubfoot: the Uganda clubfoot project: 201 (abstract).
J Invest Med. 2006;54:S114–S115.
4. Tindall AJ, Steinlechner CW, Lavy CB, et al. Results of manipulation
of idiopathic clubfoot deformity in malawi by orthopaedic clinical
officers using the Ponseti method: a realistic alternative for the
developing world? J Pediatr Orthop. 2005;25:627–629.
5. Owen RM, Norgrove Penny J, Mayo AC, et al. A collaborative
public health approach to clubfoot intervention in 10 low-income and
middle-income countries: 2-year outcomes and lessons learnt. J Pediatr
Orthop Br. 2012;21:361–365.
6. Lavy CB, Mannion SJ, Mkandawire NC, et al. Club foot treatment in
Malawi—a public health approach. Disabil Rehabil. 2007;29:857–862.
7. McElroy T, Konde-Lule J, Neema S, et al. The Ugandan Sustainable
Clubfoot Care Project. Understanding the barriers to clubfoot
treatment adherence in Uganda: a rapid ethnographic study. Disabil
Rehabil. 2007;29:845–855.
8. World Bank. Development Indicators. April 15, 2011. Available at:
http://issuu.com/world.bank.publications/docs/9780821387092. Ac-
cessed on February 28, 2012.
9. Ponseti IV, Zhivkov M, Davis N, et al. Treatment of the complex
idiopathic clubfoot. Clin Orthop Relat Res. 2006;451:171–176.
r 2013 Lippincott Williams & Wilkins