ORIGINAL ARTICLE

Patient-Reported Outcome Measurement Information

System (PROMIS) Scores in Pediatric Idiopathic
Toe Walkers
Sarah E. Lindsay, MD,* Jeremy Bauer, MD,† Daniel Bouton, MD,† Patrick Do, BS,†
Carly Woodmark, MSc,‡ Susan Sienko, PhD,† and Ellen M. Raney, MD, FAAP,†

Mobility scores than children did. There were no other significant

Background: Idiopathic toe walking (ITW) can result in early differences between patient-reported and parent-reported
contact with the health care system and be distressing for patients PROMIS scores. PROMIS scores did not differ significantly
and their families. The natural history of ITW is poorly char- between those <10 years and those ≥ 10 years.
acterized. Deciding how and when to intervene can be difficult. Conclusion: In this cohort of 45 otherwise healthy children
Patient-reported outcomes are utilized in the clinical setting to without other neurologic diagnoses, there are both clinically and
assess patient factors and indications that may better inform statistically significant differences in PROMIS means between
treatment plans. Patient-Reported Outcomes Measurement In- our cohort and the healthy age-matched population. These dif-
formation System (PROMIS) is an instrument designed to collect ferences manifested in worse peer relationships and mobility
patient-reported outcomes. Minimum clinically important dif- scores.
ferences in PROMIS metrics have been established to facilitate Level of Evidence: Level IV.
clinical relevance and utility of these metrics. The purpose of this
study was to characterize the patient perspective of ITW by Key Words: idiopathic toe walking, PROMIS, patient outcomes
utilizing the PROMIS scores. (J Pediatr Orthop 2022;42:e878–e881)
Methods: Retrospective chart review was performed to identify
children aged 5 to 17 with a diagnosis of ITW treated at a single
tertiary care center between 2017 and 2020. Inclusion criteria
were a diagnosis of ITW and completion of a PROMIS ques-
tionnaire. Exclusion criteria were neurologic disease, autism, and
previous surgical treatment. Demographic, physical exam,
I diopathic toe walking (ITW) is a diagnosis of exclusion
that describes a condition in which a child over the age of
3 who consistently walks in a toe-toe gait pattern in the
treatment, and available motion analysis data were collected. absence of a neurologic or developmental diagnosis.1
PROMIS scores for the following domains were available: Mo- Children with ITW may or may not have limitation in
bility, Peer Relationships, and Pain Interference. dorsiflexion. The prevalence of toe walking in children is
Results: Forty-five children were enrolled. Seventy-three percent estimated to be between 2% and 5%.2–4 The condition is
of PROMIS scores were patient reported while the remainder concerning for both children and caregivers, and it often
were parent reported. PROMIS score means for the cohort by results in early contact between the child and the health
domain were Mobility: 45.2 ± 8.2 (P < 0.000); Peer Relation- care system. The natural history of ITW is uncertain and is
ships: 46.4 ± 11.6 (P = 0.047); and Pain Interference: 47.4 ± 9.5 debated in the literature, and determining how and when
(P = 0.67). Motion analysis data, available for 11 children, noted to intervene for children with ITW can be a difficult de-
age-matched gait velocity negatively correlated (rs = −0.652, cision. Previous work has demonstrated that ITW resolves
P = 0.03) with Peer Relationships. No correlations were found spontaneously in the majority of children and that surgery
between other aspects of gait, body mass index, or limitations in should be reserved for only those with fixed contractures.3,5
dorsiflexion and PROMIS domains. Parents reported lower However, other work suggests that toe walking persists in
up to 50% of children who do not undergo casting or
From the *Oregon Health and Science University; †Shriners Hospitals surgical interventions.6,7
for Children, Portland, OR; and ‡BIO Ventures for Global Health, Treatment options include observation, serial casting,
Seattle, WA. bracing, botulinum toxin A injections, and surgery, and there
Approval was obtained from Western Institutional Review Board
(WIRB) for a retrospective chart review. This study was performed in
exists considerable controversy regarding optimal
accordance with the 1964 Declaration of Helsinki and US HIPAA management.8 A 2019 Cochrane Systematic Review by
regulations. Caserta et al9 demonstrated insufficient evidence regarding
The authors declare no conflicts of interest. outcomes for nonoperative treatment modalities for ITW.
Reprints: Ellen M. Raney, MD, FAAP, Shriners Hospitals for Children, Surgery (gastrocnemius or tendo-Achilles lengthening) is
3101 SW Sam Jackson Park Road, Portland, OR 97239. E-mail:
[email protected]. suggested for children with refractory or severe toe walking;
Copyright © 2022 Wolters Kluwer Health, Inc. All rights reserved. however, abnormal gait kinematics have been shown to
DOI: 10.1097/BPO.0000000000002198 persist in some despite interventions.10–13 Uncertain natural

e878 | www.pedorthopaedics.com J Pediatr Orthop  Volume 42, Number 8, September 2022

Copyright r 2022 Wolters Kluwer Health, Inc. All rights reserved.

J Pediatr Orthop  Volume 42, Number 8, September 2022 PROMIS Scores in Pediatric Idiopathic Toe Walkers

history combined with clinical equipoise regarding treatment The range of scores for PROMIS domains are 34 to 81 for
modalities can be challenging for both children and their Pain Interference, 14 to 59 for mobility, and 17.68 to 64.44
parents. for Peer Relationships.20
Patient-reported outcomes (PROs) are increasingly Gait analysis, at our institution, included a com-
utilized in the clinical setting to assess patient factors and prehensive history and a physical examination with goni-
indications that may better inform treatment plans. PROs ometers for joint range, complete kinematics, video
can be important in guiding patient-centered therapy and recording of gait, and kinetic analysis. After the collection
in encouraging shared decision making, which are both of anthropometric data, patients had retroreflective
crucial in situations with clinical equipoise. Patient- markers placed, and instrumented gait analysis was per-
Reported Outcomes Measurement Information System formed per method previously described.21 The children
(PROMIS) is an instrument designed to collect PROs in a walked at their self-selected speed in their typical toe
dynamic manner and is designed to measure 8 domains walking gait until a minimum of 3 trials were performed
across physical, social, and psychological health. The and recorded, and the average of the 3 trials was used in
PROMIS questionnaire is normalized to an age-matched the statistical analysis.
reference cohort, which uses a T-score metric with a mean
of 50 and SD of 10. The minimum clinically important Statistics
difference (MCID) is defined as “the smallest difference in The PROMIS scores in patients with a diagnosis of
score in domain of interest that patients perceive as im- ITW were the primary outcomes measured in this study.
portant, either beneficial or harmful, and that would lead Descriptive statistics including mean, SD, and frequency
the clinician to consider a change in the patient’s man- counts were performed to characterize the study cohort.
agement.”14 The MCID is reported to be 2 to 3 points for Data were assessed for normal distribution. On the
pain and mobility and 3 points for peer relationships.15–19 basis of the ordinal nature of the PROMIS scores, the
Utilizing PROMIS data from children with ITW may be number of subjects in the study, and the distribution of
beneficial in understanding the patient perspective and the data, Spearman Rank statistics were compared with
priorities regarding this diagnosis. Our null hypothesis is the reference population based on a 1-sample t test. Cor-
that PROMIS metrics in children with ITW do not differ relations were performed to assess the relationship be-
significantly from those of healthy children. tween PROMIS scores and passive ankle range of motion
(ROM) and gait parameters of velocity.
METHODS
Study Design RESULTS
Western Institutional Review Board (WIRB) ap- A total of 45 patients (20 girls; 25 boys) met inclusion
proved this retrospective case series study. This study was criteria for this study (Table 1). PROMIS score averages for
performed in accordance with the 1964 Declaration of the entire cohort by domain were as follows: Mobility:
Helsinki and US HIPAA regulations. Chart review was 45.2 ± 8.2 (P = < 0.000), Peer Relationships: 46.4 ± 11.6
performed to identify children aged 5 to 17 who presented (P = 0.047), and Pain Interference: 47.4 ± 9.5 (P = 0.67)
to at a single tertiary care center between December 2017 (Table 2). All the 3 PROMIS domains measured
and December 2021. Inclusion criteria were a diagnosis of demonstrated MCID. The PROMIS scores did not differ
ITW and completion of a PROMIS questionnaire. Ex- significantly between younger (< 10) and older ( ≥ 10)
clusion criteria were neurologic diagnoses (tethered cord, children with ITW. Post-treatment PROMIS scores were
cerebral palsy, muscular dystrophy), autism, and previous obtained for 20 children an average of 4.8 ± 2.0 months
surgical treatment. Demographic, physical exam, treat- from presentation PROMIS scores, and these scores did not
ment, and available motion analysis data were collected via significantly differ.
retrospective chart review. Gait analysis data from an in- Seventy-three percent of PROMIS scores were pa-
strumented gait lab were available for 11 children. tient-reported while the remainder were parent-reported.
Retrospective review of PROMIS questionnaires
was performed. Children older than 8 years of age and TABLE 1. Demographic Characteristics of the Patient Cohort
cognitively able completed PROMIS questionnaires, (n = 45) Collected in the Survey
which included 3 domains: Pain Interference, Mobility,
Age, years [mean (SD)] 10.5 (3.1)
and Peer Relationships. If a child was unable to complete Age <10 (%) 49
PROMIS questionnaires because of age or cognitive Age ≥ 10 (%) 51
ability, parents completed questionnaires by proxy. These Sex, n (%)
PROMIS domains are collected as part of routine clinical Girls 20 (44)
care at our institution. The PROMIS questionnaire is Boys 25 (56)
Race, n (%)
normalized to an age-matched control cohort, with scores White 30 (67)
ranging from 0 to 100 with a mean of 50 and a standard Other 15 (33)
deviation of 10. High scores indicate better function for BMI, percentile [mean (SD)] 19.3 (4.7)
Mobility and Peer Relationships. Lower scores for Pain BMI indicates body mass index.
Interference indicate less pain interfering with function.

Copyright © 2022 Wolters Kluwer Health, Inc. All rights reserved. www.pedorthopaedics.com | e879

Copyright r 2022 Wolters Kluwer Health, Inc. All rights reserved.

Lindsay et al J Pediatr Orthop  Volume 42, Number 8, September 2022

TABLE 2. PROMIS Score Domains in Children With ITW TABLE 3. Gait Analysis Data for the Patient Cohort
PROMIS Scores for Entire Cohort (n = 45) Dorsiflexion with knee extended, degrees [mean (SD)] −9.4 (11.9)
Mobility [mean (SD)] 45.2 (8.2)*** Peak dorsiflexion in stance, degrees [mean (SD)] −10.9 (14.0)
Peer Relationships [mean (SD)] 46.4 (11.6)*** Percentage of age-matched normal velocity, % 91.5 (8.59)
Pain Interference [mean (SD)] 47.4 (9.5) [mean (SD)]
PROMIS scores, age <10 vs. ≥ 10 Percentage of age-matched cadence, % [mean (SD)] 95.7 (8.7)
< 10 (n = 22) ≥ 0 (n = 23) Percentage of age-matched stride length, % [mean (SD)] 92.4 (8.2)
Mobility [mean (SD)] 43.2 (7.9) 47.1 (8.2)
Peer Relationships [mean (SD)] 46.3 (10.2) 46.6 (13.0)
Pain Interference [mean (SD)] 47.8 (10.5) 46.9 (8.5)
PROMIS Scores, Patient Reported vs. Parent Reported ankle stiffness, resulting in a severely restricted ROM at the
Patient Reported Parent Reported ankle joint in many patients.24 Decreased dorsiflexion is
(n = 33) (n = 12) associated with increased ankle sprains and fractures in
Mobility [mean (SD)] 47.3 (7.9) 40.6 (6.9)
Peer Relationships [mean (SD)] 46.7 (12.2) 44.2 (9.5) children.25
Pain Interference [mean (SD)] 46.8 (9.0) 49.1 (11.3) Tendo-Achilles lengthening has been shown to ob-
PROMIS Scores, presentation vs. follow-up jectively improve gait kinematics, particularly dorsiflexion in
Presentation Follow-up stance, physical exam measurements, and radiographic
(n = 20) (n = 20)
Mobility [mean (SD)] 46.9 (8.3) 46.0 (10.5)
parameters.10–12 Serial casting, bracing, and stretching appear
Peer Relationships [mean (SD)] 49.5 (2.2) 50.3 (11.2) to offer less satisfying kinematic and physical exam meas-
Pain Interference [mean (SD)] 49.7 (8.3) 46.0 (9.2) urement improvements, though this is controversial.7,26,27
The PROMIS questionnaire is normalized to age-matched controls with scores
The study by Stott et al13 report persistent changes in gait
ranging from 0 to 100, with a mean of 50 and an SD of 10. kinematics in children treated with both casting and surgery,
***P < 0.05. but note that these changes are clinically insignificant and are
ITW indicates idiopathic toe walking; PROMIS, Patient-Reported Outcome
Measurement Information System.
not detectable visually.
Few papers include patient satisfaction as an out-
come metric. Stott et al13 report that all subjects in their
Parents (mean = 40.6, SD = 6.9) reported significantly cohort are satisfied with outcomes for either casting or
(t = 2.7, P = .012) lower mobility scores than patients surgery. Westberry et al11 report that 87% of subjects were
(mean = 47.3, SD = 7.9) did. There were no other sig- satisfied with outcomes after tendo-Achilles lengthening
nificant differences between patient-reported and parent- procedure. No studies thus far include more detailed PRO
reported PROMIS scores. metrics for ITW patients.
Passive mean dorsiflexion ROM at the ankle with A 2012 JAAOS review article cautions orthopaedic
the knee in extension for the entire cohort was −9.4 ± 11.9 surgeons against the overtreatment of ITW, noting that
degrees. For those children with instrumented motion most children who are untreated do not ultimately have
analysis, the peak dorsiflexion in stance was −10.9 ± 14.0 poor functional outcomes. The article notes, however, that
(normal peak dorsiflexion in stance is 10 degrees). The toe walking “may have a psychological effect on some
average velocity as a percentage of age-matched norms parents and children, which may justify treatment of this
was 91.5% ± 8.6% (Table 3). disorder.”1 The potential psychological effects of ITW are
The age-matched gait velocity was negatively cor- frequently discussed throughout the literature, though
related (rs = −0.74, P = 0.01) with Peer Relationships. No these effects have not been explicitly studied previously.
correlations were found between other aspects of gait, In our study, our cohort had MCIDs in Mobility,
body mass index, or limitations in dorsiflexion and any of Peer Relationships, and Pain Interference from the refer-
the PROMIS domains. ence population. Differences were statistically significant
for Mobility and Peer Relationships. The impairment in
Mobility and Peer Relationship scores also met the
DISCUSSION MCID. Pain Interference scores trended lower consistent
This study explored patient perspective of ITW and with less reported pain interference but the difference did
quality of life by utilizing PROMIS scores. In our cohort of not reach statistical significance.
45 otherwise healthy children without other neurologic di- The gait velocity, as measured by computerized
agnoses, we established that there are both clinically motion analysis, in this group of children with ITW was
(MCID) and statistically significant differences in PROMIS 91% of their age-matched standards. However, within this
means between our cohort and the healthy age-matched group, lower velocities correlated with higher Peer Rela-
population. To the best of our knowledge, this is the first tionship scores. This is intriguing given that the entire
study investigating quality of life metrics in this population. cohort had lower Peer Relationship scores. Anecdotally,
ITW literature frequently cites toe walking as a source children seem to walk slower when they attempt to nor-
of significant functional impairment, noting that ITW is malize their gait by getting their heels down, which de-
associated with gait impairments. Kinematic studies of creases the visible difference in their gait. Further study
children with ITW demonstrate disruptions to the gait cy- may elucidate whether the impairment of peer relation-
cle, even when patients are asked to attempt to walk in a ships stems more from social and cosmetic concerns than
typical heel-toe progression.22,23 ITW is associated with the ability to keep up.

e880 | www.pedorthopaedics.com Copyright © 2022 Wolters Kluwer Health, Inc. All rights reserved.

Copyright r 2022 Wolters Kluwer Health, Inc. All rights reserved.

J Pediatr Orthop  Volume 42, Number 8, September 2022 PROMIS Scores in Pediatric Idiopathic Toe Walkers

The PROMIS metrics may be useful in capturing 8. Williams CM, Gray K, Davies N, et al. Exploring health
some of the quality of life metrics that impact children professionals’ understanding of evidence-based treatment for idio-
pathic toe walking. Child Care Health Dev. 2020;46:310–319.
with ITW. PROs have not previously been utilized for 9. Caserta AJ, Pacey V, Fahey M, et al. Interventions for idiopathic toe
ITW patients. Understanding how ITW affects patient walking. Cochrane Database Syst Rev. 2019;10:10.
quality of life as reflected by these metrics may allow the 10. McMulkin ML, Baird GO, Caskey PM, et al. Comprehensive
providers to engage patients and their families in shared outcomes of surgically treated idiopathic toe walkers. J Pediatr
decision making in the setting of the clinical equipoise that Orthop. 2006;26:606–611.
11. Westberry DE, Carpenter AM, Brandt A, et al. Surgical outcomes
is associated with the management of ITW. Eventually, for severe idiopathic toe walkers. J Pediatr Orthop. 2021;41:
utilizing PROMIS scores pretreatment and post-treatment e116–e124.
may better inform efficacy of treatment options. These 12. Hemo Y, Macdessi SJ, Pierce RA, et al. Outcome of patients after
measurements may be helpful adjuncts to kinematic Achilles tendon lengthening for treatment of idiopathic toe walking.
measurements in accurately capturing patient outcomes. J Pediatr Orthop. 2006;26:336–340.
13. Stott NS, Walt SE, Lobb GA, et al. Treatment for idiopathic toe-
Our study has several limitations. First, we did not walking: results at skeletal maturity. J Pediatr Orthop. 2004;24:
use foot models for our patients, as they were not the 63–69.
standard for this patient population when the pilot data 14. Guyatt GH, Osoba D, Wu AW, et al. Methods to explain the clinical
were collected. We will strongly consider the foot model significance of health status measures. Mayo Clin Proc.
2002;77:371–383.
for future studies in this population. Second, we are lim- 15. Thissen D, Liu Y, Magnus B, et al. Estimating minimally important
ited by the small size of our cohort. Much of our original difference (MID) in PROMIS pediatric measures using the scale-
cohort was excluded because of the concomitant neuro- judgment method. Qual Life Res Int J Qual Life Asp Treat Care
logic diagnoses. Third, we are a tertiary referring pediatric Rehabil. 2016;25:13–23.
hospital, and therefore, many patients with mild or self- 16. Okoroafor UC, Gerull W, Wright M, et al. The impact of social
deprivation on pediatric PROMIS health scores after upper
resolving ITW are not screened by our clinics. We an- extremity fracture. J Hand Surg. 2018;43:897–902.
ticipate that these milder cases would likely also report no 17. Yost KJ, Eton DT, Garcia SF, et al. Minimally important
changes in PROMIS scores, but a larger cohort would differences were estimated for six Patient-Reported Outcomes
better represent the population. Measurement Information System-Cancer scales in advanced-stage
cancer patients. J Clin Epidemiol. 2011;64:507–516.
18. Matsumoto H, Hyman JE, Shah HH, et al. Validation of Pediatric
CONCLUSIONS Self-Report Patient-Reported Outcomes Measurement Information
ITW is associated with detriments to Mobility, and System (PROMIS) measures in different stages of Legg-Calvé-
Perthes disease. J Pediatr Orthop. 2020;40:235–240.
Peer Relationships as captured by PROMIS metrics. This 19. Khan AA, Abarca N, Cung NQ, et al. Use of PROMIS in
is the first study to evaluate PRO metrics in patients with assessment of children with Ponseti-treated idiopathic clubfoot:
ITW and helps better characterize why this population better scores with greater than 3 years of brace use. J Pediatr Orthop.
seeks medical intervention. Future work may be directed 2020;40:526–530.
at the efficacy of treatment options for children with ITW. 20. Rothrock NE, Amtmann D, Cook KF. Development and validation
of an interpretive guide for PROMIS scores. J Patient-Rep
Outcomes. 2020;4:16.
REFERENCES 21. Perry J, Burnfield JM. Gait Analysis: normal and pathological
1. Oetgen ME, Peden S. Idiopathic toe walking. J Am Acad Orthop function. J Sports Sci Med. 2010;9:353.
Surg. 2012;20:292–300. 22. Westberry DE, Davids JR, Davis RB, et al. Idiopathic toe walking:
2. Ruzbarsky JJ, Scher D, Dodwell E. Toe walking: causes, epidemiol- a kinematic and kinetic profile. J Pediatr Orthop. 2008;28:352–358.
ogy, assessment, and treatment. Curr Opin Pediatr. 2016;28:40–46. 23. Alvarez C, Vera De, Beauchamp M, et al. Classification of idiopathic
3. Engström P, Tedroff K. Idiopathic toe-walking: prevalence and toe walking based on gait analysis: development and application of
natural history from birth to ten years of age. J Bone Joint Surg Am. the ITW severity classification. Gait Posture. 2007;26:428–435.
2018;100:640–647. 24. Engelbert R, Gorter JW, Uiterwaal C, et al. Idiopathic toe-walking
4. Engström P, Tedroff K. The prevalence and course of idiopathic toe- in children, adolescents and young adults: a matter of local or
walking in 5-year-old children. Pediatrics. 2012;130:279–284. generalised stiffness? BMC Musculoskelet Disord. 2011;12:61.
5. Hirsch G, Wagner B. The natural history of idiopathic toe-walking: 25. Tabrizi P, McIntyre WM, Quesnel MB, et al. Limited dorsiflexion
a long-term follow-up of fourteen conservatively treated children. predisposes to injuries of the ankle in children. J Bone Joint Surg Br.
Acta Paediatr Oslo Nor 1992. 2004;93:196–199. 2000;82:1103–1106.
6. Eastwood DM, Menelaus MB, Dickens DR, et al. Idiopathic toe- 26. Thielemann F, Rockstroh G, Mehrholz J, et al. Serial ankle casts for
walking: does treatment alter the natural history? J Pediatr Orthop patients with idiopathic toe walking: effects on functional gait
Part B. 2000;9:47–49. parameters. J Child Orthop. 2019;13:147–154.
7. Davies K, Black A, Hunt M, et al. Long-term gait outcomes 27. Stricker SJ, Angulo JC. Idiopathic toe walking: a comparison of
following conservative management of idiopathic toe walking. Gait treatment methods. J Pediatr Orthop. 1998;18:289–293.
Posture. 2018;62:214–219.

Copyright © 2022 Wolters Kluwer Health, Inc. All rights reserved. www.pedorthopaedics.com | e881

Copyright r 2022 Wolters Kluwer Health, Inc. All rights reserved.