HipScreen: a valid mobile app to measure hip migration in children with cerebral palsy in the community setting - PMC Skip to main content An official website of the United States government Here's how you know Here's how you know Official websites use .gov A .gov website belongs to an official government organization in the United States. Secure .gov websites use HTTPS A lock ( Lock Locked padlock icon ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites. Search Log in Dashboard Publications Account settings Log out Search… Search NCBI Primary site navigation Search Logged in as: Dashboard Publications Account settings Log in Search PMC Full-Text Archive Search in PMC Journal List User Guide PERMALINK Copy As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsement of, or agreement with, the contents by NLM or the National Institutes of Health. Learn more: PMC Disclaimer | PMC Copyright Notice Bone Jt Open . 2026 Apr 15;7(4):540–548. doi: 10.1302/2633-1462.74.BJO-2025-0322.R1 Search in PMC Search in PubMed View in NLM Catalog Add to search HipScreen: a valid mobile app to measure hip migration in children with cerebral palsy in the community setting Akib M Khan Akib M Khan , MBChB, MRCS, PGDipMedEd (Cambridge), SICOT (Dip), FEBOT, FRCS (Tr&Orth) 1 Department of Paediatric Orthopaedic Surgery, Chelsea and Westminster Hospital NHS Foundation Trust, London, UK Consultant Orthopaedic Surgeon Find articles by Akib M Khan 1, ✉ , John Amen John Amen , MD 2 Division of Paediatric Orthopaedics, Department of Orthopaedic Surgery, Ain Shams University, Cairo, Egypt Consultant Orthopaedic Surgeon Find articles by John Amen 2 , Oliver Perkins Oliver Perkins , MBChB, BSc, MRCS 3 Department of Paediatric Orthopaedic Surgery, Evelina London Children’s Hospital, Guy’s and St Thomas’ NHS Foundation Trust, London, UK Orthopaedic Registrar Find articles by Oliver Perkins 3 , Konstantinos Kafchitsas Konstantinos Kafchitsas , MD, PhD 4 Department of Orthopaedics and Traumatology, University of Patras, Patras, Greece Consultant Orthopaedic Surgeon Find articles by Konstantinos Kafchitsas 4 , Stephen J Cooke Stephen J Cooke , MB ChB, BSc 5 Department of Paediatric Orthopaedic Surgery, University Hospitals Coventry and Warwickshire NHS Trust, Coventry, UK Consultant Orthopaedic Surgeon Find articles by Stephen J Cooke 5 , Michail Kokkinakis Michail Kokkinakis , PhD, PGCert (Distinction) 6 Department of Paediatric Orthopaedic Surgery, Evelina London Children’s Hospital, Guy’s and St Thomas’ NHS Foundation Trust, London, UK 7 King’s College London, London, UK Consultant Orthopaedic Surgeon, Honorary Reader (Associate Professor) Find articles by Michail Kokkinakis 6, 7 ; HipScreen Study collaborators Author information Article notes Copyright and License information 1 Department of Paediatric Orthopaedic Surgery, Chelsea and Westminster Hospital NHS Foundation Trust, London, UK 2 Division of Paediatric Orthopaedics, Department of Orthopaedic Surgery, Ain Shams University, Cairo, Egypt 3 Department of Paediatric Orthopaedic Surgery, Evelina London Children’s Hospital, Guy’s and St Thomas’ NHS Foundation Trust, London, UK 4 Department of Orthopaedics and Traumatology, University of Patras, Patras, Greece 5 Department of Paediatric Orthopaedic Surgery, University Hospitals Coventry and Warwickshire NHS Trust, Coventry, UK 6 Department of Paediatric Orthopaedic Surgery, Evelina London Children’s Hospital, Guy’s and St Thomas’ NHS Foundation Trust, London, UK 7 King’s College London, London, UK 1 University Hospitals Coventry & Warwickshire NHS Trust, Coventry, UK 2 Royal Wolverhampton NHS Trust, Wolverhampton, UK 3 Staffordshire and Stoke On Trent NHS Partnership Trust, Stoke On Trent, UK 4 Physio CYP, Birmingham, UK 5 University Hospitals Coventry & Warwickshire NHS Trust, Coventry, UK 6 Evelina London Children’s Hospital, London, UK 7 Queen Mary’s Hospital, Sidcup, London, UK 8 Rainbow Centre, East Kent University Foundation Trust, London, UK ✉ Correspondence should be sent to Akib M. Khan. E-mail: [email protected] The authors declare no competing interests relevant to this work. Roles Akib M Khan : MBChB, MRCS, PGDipMedEd (Cambridge), SICOT (Dip), FEBOT, FRCS (Tr&Orth) , Consultant Orthopaedic Surgeon John Amen : MD , Consultant Orthopaedic Surgeon Oliver Perkins : MBChB, BSc, MRCS , Orthopaedic Registrar Konstantinos Kafchitsas : MD, PhD , Consultant Orthopaedic Surgeon Stephen J Cooke : MB ChB, BSc , Consultant Orthopaedic Surgeon Michail Kokkinakis : PhD, PGCert (Distinction) , Consultant Orthopaedic Surgeon, Honorary Reader (Associate Professor) Collection date 2026 Apr. © 2026 Khan et al. This article is distributed under the terms of the Creative Commons Attributions (CC BY 4.0) licence ( https://creativecommons.org/licenses/by/4.0/ ), which permits unrestricted use, distribution, and reproduction in any medium or format, provided the original author and source are credited. PMC Copyright notice PMCID: PMC13080576 PMID: 41983275 Abstract Aims The Reimers migration percentage (MP) is the gold-standard radiological parameter used to aid decision-making regarding surgical management of hip displacement in cerebral palsy (CP). Accurate measurement is important to risk stratify patients and allow timely onward referral from community teams to paediatric orthopaedic services. We performed a study to determine whether experts and novices could use a free smartphone app (HipScreen (HS) app) as a valid method for measuring MP in CP. Methods Using the HS app, two groups measured MP in 20 pelvis radiographs (40 hips) at weeks 0 and 2. The ‘inexperienced’ group included four community physiotherapists with no CP experience. The ‘experienced’ group included four community physiotherapists and two paediatricians with CP experience. All participants watched the tutorial videos on the HS app website. HS app measurements were then compared with gold-standard picture archiving and communication system (PACS) measurements conducted by a senior paediatric orthopaedic surgeon. Modified Pearson correlation (r) was used to determine inter-rater reliability between HS app and PACS measurements. Intraclass correlation coefficient (ICC) was then used to assess intrarater reliability. The mean absolute deviation (MAD) was calculated to compare raters with the gold standard. Results HS app measurements in the experienced and inexperienced groups showed highly significant correlation with the gold-standard measurements (p < 0.001). There were no significant differences between intra- or inter-group measurements at weeks 0 and 2 with r > 0.86 and p < 0.001. Both inter- and intra-rater reliability were excellent with ICC > 0.9. There was no statistically significant difference in the MAD within individual measurements and compared with the gold standard. Conclusion The HS app is accurate in measuring MP in CP when used by non-specialists and specialists. Non-specialists do not require additional supplementary training. These findings suggest app efficacy and safety in regional and national hip surveillance programmes for children with CP. Cite this article: Bone Jt Open 2026;7(4):540–548. Keywords: Cerebral palsy, CP, Hip migration, Hip dislocation, hips, radiographs, intraclass correlation coefficient (ICC), physiotherapists, paediatric orthopaedic, Hip displacement, paediatric orthopaedic surgeon, paediatricians, Pearson correlation coefficient Introduction Cerebral palsy (CP) describes a constellation of conditions related to a non-progressive insult to the developing brain causing disordered movement or posture. Hip displacement is a common musculoskeletal problem in CP, more prevalent in nonambulant children with higher levels of neurological and functional impairment, with 90.5% of those with a Gross Motor Function Classification System (GMFCS) V affected. 1 The natural history of hip dysplasia in CP has been studied extensively, and the disease-specific validated Child Health Index of Life with Disabilities (CPCHILD) shows a significant reduction in health-related quality of life in children with CP who have dislocated hips. 3 Early identification of children who are ‘at risk’ allows a ‘proactive’ rather than ‘reactive’ approach to hip displacement, with earlier intervention associated with improved CPCHILD scores. 4 Therefore, evidence-based validated surveillance tools for hip dysplasia are important in improving care for CP patients. Several hip surveillance programmes have been developed which use radiological assessments to identify ‘at risk’ hips. 5 , 6 Hip surveillance programmes rely on clinical and radiological assessments at appropriate intervals and frequencies as determined by the relative risk of hip displacement per GMFCS level. The migration percentage (MP) is the gold-standard radiological measure of hip displacement, which has been extensively studied as a valid and reliable quantitative measure. 7 , 8 Prerequisites for effective use of MP in hip surveillance include the use of radiographs of acceptable quality interpreted by qualified professionals and means for onward referral at set thresholds. Referral thresholds are not standardized, however MP thresholds between > 30% and 40% usually warrant referral to orthopaedic services. 9 MP measurements are conducted by clinicians who are trained to do so, including orthopaedic surgeons, specialist physiotherapists, and/or radiologists. 7 Access to this group is variable depending on local service provisions, with the potential for delays in interpretation and referral. In many healthcare settings, there is limited access to these professionals, which has created the need for validated tools to screen patients with this condition. The HipScreen (HS) app (Shriner’s Hospital for Children, USA) has been validated for use in hip surveillance of patients with hip subluxation in CP. 10 It is a free resource available for download on different smartphone models. The HS app allows measurement of MP using the classic method (lateral border of the acetabulum) rather than the modified method. 11 The app divides the ossific nucleus into ten partitions at 10% intervals of the entire width of the ossific nucleus ( Figure 1 ). Our previous study has examined its accuracy, reliability, and discriminatory ability in measuring migration percentage among healthcare workers with experience in treating patients with CP. 12 We aimed to investigate whether the use of the HS app could be used by a broader range of community-based healthcare professionals. We also aimed to assess whether additional training was necessary for novice users to increase accuracy of measurements. Fig. 1. Open in a new tab a) Home screen of the HipScreen (HS) app with a red arrow pointing to the ruler function. b) Ruler with the red line indicating 30% migration percentage. c) Rotate function to correct pelvic obliquity; in this radiograph, counter clockwise (CCW) is needed (red curved arrow). Permission obtained from original authors. 12 Methods Ethical approval for this study was sought and deemed unnecessary by the Institutional Review Board at the Evelina London Children’s Hospital. Sample size was calculated using PASS v. 15 (NCSS, USA) with the type-1 error (α) rate of 0.05 and a power of 80%. This determined that a total of 35 hips was required. We anticipated a dropout rate of 10% and therefore deemed the use of 40 hips for analysis to be adequate. We identified 100 consecutive standardized pelvic radiographs from our CP surveillance programme and selected every fifth radiograph for inclusion. Our CP surveillance programme database included children of any age, CP type, and functional GMFCS level. All of our hip radiographs were obtained under the supervision of a senior radiographer with no forward or backward pelvic tilt, and with femora in neutral position so as to reduce the risk of measurement error. 13 , 14 We excluded any with a fully contained hip, completely dislocated hip, or previous hip surgery on either one or both sides. In total, 20 radiographs (40 hips) were used in this study. Of note, as the same methodology was used, these were the same standardized and anonymized pelvic radiographs from our previous study. 12 Gold-standard measurements for each radiograph were conducted and set using picture archiving and communication system (PACS) measurements by the senior author (MK) ( Table I ). A separate paediatric orthopaedic consultant (SJC) also measured the MP using both PACS and HS app to prove the validity of the software and confirm gold-standard measurements. We believe our methodology of using two experienced surgeons in setting the gold-standard measurements is sound, and the use of PACS matches globally accepted practice. 14 Table I. Gold-standard measurements for each radiograph conducted by the senior author. Patient number Right Left 1 40 35 2 40 30 3 30 35 4 65 30 5 50 20 6 30 60 7 30 35 8 30 40 9 40 25 10 35 30 11 55 55 12 35 30 13 40 65 14 45 50 15 30 30 16 40 35 17 30 30 18 40 45 19 30 25 20 20 45 Open in a new tab The anonymized radiographs were inserted into two PowerPoint presentations (PPT1 and PPT2; Microsoft, USA) in different random orders. Two groups of professionals were identified from two geographically diverse areas of the UK. The first ‘inexperienced’ group had no experience in managing children with CP while the other ‘experienced’ group were professionals who routinely treat CP patients. The inexperienced group was composed of four community physiotherapists and the experienced group of four community physiotherapists and two community paediatricians. All participants were directed to complete the free online training resources provided by the HS app developers on their website. 15 After each participant confirmed they had completed the online training, they were asked to complete MP measurements on PPT1. Following the first set of measurements, the inexperienced group were then given individual training sessions with the senior author (MK). All participants were asked to complete MP measurements on PPT2 two weeks later. All measurements were rounded to the nearest 5% which is in-keeping with the training material provided by the app developers. 16 Statistical analysis Measurements from all participants were transcribed into table format and statistical analysis was performed using SPSS v. 27 (IBM, USA). The interobserver correlation was determined using the modified Pearson’s correlational coefficient (r), a reliability measurement between -1 (reverse reliability) and + 1 (perfect reliability). Measurements < 0.7 are considered unreliable, 0.7 to 0.8 moderately reliable, > 0.8 good reliability, and > 0.9 highly significant reliability. Significance calculations were also performed to determine a probability assessment of measurements. Intra-rater correlation was determined using the intraclass correlation coefficient (ICC). A score below 0.5 indicates poor reliability, 0.5 to 0.75 moderate reliability, 0.75 to 0.90 good reliability, and > 0.90 excellent reliability. 17 Interobserver correlation in the experienced and inexperienced groups was calculated. Initially a normality test using the Kolmogorov-Smirnof test was conducted. In cases of normality, a paired t -test was performed and in case of not normal distributed data, a Wilcoxon-signed rank test was performed. A p-value < 0.05 was deemed statistically significant. In case of repeated comparisons (multiple null hypotheses with same data) the Holm-Bonferoni correction was also performed considering that this would otherwise increase the type-II error rate. The mean absolute deviation (MAD) was calculated from the mean absolute difference taken between raters at week 0 and week 2. It was also calculated between the gold-standard measurement and the raters’ measurements at both week 0 and week 2. The SD for each dataset was also calculated. Results Interclass comparison All HS app measurements showed high correlation with the gold-standard PACS measurements, with the Pearson coefficient r showing high correlation (> 0.9) in each measurement at both timepoints in the experienced ( Table II ) and inexperienced ( Table III ) groups. Further analysis of inter-rater reliability showed no single statistically significant measurement compared with the gold standard by any participant in either group at either timepoint ( Table IV and Table V ). There were only two statistically significant differences between experienced participants (experienced rater 5 vs rater 1 and 3 at the two-week measurements). However, all three raters were not statistically significant compared with the gold standard, and this finding is not thought to represent any clinical importance. Table II. Validity assessment demonstrating interobserver correlation in measurements by the experienced group at week 0 and week 2, and comparison with the gold-standard picture archiving and communication system measurements. r Wk 0 Exp1 Exp2 Exp3 Exp4 Exp5 Exp6 GS PACS Exp1 1 0.934 0.918 0.930 0.910 0.890 0.917 Exp2 1 0.951 0.974 0.968 0.947 0.964 Exp3 1 0.953 0.941 0.896 0.926 Exp4 1 0.972 0.943 0.946 Exp5 1 0.948 0.948 Exp6 1 0.931 r Wk 2 Exp1 1 0.975 0.962 0.955 0.941 0.932 0.959 Exp2 1 0.966 0.971 0.948 0.923 0.968 Exp3 1 0.954 0.915 0.908 0.953 Exp4 1 0.939 0.910 0.955 Exp5 1 0.902 0.933 Exp6 1 0.915 Open in a new tab Exp, experienced practitioner; GS PACS, gold-standard picture archiving and communication system; r, Pearson’s correlation coefficient. Table III. Validity assessment demonstrating interobserver correlation in measurements by the inexperienced group at week 0 and week 2, and comparison with the gold-standard picture archiving and communication system measurements. r Wk 0 InExp1 InExp2 InExp3 InExp4 GS PACS InExp1 1 0.93 0.856 0.924 0.960 InExp2 1 0.866 0.958 0.930 InExp3 1 0.891 0.860 InExp4 1 0.965 r Wk 2 InExp1 1 0.950 0.956 0.961 0.970 InExp2 1 0.926 0.941 0.946 InExp3 1 0.928 0.940 InExp4 1 0.950 Open in a new tab GS PACS, gold-standard picture archiving and communication system; InExp, inexperienced practitioner; r, Pearson’s correlation coefficient. Table IV. Significance testing of interobserver correlation in measuring experienced group at week 0 and week 2, including comparison with the gold-standard picture archiving and communication system measurements. p-value Wk 0 Exp1 Exp2 Exp3 Exp4 Exp5 Exp6 GS PACS Exp1 0.170 0.443 0.200 0.064 0.160 0.266 Exp2 0.636 0.992 0.105 0.725 0.549 Exp3 0.650 0.280 0.536 0.805 Exp4 0.523 0.805 0.716 Exp5 0.696 0.716 Exp6 0.361 p-value Wk 2 Exp1 0.513 0.942 0.578 0.025 0.305 0.866 Exp2 0.505 0.910 0.105 0.725 0.548 Exp3 0.566 0.027 0.281 0.925 Exp4 0.074 0.621 0.611 Exp5 0.201 0.300 Exp6 0.334 Open in a new tab p-values calculated using Wilcoxon-signed rank test. Exp, experienced practitioner; GS PACS, gold-standard picture archiving and communication system. Table V. Significance testing of interobserver correlation in measuring inexperienced group at week 0 and week 2, including comparison with the gold-standard picture archiving and communication system measurements. p-value Wk 0 InExp1 InExp2 InExp3 InExp4 GS PACS InExp1 0.511 0.154 0.595 0.613 InExp2 0.385 0.880 0.821 InExp3 0.284 0.316 InExp4 0.855 p-value Wk 2 InExp1 0.151 0.138 0.831 0.847 InExp2 0.950 0.098 0.124 InExp3 0.097 0.142 InExp4 0.854 Open in a new tab p-values calculated using Wilcoxon-signed rank test. GS PACS, gold-standard picture archiving and communication system; InExp, inexperienced practitioner; p, p-value; Wk, Week. Analysis of the MAD using a paired t -test demonstrated no statistically significant difference between the inexperienced and experienced raters (p = 0.399, 95% CI -0.326 to 0.758). Overall, these findings validate the interclass reliability of the HS app in accurately determining MP in CP in inexperienced and experienced healthcare workers over time. Intraclass comparison ICC showed excellent correlation (ICC > 0.9) in both groups ( Table VI and Table VII ). The MAD showed no statistically significant difference between week zero calculations and the gold standard (p = 0.525, 95% CI -0.227 to 0.420) and week 2 calculations and the gold standard (p = 0.626, 95% CI -0.170 to 0.271) ( Table VIII ). Table VI. Intraclass correlation between week 0 and week 2 in the experienced group. ICC Exp1 Exp2 Exp3 Exp4 Exp5 Exp6 Exp1 0.972 Exp2 0.986 Exp3 0.971 Exp4 0.985 Exp5 0.969 Exp6 0.954 Open in a new tab Exp, experienced practitioner; ICC, intraclass correlation coefficient. Table VII. Intraclass correlation between week 0 and week 2 in the inexperienced group. ICC InExp1 InExp2 InExp3 InExp4 InExp1 0.960 InExp2 0.968 InExp3 0.909 InExp4 0.982 Open in a new tab ICC, intraclass correlation coefficient; InExp, inexperienced practitioner. Table VIII. Mean absolute deviation between measurements taken at week 0 and week 2 as compared with the gold-standard measurements. Group member and measurement MAD (%) SD (%) Exp1 Week 0 vs Week 2 2.525 2.341 Week 0 vs gold standard 2.969 2.438 Week 2 vs gold standard 3.000 1.432 Exp2 Week 0 vs Week 2 3.019 1.475 Week 0 vs gold standard 2.844 2.183 Week 2 vs gold standard 2.625 0.791 Exp3 Week 0 vs Week 2 3.281 1.631 Week 0 vs gold standard 2.625 0.791 Week 2 vs gold standard 2.719 1.631 Exp4 Week 0 vs Week 2 1.750 0.975 Week 0 vs gold standard 2.125 0.813 Week 2 vs gold standard 2.303 1.144 Exp5 Week 0 vs Week 2 3.000 2.181 Week 0 vs gold standard 2.625 1.312 Week 2 vs gold standard 2.881 2.705 Exp6 Week 0 vs Week 2 1.938 1.497 Week 0 vs gold standard 2.500 1.013 Week 2 vs gold standard 2.500 1.013 InExp1 Week 0 vs Week 2 2.291 1.439 Week 0 vs gold standard 2.469 1.630 Week 2 vs gold standard 2.548 1.548 InExp2 Week 0 vs Week 2 2.366 2.531 Week 0 vs gold standard 2.805 2.710 Week 2 vs gold standard 2.080 1.339 InExp3 Week 0 vs Week 2 1.883 1.098 Week 0 vs gold standard 2.025 1.154 Week 2 vs gold standard 3.125 1.550 InExp4 Week 0 vs Week 2 2.938 4.008 Week 0 vs gold standard 2.774 4.016 Week 2 vs gold standard 2.731 1.086 Open in a new tab Exp, experienced practitioner; InExp, inexperienced practitioner; MAD, mean absolute deviation. Effect of training for the inexperienced group Further statistical testing was performed assessing changes in scores between week 0 and week 2 for the inexperienced raters ( Table IX ). This revealed improvements in scores for two of the four participants following the in-person training. However, it should be noted that all scores in week 0 were already significantly correlated with the gold standard, and this improvement is unlikely to be of any clinical significance. Table IX. Comparison of week 0 and week 2 scores for inexperienced raters. Rater p-value * InExp1 0.055 InExp2 0.001 InExp3 0.009 InExp4 0.429 Open in a new tab * Wilcoxon-signed rank test. InExp, inexperienced practitioner. Discussion CP hip surveillance programmes first began in Sweden in 1994 with the Uppföljningsprogram för cerebral pares (CPUP), which demonstrated a clear reduction in hip displacement through repeated, standardized assessments. 18 Subsequent programmes have achieved similar results including the Cerebral Palsy Integrated Pathway Scotland (CPIPS) and the Cerebral Palsy Integrated Pathway (CPIP) in England and Wales. 1 The current challenges faced by hip surveillance programmes are multiple and summarized well in Hughes et al’s 19 recent work. The HS app may play a role in addressing some of these, including reducing resource requirements (i.e. more inexperienced users can accurately determine MP), consistent reliability of measurements, and standardizing the expanded use of technology. Additionally, the ability to use this tool in the community allows improved equity of access to care across geographically diverse regions. Different protocols exist for timing and adequacy of radiographs, experience of those reporting the imaging and subsequent referral pathways. Standardized and successful completion of each of these steps should be the goal with a view to reducing errors and improving equitable care. Unfortunately, the standard of service received by children can vary significantly even within well-established programmes in developed countries. For example, in England, regional health provision is determined by integrated care boards (ICBs) that commission care services for children. According to the NHS England Commissioning framework for children and young people with CP, as of 2025, “8 out of 42 ICBs do not have a provider overseeing the care for people with cerebral palsy.” 20 Another example of variability in regional services can be seen in Canada, where reporting of MP by radiologists is as low as 20% in paediatric hospitals and 3% in community hospital settings, despite being used routinely by the surveillance programme coordinators. 21 From a practical perspective, variation can also exist due to local and regional factors including appropriate transportation for heavily disabled children, location of clinical assessment, imaging facility provisions, access to tertiary care, and funding. From a financial perspective, the use of a validated app in the community can also be cost-effective for the health service by reducing transportation needs. Case studies have shown that the average yearly cost for families to attend tertiary clinic appointments is up to £3,500. 20 Reducing this financial burden for families also minimizes time away from work and education for children and their families. Currently, most radiographs are interpreted manually by end-users (such as physiotherapists or surgeons). However, even when manual measurements are conducted, differences can exist between clinicians. Faraj et al 22 conducted a study with two clinicians measuring MP at time 0 and six weeks later. They found significant intrameasurer and intrasessional errors of up to 22.4% (with an absolute difference between successive measurements range of 0% to 23%). Inter-measurer absolute difference was as high as 26.5% which may be considered clinically unacceptable. It may be postulated that with more clinical experience, the accuracy of MP measurements may improve. However, Analan et al 23 found that the seniority of clinicians had no bearing on the accuracy of measurements. Of note, their study involved the use of PACS systems and psychiatrists conducting the measurements. They found that both the novice and experienced measurers were highly accurate in comparison with the gold standard. Our study demonstrated a similar finding, in that both inexperienced and experienced users were as accurate in their measurements. Another question is whether measurers in different health professional roles (i.e. physiotherapist, paediatrician, surgeon) have significant differences in measurement results. Our findings have demonstrated no difference between the HS app measurements by health professionals in any of these roles. We have shown that the HS app is a standardized, user-friendly tool which, following simple online training, enables users to make accurate and repeatable measurements of MP regardless of user experience. Previous studies have shown the reliability of using technology in reducing variation and inconsistencies in interpreting MP calculations. Kulkarni et al 10 (developers of the HS app) studied a diverse group of users analyzing the MP in 40 hips over two timepoints, and found the HS app to be accurate with a sensitivity of 94% and specificity of 66% with good reliability between timepoints (ICC = 0.83). These findings mirror our own, which showed accuracy with all users demonstrating no significant difference from the gold-standard measurements (p < 0.001) and excellent reliability between timepoints (ICC > 0.9). The previous arm of this study investigated the role of healthcare professionals in a hospital setting using the HS app in measuring MP in CP. 12 It involved training of each member of the healthcare team prior to conducting their first round of measurements. The study found good inter- and intrarater reliability in measurements across different medical and allied health specialties. Our current study adds to this, in that it examines the use of the HS app in a community setting across a wide geographical area, and with no didactic training prior to the first round of measurements, in both experienced and inexperienced healthcare practitioners. The only educational material mandated was the freely available resources on the HS app website. Our study found that the initial measurements at week 0 were accurate in both the experienced and inexperienced groups and, although specific training by a senior clinician improved accuracy, this was not thought to be clinically significant, suggesting that the existing training materials are effective. The use of AI will bring the ability to automate analysis of CP hip radiographs through machine-learning (ML). An example of a system that shows excellent promise is BoneFinder (The University of Manchester, UK). 24 The developers of the system used 1,650 pelvic radiographs in children with CP to compare BoneFinder with manual measurements made by five experts. They determined a MAD of 5.7% (SD 8.5%) with an excellent ICC of 0.91 for MP. This finding was mirrored by work conducted by Hughes et al, 25 who used BoneFinder to interrogate 1,018 CP hips from a national database. They showed good ICC (0.80 (95% CI 0.65 to 0.87)) between its results and CPIPS measurements (entered manually by end-users). 25 The reliability of the system was poorer when image artefact (such as a groin shield), aberrant anatomy, or radiographs with inadequate visualization of the lesser trochanter. Another ML model (SVM Model, Turkey) developed by Ertan Birsel et al 26 also demonstrated excellent ICC (> 0.95) and an accuracy of > 90%. Of note, they excluded any radiograph that was not deemed to meet strict standardization criteria, which may explain the excellent results when compared with BoneFinder. While these newer ML methods hold promise, the role of manual measurements still exists, particularly in those who do not have access to newer technology, or in cases where artefact or anatomical variations are present. Although we commend the HS app as a reliable tool for the diagnosis and accurate measurement of MP in children with CP, there are several limitations to our study. The HS app requires rounding of measurements to the closest 5%, which leads to results that may be of lower accuracy than standard PACS measurements (where increments of < 1% may be measured). However, our findings show that HS app measurements were comparable with the gold-standard PACS measurements as well as being reliable within the internal measurements between timepoints. This suggests that the 5% incremental cutoff is not of significance compared with using PACS measurements. In fact, Reimers 27 reported a standard error of 10% for MP based on an estimate of errors that occur on the line to the nearest millimetre. Similarly, the Melbourne group also found rate of variability in expert measurements on a single radiograph to 5.8% of the true value. 8 A pragmatic limitation of our study is the requirement of users to own a smartphone that can host the HS app. A large proportion of the clinical workforce in our healthcare system have smartphones, and therefore this consideration may be more significant in resource-poor settings. However, regardless of healthcare setting, there are also questions regarding data protection and ensuring clinicians are compliant with local standards when using a third-party application. We used screenshot images of PACS radiographs on PowerPoint slides, which is a relatively low fidelity method for displaying digital images. Although accessing higher-resolution images may be possible in hospital settings, this may not be the case in community care, and therefore lower-fidelity displays may in fact be more realistic. Additionally, the users did not trial the HS app in non-digitally image-based settings, therefore it is unknown how the HS app would perform with acetate films. Another assumption was that, although the inexperienced group of physiotherapists were not specifically trained to manage CP in their daily practice, they may have prior experience during their qualification. In conclusion, accurate measurement of the MP is important in ensuring an effective CP Hip surveillance programme. Our study has shown that the use of the free HS app is reliable, accurate, and safe in the community setting with both allied health practitioners who have and have not previously had experience in treating children with CP. Those who have not had experience treating CP children had improvement of scores when individual expert training was provided, however we found that this is unlikely to be clinically significant, given excellent initial scores achieved (in-keeping with the gold standard). The efficacy of the free online training material in obtaining accurate readings is sustained over multiple weeks. Take home message - This study validated the free HipScreen app as an accurate method for measuring Reimers migration percentage in children with cerebral palsy (CP), comparing both experienced and inexperienced community practitioners against gold-standard measurements by a senior paediatric orthopaedic surgeon. - The app showed excellent agreement with gold-standard measurements, with strong inter- and intrarater reliability and no significant differences between specialists and non-specialists, even without extra training. These findings support the app’s safe use in regional and national CP hip surveillance pathways, helping community teams identify hip displacement earlier and streamline timely referral to paediatric orthopaedic services. Author contributions A. M. Khan: Conceptualization, Data curation, Methodology, Project administration, Writing – original draft, Writing – review & editing J. Amen: Methodology, Supervision, Writing – review & editing O. Perkins: Data curation, Methodology, Resources K. Kafchitsas: Data curation, Formal analysis, Methodology, Supervision, Validation, Writing – review & editing S. J. Cooke: Data curation, Methodology, Supervision, Validation, Writing – review & editing M. Kokkinakis: Conceptualization, Methodology, Supervision, Validation, Writing – review & editing Funding statement The author(s) received no financial or material support for the research, authorship, and/or publication of this article. ICMJE COI statement The authors declare no competing interests relevant to this work. Data sharing The data that support the findings for this study are available to other researchers from the corresponding author upon reasonable request. Acknowledgements The authors would like to extend our gratitude to the study participants and the senior radiographers, including Superintendent Radiographer Lucy Clough, at the Evelina London Children's Hospital. HipScreen Study collaborators: • Laura Ball, University Hospitals Coventry & Warwickshire NHS Trust, Coventry, UK. • Andre King and Sophie Billingham, Royal Wolverhampton NHS Trust, Wolverhampton, UK. • Joel Bowpitt, Staffordshire and Stoke On Trent NHS Partnership Trust, Stoke On Trent, UK. • Sunita Dhindsa, University Hospitals Coventry & Warwickshire NHS Trust, Coventry, UK. Anita Patel, Lucy Clough, Claire Wicks, and Jill Cadwgan, Evelina London Children’s Hospital, London, UK. • Ola Pieno, Beckie Corps, and Rachel Heayberd, Queen Mary’s Hospital, Sidcup, London, UK. • Suzi Davenport, Rainbow Centre, East Kent University Foundation Trust, London, UK. Open access funding The open access fee for this article was self-funded. © 2026 Khan et al. This article is distributed under the terms of the Creative Commons Attributions (CC BY 4.0) licence ( https://creativecommons.org/licenses/by/4.0/ ), which permits unrestricted use, distribution, and reproduction in any medium or format, provided the original author and source are credited. Contributor Information Akib M. Khan, Email: [email protected]. Collaborators: Study collaborators, Laura Ball , Andre King , Sophie Billingham , Joel Bowpitt , Alexandra Webster , Sunita Dhindsa , Anita Patel , Lucy Clough , Ola Pieno , Beckie Corps , Rachel Heayberd , Suzi Davenport , Claire Wicks , and Jill Cadwgan Data Availability The data that support the findings for this study are available to other researchers from the corresponding author upon reasonable request. References 1. Wordie SJ, Robb JE, Hägglund G, Bugler KE, Gaston MS. Hip displacement and dislocation in a total population of children with cerebral palsy in Scotland.
Bone Joint J. 2020;102-B(3):383–387. doi: 10.1302/0301-620X.102B3.BJJ-2019-1203.R1. [ DOI ] [ PubMed ] [ Google Scholar ] 2. Palisano R, Rosenbaum P, Walter S, Russell D, Wood E, Galuppi B. Development and reliability of a system to classify gross motor function in children with cerebral palsy.
Dev Med Child Neurol. 1997;39(4):214–223. doi: 10.1111/j.1469-8749.1997.tb07414.x. [ DOI ] [ PubMed ] [ Google Scholar ] 3. Ramstad K, Jahnsen RB, Terjesen T. Severe hip displacement reduces health-related quality of life in children with cerebral palsy.
Acta Orthop. 2017;88(2):205–210. doi: 10.1080/17453674.2016.1262685. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 4. DiFazio R, Vessey JA, Miller P, Van Nostrand K, Snyder B. Postoperative complications after hip surgery in patients with cerebral palsy: a retrospective matched cohort study.
J Pediatr Orthop. 2016;36(1):56–62. doi: 10.1097/BPO.0000000000000404. [ DOI ] [ PubMed ] [ Google Scholar ] 5. Hägglund G, Andersson S, Düppe H, Lauge-Pedersen H, Nordmark E, Westbom L. Prevention of dislocation of the hip in children with cerebral palsy. The first ten years of a population-based prevention programme.
J Bone Joint Surg Br. 2005;87-B(1):95–101. [ PubMed ] [ Google Scholar ] 6. Howard JJ, Willoughby K, Thomason P, Shore BJ, Graham K, Rutz E. Hip surveillance and management of hip displacement in children with cerebral palsy: clinical and ethical dilemmas.
J Clin Med. 2023;12(4):1651. doi: 10.3390/jcm12041651. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 7. Kulkarni VA, Davids JR, Boyles AD, Cung NQ, Bagley A. Reliability and efficiency of three methods of calculating migration percentage on radiographs for hip surveillance in children with cerebral palsy.
J Child Orthop. 2018;12(2):145–151. doi: 10.1302/1863-2548.12.170189. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 8. Parrott J, Boyd RN, Dobson F, et al. Hip displacement in spastic cerebral palsy: repeatability of radiologic measurement.
J Pediatr Orthop. 2002;22(5):660–667. [ PubMed ] [ Google Scholar ] 9. Shore BJ, Shrader MW, Narayanan U, Miller F, Graham HK, Mulpuri K. Hip surveillance for children with cerebral palsy: a survey of the POSNA membership.
J Pediatr Orthop. 2017;37(7):e409–e414. doi: 10.1097/BPO.0000000000001050. [ DOI ] [ PubMed ] [ Google Scholar ] 10. Kulkarni VA, Cung-Shahlaie NQ, Bagley AM, Yang NT, Taylor SL, Davids JR. HipScreen mobile app for the measurement of hip migration percentage in children with cerebral palsy: accuracy, reliability, and discriminatory ability.
Dev Med Child Neurol. 2023;65(11):1486–1492. doi: 10.1111/dmcn.15612. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 11. Wek C, Chowdhury P, Smith C, Kokkinakis M. Is the Gothic Arch a reliable radiographic landmark for migration percentage in children with cerebral palsy?
J Child Orthop. 2020;14(5):397–404. doi: 10.1302/1863-2548.14.200008. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 12. Amen J, Perkins O, Cadwgan J, Cooke SJ, Kafchitsas K, Kokkinakis M. Measuring hip migration percentage in cerebral palsy using the HipScreen app.
Bone Jt Open. 2023;4(5):363–369. doi: 10.1302/2633-1462.45.BJO-2023-0031.R1. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 13. Cliffe L, Sharkey D, Charlesworth G, Minford J, Elliott S, Morton RE. Correct positioning for hip radiographs allows reliable measurement of hip displacement in cerebral palsy.
Dev Med Child Neurol. 2011;53(6):549–552. doi: 10.1111/j.1469-8749.2011.03970.x. [ DOI ] [ PubMed ] [ Google Scholar ] 14. Wynter M, Gibson N, Kentish M, Love S, Thomason P, Kerr Graham H. The consensus statement on hip surveillance for children with cerebral palsy: Australian Standards of Care.
J Pediatr Rehabil Med. 2011;4(3):183–195. doi: 10.3233/PRM-2011-0174. [ DOI ] [ PubMed ] [ Google Scholar ] 15. Kulkarni V.HipScreen: Hip surveillance at your fingertips. Shriners Hospital for Children. [17 March 2026]. https://www.hipscreen.org date last. accessed. 16. No authors listed Using the HipScreen Ruler: Comprehensive Overview. hipscreen.org. 2018. [7 April 2026]. https://www.hipscreen.org/_files/ugd/1b38da_7c81cc1db6f74850a30e78e6df13f07f.pdf date last. accessed. 17. Koo TK, Li MY. A guideline of selecting and reporting intraclass correlation coefficients for reliability research.
J Chiropr Med. 2016;15(2):155–163. doi: 10.1016/j.jcm.2016.02.012. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 18. Hägglund G, Alriksson-Schmidt A, Himmelmann K, Rodby Bousquet E, Åstrand P. Significant medical improvements with CPUP - a combined follow-up program and national quality registry for cerebral palsy.
Lakartidn. 2023;120:23136. [ PubMed ] [ Google Scholar ] 19. Hughes K, Cooke SJ, Perry DC, Gaston MS. Hip surveillance in children with cerebral palsy in the UK: history, challenges, and future directions.
Bone Joint J. 2025;107-B(7):673–681. doi: 10.1302/0301-620X.107B7.BJJ-2025-0167.R1. [ DOI ] [ PubMed ] [ Google Scholar ] 20. NHS England Commissioning framework for children and young people with cerebral palsy. 2025. [7 April 2026]. https://www.england.nhs.uk/long-read/commissioning-framework-for-children-and-young-people-with-cerebral-palsy/ date last. accessed. [ DOI ] [ PubMed ] 21. Miller SD, Coates J, Bone JN, Farr J, Mulpuri K. A review of radiology reports from hip surveillance radiographs for children with cerebral palsy.
J Pediatr Orthop. 2022;42(7):e742–e746. doi: 10.1097/BPO.0000000000002183. [ DOI ] [ PubMed ] [ Google Scholar ] 22. Faraj S, Atherton WG, Stott NS. Inter- and intra-measurer error in the measurement of Reimers’ hip migration percentage.
J Bone Joint Surg Br. 2004;86-B(3):434–437. doi: 10.1302/0301-620x.86b3.14094. [ DOI ] [ PubMed ] [ Google Scholar ] 23. Analan PD, Yilmaz EE, Adam M, Leblebici B. The effect of physician experience on the measurement reliability of the Reimers’ hip migration percentage in children with cerebral palsy.
J Phys Ther Sci. 2015;27(10):3255–3258. doi: 10.1589/jpts.27.3255. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 24. Thompson P, Khattak M, Joseph PJ, et al. Automating radiological measurements of the hip in children with cerebral palsy.
Bone Joint J. 2025;107-B(1):124–132. doi: 10.1302/0301-620X.107B1.BJJ-2024-0894. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 25. Hughes K, Luzar J, Lang J, Perry DC, Gaston MS. Fully automated measurement of paediatric cerebral palsy pelvic radiographs with BoneFinder: external validation using a national surveillance database.
Bone Joint J. 2025;107-B(7):752–760. doi: 10.1302/0301-620X.107B7.BJJ-2024-1575.R1. [ DOI ] [ PubMed ] [ Google Scholar ] 26. Ertan Birsel S, Demirci E, Seker A, Usta Ayanoğlu KY, Oncu E, Ciftci F. Machine learning-assisted classification of hip conditions in pediatric cerebral palsy patients using migration percentage measurements.
Bone Rep. 2025;25:101852. doi: 10.1016/j.bonr.2025.101852. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 27. Reimers J. The stability of the hip in children. A radiological study of the results of muscle surgery in cerebral palsy.
Acta Orthop Scand Suppl. 1980;184:1–100. doi: 10.3109/ort.1980.51.suppl-184.01. [ DOI ] [ PubMed ] [ Google Scholar ] Associated Data This section collects any data citations, data availability statements, or supplementary materials included in this article. Data Availability Statement The data that support the findings for this study are available to other researchers from the corresponding author upon reasonable request. Articles from Bone & Joint Open are provided here courtesy of British Editorial Society of Bone and Joint Surgery ACTIONS View on publisher site PDF (760.3 KB) Cite Collections Permalink PERMALINK Copy RESOURCES Similar articles Cited by other articles Links to NCBI Databases Cite Copy Download .nbib .nbib Format: AMA APA MLA NLM Add to Collections Create a new collection Add to an existing collection Name your collection * Choose a collection Unable to load your collection due to an error Please try again Add Cancel Follow NCBI NCBI on X (formerly known as Twitter) NCBI on Facebook NCBI on LinkedIn NCBI on GitHub NCBI RSS feed Connect with NLM NLM on X (formerly known as Twitter) NLM on Facebook NLM on YouTube National Library of Medicine 8600 Rockville Pike Bethesda, MD 20894 Web Policies FOIA HHS Vulnerability Disclosure Help Accessibility Careers NLM NIH HHS USA.gov Back to Top