Abstract
We critically evaluated the evidence cited by the American Academy of Pediatrics (AAP) and the Royal College of Paediatrics and Child Health (RCPCH) supporting the diagnostic significance of fractures in suspected physical child abuse. Primary studies cited by the AAP Clinical Report and the RCPCH Systematic Review were reviewed with respect to study design, methods used to classify cases as abuse, and the potential for incorporation bias and outcome misclassification. Sixty-seven studies were identified, of which 35 reported measures of diagnostic utility. All were observational, predominantly retrospective studies. No study employed an independent gold standard for the diagnosis of abuse, and all relied on imperfect reference standards that varied in their methods of case classification. The evidence demonstrated substantial methodological limitations, primarily the risks of incorporation bias and circular reasoning.
Overall, the evidence underpinning current guidance on fractures in suspected child abuse is characterised by methodological flaws that limit the scientific confidence that can be placed in the reported diagnostic associations. The substantial uncertainty surrounding the diagnostic significance attributed to fractures in the evaluation of suspected child abuse should be explicitly recognised and communicated.
Keywords: Child abuse, Infant fractures, Evidence based medicine, Diagnostic uncertainty
1. Introduction
In pediatric medicine, certain skeletal injuries are widely regarded as important indicators of possible physical child abuse. These diagnostic associations have been incorporated into clinical guidance documents, including the American Academy of Pediatrics (AAP) Evaluating Young Children With Fractures for Child Abuse: Clinical Report [1] and the Royal College of Paediatrics and Child Health (RCPCH) Systematic Review on Fractures [2], both of which are widely used to inform the clinical evaluation of infants presenting with these injuries.
While these documents provide comprehensive literature reviews and synthesize evidence for clinical decision-making, they fail to critically evaluate the underlying methodological assumptions, the strength of the evidence supporting their conclusions, or the associated uncertainties.
In evidence-based medicine, the confidence placed in a diagnostic finding depends upon the quality of the underlying evidence and the associated uncertainty. For example, the diagnostic certainty associated with a screening mammogram differs fundamentally from that provided by histopathological confirmation of breast cancer. Accurate communication of diagnostic uncertainty is therefore an essential component of informed decision-making. The explicit communication of uncertainty is a fundamental principle of both classical [3] and modern [4] scientific practice.
The importance of evaluating the strength of evidence underpinning clinical guidance is illustrated by a recent analysis of AAP Clinical Practice Guidelines, which found that only 10.6% of 236 recommendations across 14 guidelines were supported by Level A (high-quality) evidence [5]. This analysis highlights the importance of understanding the methodological foundations of pediatric guidance documents and the strength of the evidence on which they rely.
In the present study, we critically review the evidence underpinning the AAP Clinical Report and RCPCH Systematic Review relating to the diagnostic significance of rib fractures, classic metaphyseal lesions (CMLs), and long-bone fractures in suspected physical child abuse. These fracture types are identified in the AAP Clinical Report as being “highly specific” for abuse.
Both the AAP and RCPCH have already undertaken comprehensive literature searches. Our objective is neither to replicate their search strategies nor to evaluate their study selection and inclusion criteria. Rather, we critically evaluate the underlying methodological characteristics and assumptions of their included studies, providing an appraisal of the evidence base underpinning clinical recommendations.
Particular attention is given to the methods used to classify cases as abuse. As previously acknowledged [6,7], there is no “gold standard” for the identification of physical abuse when evaluating pediatric fractures. Consequently, it is of fundamental importance to analyse the imperfect reference standards that are employed, and whether they are vulnerable to incorporation bias, circular reasoning, and outcome misclassification [8,9].
Other potential sources of uncertainty are also explored.
Our analysis provides a critical appraisal of the methodological foundations of the evidence base supporting the use of fractures in the diagnosis of suspected physical child abuse.
2. Methods
2.1. Literature selection
References were compiled from the American Academy of Pediatrics (AAP) Evaluating Young Children With Fractures for Child Abuse: Clinical Report [1] and the Royal College of Paediatrics and Child Health (RCPCH) Systematic Review on Fractures [2] concerning the association between rib fractures, classic metaphyseal lesions (CMLs), and long-bone fractures and suspected physical child abuse.
The AAP Clinical Report and RCPCH Systematic Review each undertook comprehensive searches of the literature using complementary methodologies, comprising a formal systematic review in the case of the RCPCH and an extensive expert-led literature review in the case of the AAP. Accordingly, the present study did not undertake a de novo systematic search of literature but instead critically evaluated the primary studies underpinning these influential guidance documents.
Articles from the AAP Clinical Report were included if they were cited within the Fractures With High Specificity for Abuse subsection of the Fracture Specificity for Abuse section. For the RCPCH Systematic Review, articles were selected from the Which fractures are indicative of abuse? section, specifically subsections 1.1–1.4 and 1.6.
Studies were included if they investigated the association between rib fractures, classic metaphyseal lesions, or long-bone fractures and abuse. Studies examining alternative causes of fractures (for example cardiopulmonary resuscitation) without comparison with abusive injury were excluded, as were studies concerned solely with fracture biomechanics. Where systematic reviews were cited by the guidelines, the original primary studies cited within those reviews were retrieved and evaluated individually. Duplicate publications were removed to generate a single dataset of unique studies.
Each included study was independently reviewed by two authors. Where differences arose, the studies were reviewed jointly and consensus was reached.
2.2. Study details
Data extracted included publication year, study design, fracture type, the reference standard used to classify cases as abuse or non-abuse, and whether the study provided measures such as specificity or positive predictive value.
2.3. Assessment of reference standards
As no definitive gold standard exists for determining whether pediatric fractures are attributable to abuse, studies were reviewed with particular attention to the reference standards used for case classification.
The reference standards employed were evaluated with respect to their transparency, reproducibility, independence from the index diagnosis, and susceptibility to incorporation bias, circular reasoning, and outcome misclassification. Particular attention was given to whether the case-classification process was explicitly described, whether fracture findings may have contributed to the determination of abuse, and whether sufficient methodological detail was provided to permit independent evaluation of the reported diagnostic associations.
3. Results
3.1. Included literature
The AAP Clinical Report cited 16 studies relating to rib fractures, 3 relating to classic metaphyseal lesions (CMLs), and 22 relating to long-bone fractures. The RCPCH Systematic Review cited 19 rib fracture studies, 14 CML studies, and 32 long-bone studies.
After removal of duplicate citations, 24 studies relating to rib fractures, 17 relating to CMLs, and 36 relating to long-bone fractures remained. Several studies investigated more than one fracture type and are therefore represented in multiple categories. Overall, 67 unique primary studies met the inclusion criteria. A complete list of included studies is provided as Supplementary Material.
Thirty-five of the 67 studies reported measures of diagnostic utility, including positive predictive values, likelihood ratios, or related diagnostic performance measures. Fourteen of the 35 reported such measures for rib fractures [[10], [11], [12], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23]], four for CMLs [22,[24], [25], [26]], eleven for humeral fractures [14,19,[21], [22], [23],[27], [28], [29], [30], [31], [32]], twenty for femoral fractures [14,[21], [22], [23], [24],27,30,[32], [33], [34], [35], [36], [37], [38], [39], [40], [41], [42], [43], [44]], with several studies reporting diagnostic measures for more than one fracture type.
3.2. Study types
All 67 included studies, including the 35 reporting measures of diagnostic utility, were observational in design, comprising predominantly retrospective cross-sectional studies, case-control studies, and small case series. This is consistent with what was found in previous reviews [6,7,45].
3.3. Reference standards
Substantial variation was observed in the methods used to classify cases as abuse across the included studies. Sixteen of the 67 studies (24%) did not report how cases were classified as abuse. Among the remaining 51 studies, multidisciplinary team (MDT) determination was the most frequently employed reference standard (34 studies, 67%), followed by predefined criteria (11 studies, 22%), author classification without predefined criteria (7 studies, 14%), court determinations (5 studies, 10%), and confessions or admissions (3 studies, 6%). Several studies employed more than one reference standard (Fig. 1).
Fig. 1.
Method for Categorisation as Abuse for all 67 included studies, Multidisciplinary Team determination (MDT, 15 studies), predefined criteria (10, 7 of which use the criteria of Thomas 1991), author classification (5), court determinations (2), diagnostic coding (2), confessions (2). Some studies use more than one reference standard.
Among the 35 studies reporting measures of diagnostic utility, five studies did not specify the method by which cases were classified. Reference standards comprised Multidisciplinary Team determination (MDT, 15 studies), predefined criteria (10, 7 of which use the criteria of Thomas 1991 [27]), author classification (5), court determinations (2), diagnostic coding (2), confessions or admissions (2) (Fig. 2).
Fig. 2.
Method for Categorisation as Abuse for 35 studies that provide measures of diagnostic utility, as for Fig. 1.
4. Discussion
This review critically evaluates the evidence underpinning the AAP Evaluating Young Children With Fractures for Child Abuse: Clinical Report and the RCPCH Systematic Review on Fractures relating to the diagnostic significance of rib fractures, classic metaphyseal lesions (CMLs), and long-bone fractures in suspected physical child abuse.
4.1. Reference standards
As there is no gold standard for the diagnosis of physical abuse in children presenting with fractures, all included studies relied upon imperfect methods of case classification. The validity of any diagnostic accuracy study depends fundamentally on the validity and independence of the reference standard used to define disease. Where that reference standard is imperfect, reported measures of diagnostic performance become susceptible to incorporation bias, circular reasoning, and outcome misclassification.
4.1.1. Studies without a reported reference standard
Sixteen of the 67 included studies did not report how cases were classified as abuse. The absence of a transparent classification method represents a fundamental methodological weakness because it precludes independent evaluation of the reference standard, prevents reproducibility, and renders meaningful critical appraisal impossible.
Although many of these studies were published several decades ago, their continued citation within contemporary guidance documents means that these methodological limitations remain relevant when considering the evidence base and its historic roots. Transparent reporting of the methods used to establish diagnostic outcomes is a fundamental requirement of diagnostic research and is particularly important where no independent reference standard exists.
4.1.2. Multidisciplinary team assessment
For studies in which a reference standard was reported, most (34/51) employed multidisciplinary team (MDT) determination for case classification. Although MDT assessment reflects contemporary clinical practice, its validity as a research reference standard depends upon the transparency and independence of the methods by which diagnostic conclusions are reached [46].
However, among the studies reviewed, the MDT process was generally described only superficially. Most studies (24/34) did not specify the criteria used to determine abuse (10 employed predefined criteria, discussed below). None described the reasoning underpinning the final classification of individual cases or disclosed whether the index fracture formed part of the information considered during MDT deliberations.
Given that rib fractures, classic metaphyseal lesions, and certain long-bone fractures have long been regarded as indicators of abuse, it is not unreasonable to conclude that the fracture findings themselves contributed to the MDT determination of abuse. Where independence of a reference standard from the index test has not been demonstrated, there is a high risk of circular reasoning and incorporation bias [47].
Unlike other domains of diagnostic medicine, there is no independent ground truth or clinical follow-up against which MDT determinations of abuse can subsequently be validated, and thus no feedback loop to correct clinical assessments. Consequently, the true diagnostic accuracy of these classifications remains unmeasurable, preventing systematic errors in the diagnostic process from being identified or mitigated. This creates an inherent risk of persistent misclassification.
Similar methodological concerns apply to studies in which abuse was classified solely by study authors without predefined criteria or detailed reporting of the underlying reasons for those decisions.
4.1.3. Criteria-based reference standards
Eleven of the 51 studies (22%) provided predetermined criteria for classification, most commonly those prescribed by Thomas et al. (1991), where “definite abuse” required: Serious trauma must have been involved, not reflected in history; Serious trauma must have been involved, with a suspicious delay in reporting; Multiple fractures found, not explained by history; or Healing fractures found, not explained by history. The criteria also specify “likely abuse” where the injury is severe and the history does not reflect sufficient severity.
This method of classification has significant shortcomings. Firstly, none of the predetermined criteria are themselves validated and all function as imperfect reference standards. Secondly, the criteria involve high levels of subjective judgement regarding what fractures are indicative of “serious trauma,” whether that trauma is “reflected in history” or a delay in reporting is “suspicious,” and whether a particular infant is susceptible due to medical factors.
Determining whether a history is sufficient and the findings explained not only requires subjective judgements related to the force required to cause particular findings, but also the force required to cause those findings in particular infants. This is inherently influenced by biological and medical variables and by judgements of their clinical implications. Prior beliefs regarding which fractures are attributed to abusive trauma inevitably inform these subjective judgements. Consequently, the index test and reference standard cannot be regarded as independent.
None of the studies using predetermined criteria provided information as to how cases met the criteria, which criteria were employed most commonly, whether the index test was known to those making assessments, and whether it was considered in assessing whether cases met the criteria.
The use of predetermined criteria in the included studies therefore carries serious risk of incorporation bias, circular reasoning, and outcome misclassification.
4.1.4. Judicial findings
Using court decisions or judicial findings to classify cases as abusive also creates a very serious risk of circular reasoning. Because convictions rely on expert opinion evidence that findings like metaphyseal or rib fractures are specific for abuse, selecting cohorts based on those legal outcomes ensures those findings are present.
4.1.5. Diagnostic codes
A similar problem occurs when diagnostic codes are used to classify cases as abusive; because of the belief that certain findings are diagnostic of abuse, cases with “abuse” codes will exhibit those findings, leading to very serious risk of circular reasoning and incorporation bias. Using diagnostic codes identifies which findings are regularly used in diagnosing abuse, but provides no independent verification that those diagnoses were accurate.
4.1.6. Confessions
Categorisation by confessions also fails to provide an independent reference standard for validating fracture etiology. In the included studies, confessions to causing fractures were obtained after a medical evaluation had identified findings believed to be diagnostic of abuse. This creates a selection bias: cases without such findings are not investigated, and caregivers are not subjected to questioning and pressures that lead to admissions. Using these confessions to validate the diagnostic significance of specific fractures therefore results in a very serious risk of incorporation bias, as the medical findings were the catalyst for the investigation and the subsequent statement.
Furthermore, the reliability of this classification method is fundamentally compromised by well-documented risk factors for false confessions [48]. These include coercive questioning by investigators already convinced of guilt based on medical assumptions, the prospect of plea bargains and/or hopes for better sentences, the hope that a confession will facilitate the return of the child or other siblings from foster care, and the pressure to end long-running, financially draining, and seemingly hopeless legal proceedings where the accused is totally outgunned in access to experts and resources [49].
Accordingly, confessions do not provide an independent scientific reference standard for validating the diagnostic significance attributed to fractures.
4.2. Alternative explanations
The limitations of the reference standards are not the only source of uncertainty within the evidence base. Fractures represent an indirect marker of possible abuse rather than a direct measure of abusive intent, and their interpretation is influenced by additional sources of uncertainty.
Fractures have a range of potential aetiologies, including accidental trauma, genetic disorders, endocrine disease, metabolic bone disease, nutritional deficiencies, and developmental factors. Differential diagnosis in an infant presenting with fractures requires evaluating each of these potential causes, all of which carry diagnostic challenges. Crucially, these factors do not operate in isolation; rather, they interact within individual children to produce a continuous distribution of skeletal fragility rather than a binary distinction between “normal” and “abnormal” bone strength.
Furthermore, the force required to cause fractures in infants is not well constrained [50], and assessing whether a reported mechanism of injury is biomechanically plausible depends on the mechanical properties of the individual child's skeleton. Because skeletal strength cannot be measured directly in vivo and must instead be inferred from clinical, radiological, and laboratory findings, each of which has inherent limitations [51], an individual child's specific susceptibility to fracture represents a significant additional layer of uncertainty when interpreting diagnostic significance in suspected physical abuse.
4.3. Radiology
Uncertainty regarding individual skeletal fragility is compounded by the inherent limitations of radiological assessment. Conventional radiography is relatively insensitive to reduced bone mineralisation, with radiographic osteopenia generally not becoming apparent until approximately 30–40% of bone mineral has been lost. Accordingly, the absence of radiographic evidence of osteopenia does not exclude clinically relevant skeletal fragility [52].
Furthermore, significant inter-observer variation exists among radiologists regarding the detection and characterization of infant rib fractures, with diagnostic agreement varying depending on the specific imaging protocol and projections utilized [53]. As has been observed previously, there is no gold standard for validating their diagnosis of “classic metaphyseal lesions” on initial imaging [54] and even when follow up imaging is available, the diagnosis is determined by consensus rather than typical fracture healing criteria. In addition, classic metaphyseal lesions can be difficult to distinguish from metaphyseal abnormalities associated with metabolic bone disease, particularly when disease is mild or partially treated [55].
Consequently, uncertainty in radiological diagnosis represents an additional source of uncertainty when evaluating the diagnostic significance attributed to fractures in suspected physical abuse.
5. Limitations
This review has some limitations. First, it was not a de novo systematic review of the literature. This approach was chosen because these guidance documents already undertook comprehensive literature reviews using complementary methodologies and represent the principal evidence base underpinning current clinical recommendations. We are not aware of any methodologically sound studies concerning the diagnostic significance of rib fractures, classic metaphyseal lesions, or long-bone fractures that were omitted from these reviews.
Second, it is recognised that ethical and practical constraints make prospective diagnostic studies of suspected child abuse difficult to undertake. However, these constraints do not diminish the importance of transparently acknowledging the methodological limitations of the available evidence when interpreting reported diagnostic associations.
Finally, the scope of this review was limited to evaluating the methodological characteristics of the evidence supporting the diagnosis of abuse. It did not examine the evidence relating to alternative medical causes of fractures, which represents an important but separate area of investigation.
6. Conclusions
The evidence underpinning the AAP Clinical Report and the RCPCH Systematic Review is characterised by critical methodological flaws. A substantial proportion of the cited studies did not report how cases were classified as abuse, while those that did relied upon imperfect reference standards that carried significant potential for incorporation bias, circular reasoning, and outcome misclassification.
These methodological limitations are further compounded by uncertainties surrounding alternative medical explanations for fractures, individual variation in skeletal fragility, and the inherent limitations of radiological assessment.
Although the AAP Clinical Report and the RCPCH Systematic Review provide influential guidance regarding the diagnostic significance of fractures in suspected child abuse, neither evaluates the quality or the methodological certainty of the evidence underpinning those recommendations. The substantial uncertainty surrounding the diagnostic significance attributed to fractures when evaluating suspected physical child abuse should be explicitly recognised and communicated.
CRediT authorship contribution statement
Chris Brook: Writing – original draft, Formal analysis, Conceptualization. Julie Mack: Writing – review & editing, Data curation, Conceptualization. Waney Squier: Writing – review & editing, Data curation, Conceptualization. Ankin Güvencel: Data curation, Conceptualization. Cyrille Rossant: Writing – review & editing, Conceptualization. Anders Eriksson: Writing – review & editing, Data curation, Conceptualization.
Role of funder/sponsor (if any)
N/A.
Clinical trial registration (if any)
N/A.
Funding/support
No funding was secured for this study.
Declaration of competing interest
Julie Mack, Waney Squier and Anders Eriksson have testified in court regarding alleged abuse cases. Cyrille Rossant is the co-founder and former President of a French non-profit organization (Adikia) providing access to support groups to families facing contested medical determinations of child abuse (unpaid activity). Other authors have no conflicts of interest to disclose.
Footnotes
Supplementary data to this article can be found online at https://doi.org/10.1016/j.fsisyn.2026.100740.
Appendix A. Supplementary data
The following is the Supplementary data to this article:
References
- 1.Haney Suzanne, Scherl Susan, DiMeglio Linda, Perez-Rossello Jeannette, Servaes Sabah, Merchant Nadia, the Council on Child Abuse and Neglect Section on orthopaedics, section on radiology, and section on endocrinology, and the society for pediatric radiology; evaluating young children with fractures for child abuse: clinical report. Pediatrics. February 2025;155(2) doi: 10.1542/peds.2024-070074. [DOI] [PubMed] [Google Scholar]
- 2.Royal College of Paediatrics and Child Health . RCPCH; London: 2020 Sep. Fractures: Systematic Review.https://childprotection.rcpch.ac.uk/child-protection-evidence/fractures-systematic-review/ Available from: [Google Scholar]
- 3.Keynes J.M. Macmillan & Co; 1921. A Treatise on Probability. [Google Scholar]
- 4.International Committee of Medical Journal . Recommendations for the Conduct, Reporting, Editing, and Publication of Scholarly Work in Medical Journals. ICMJE; 2024. https://www.icmje.org/icmje-recommendations.pdf Available at: [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Matheny Antommaria Armand H., Kelleher Matthew, Peterson Rachel J. Quality of evidence and strength of recommendations in American academy of pediatrics' guidelines. Pediatrics. April 2025;155(4) doi: 10.1542/peds.2024-067836. [DOI] [PubMed] [Google Scholar]
- 6.Mitchell I.C., Norat B.J., Auerbach M., Bressler C.J., Como J.J., Escobar M.A., Jr., Flynn-O'Brien K.T., Lindberg D.M., Nickoles T., Rosado N., Weeks K., Maguire S. Identifying maltreatment in infants and young children presenting with fractures: does age matter? Acad. Emerg. Med. 2021 Jan;28(1):5–18. doi: 10.1111/acem.14122. [DOI] [PubMed] [Google Scholar]
- 7.Maguire S., Cowley L., Mann M., Kemp A. What does the recent literature add to the identification and investigation of fractures in child abuse: an overview of review updates 2005–2013. Evid. Base Child Health. 2013;8:2044–2057. [Google Scholar]
- 8.Whiting P., Rutjes A.W., Reitsma J.B., Bossuyt P.M., Kleijnen J. Sources of variation and bias in studies of diagnostic accuracy: a systematic review. Ann. Intern. Med. 2004 Feb 3;140(3):189–202. doi: 10.7326/0003-4819-140-3-200402030-00010. [DOI] [PubMed] [Google Scholar]
- 9.Mower W.R. Evaluating bias and variability in diagnostic test reports. Ann. Emerg. Med. 1999 Jan;33(1):85–91. doi: 10.1016/s0196-0644(99)70422-1. PMID: 9867892. [DOI] [PubMed] [Google Scholar]
- 10.Thomas P.S. Rib fractures in infancy. Ann Radiol (Paris) 1977;20(1):115–122. [PubMed] [Google Scholar]
- 11.Schweich P., Fleisher G. Rib fractures in children. Pediatr. Emerg. Care. 1985;1(4):187–189. doi: 10.1097/00006565-198512000-00004. [DOI] [PubMed] [Google Scholar]
- 12.Bulloch B., et al. Cause and clinical characteristics of rib fractures in infants. Pediatrics. 2000;105 doi: 10.1542/peds.105.4.e48. [DOI] [PubMed] [Google Scholar]
- 13.Leventhal J.M., Larson I.A., Abdoo D., et al. Are abusive fractures in young children becoming less common? Changes over 24 years. Child Abuse Negl. 2007;31(3):311–322. doi: 10.1016/j.chiabu.2006.07.004. [DOI] [PubMed] [Google Scholar]
- 14.Leventhal J.M., Martin K.D., Asnes A.G. Incidence of fractures attributable to abuse in young hospitalized children: results from analysis of a United States database. Pediatrics. 2008;122(3):599–604. doi: 10.1542/peds.2007-1959. [DOI] [PubMed] [Google Scholar]
- 15.Darling S.E., Done S.L., Friedman S.D., et al. Frequency of intrathoracic injuries in children younger than 3 years with rib fractures. Pediatr. Radiol. 2014;44(10):1230–1236. doi: 10.1007/s00247-014-2988-y. [DOI] [PubMed] [Google Scholar]
- 16.Cosway B., Mathura N., Mott A., et al. Occult rib fractures: defining the cause. Child Abuse Rev. 2015;24(1):6–15. [Google Scholar]
- 17.Strouse P.J., Owings C.L. Fractures of the first rib in child abuse. Radiology. 1995;197(3):763–765. doi: 10.1148/radiology.197.3.7480753. [DOI] [PubMed] [Google Scholar]
- 18.Cadzow S.P., Armstrong K.L. Rib fractures in infants: red alert! the clinical features, investigations and child protection outcomes. J. Paediatr. Child Health. 2000;36(4):322–326. doi: 10.1046/j.1440-1754.2000.00515.x. [DOI] [PubMed] [Google Scholar]
- 19.Barsness K.A., Cha E.S., Bensard D.D., et al. The positive predictive value of rib fractures as an indicator of nonaccidental trauma in children. J. Trauma. 2003;54(6):1107–1110. doi: 10.1097/01.TA.0000068992.01030.A8. [DOI] [PubMed] [Google Scholar]
- 20.Garcia V.F., Gotschall C.S., Eichelberger M.R., et al. Rib fractures in children: a marker of severe trauma. J. Trauma. 1990;30(6):695–700. [PubMed] [Google Scholar]
- 21.Pandya N.K., Baldwin K., Wolfgruber H., et al. Child abuse and orthopaedic injury patterns: analysis at a level I pediatric trauma center. J. Pediatr. Orthop. 2009;29(6):618–625. doi: 10.1097/BPO.0b013e3181b2b3ee. [DOI] [PubMed] [Google Scholar]
- 22.Worlock P., Stower M., Barbor P. Patterns of fractures in accidental and non-accidental injury in children: a comparative study. Br. Med. J. 1986;293(6539):100–102. doi: 10.1136/bmj.293.6539.100. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Kowal-Vern A., Paxton T.P., Ros S.P., et al. Fractures in the Under-3-Year-Old age cohort. Clin. Pediatr. 1992;31(11):653–659. doi: 10.1177/000992289203101103. [DOI] [PubMed] [Google Scholar]
- 24.Anderson W.A. The significance of femoral fractures in children. Ann. Emerg. Med. 1982;11(4):174–177. doi: 10.1016/s0196-0644(82)80492-7. [DOI] [PubMed] [Google Scholar]
- 25.McClelland C.Q., Heiple K.G. Fractures in the first year of life: a diagnostic dilemma? Am. J. Dis. Child. 1982;136(1):26–29. doi: 10.1001/archpedi.1982.03970370028006. [DOI] [PubMed] [Google Scholar]
- 26.Thackeray J.D., Wannemacher J., Adler B.H., et al. The classic metaphyseal lesion and traumatic injury. Pediatr. Radiol. 2016;46:1128–1133. doi: 10.1007/s00247-016-3568-0. [DOI] [PubMed] [Google Scholar]
- 27.Thomas S.A., Rosenfield N.S., Leventhal J.M., et al. Long-bone fractures in young children: distinguishing accidental injuries from child abuse. Pediatrics. 1991;88(3):471–476. [PubMed] [Google Scholar]
- 28.Leventhal J.M., Thomas S.A., Rosenfield N.S., et al. Fractures in young children. Distinguishing child abuse from unintentional injuries. Am. J. Dis. Child. 1993;147(1):87–92. doi: 10.1001/archpedi.1993.02160250089028. [DOI] [PubMed] [Google Scholar]
- 29.Strait R.T., Siegel R.M., Shapiro R.A. Humeral fractures without obvious etiologies in children less than 3 years of age: when is it abuse? Pediatrics. 1995;96(4 Pt 1):667–671. [PubMed] [Google Scholar]
- 30.Rosenberg N., Bottenfield G. Fractures in infants: a sign of child abuse. Ann. Emerg. Med. 1982;11(4):178–180. doi: 10.1016/s0196-0644(82)80493-9. [DOI] [PubMed] [Google Scholar]
- 31.Shaw B.A., Murphy K.M., Shaw A., Janecek J., Kay A.J. Humerus shaft fractures in young children: accident or abuse? J. Pediatr. Orthop. 1997;17(3):293–297. [PubMed] [Google Scholar]
- 32.Mitchell P.D., Brown R., Wang T., et al. Multicentre study of physical abuse and limb fractures in young children in the East Anglia Region, UK. Arch. Dis. Child. Oct 2019;104(10):956–961. doi: 10.1136/archdischild-2018-315035. [DOI] [PubMed] [Google Scholar]
- 33.Beals R.K., Tufts E. Fractured femur in infancy: the role of child abuse. J. Pediatr. Orthop. 1983;3(5):583–586. doi: 10.1097/01241398-198311000-00004. [DOI] [PubMed] [Google Scholar]
- 34.Gross R.H., Stranger M. Causative factors responsible for femoral fractures in infants and young children. J. Pediatr. Orthop. 1983;3(3):341–343. doi: 10.1097/01241398-198307000-00012. [DOI] [PubMed] [Google Scholar]
- 35.Wellington P., Bennet G.C. Fractures of the femur in childhood. Injury. 1987;18(2):103–104. doi: 10.1016/0020-1383(87)90183-5. [DOI] [PubMed] [Google Scholar]
- 36.Dalton H.J., Slovis T., Heifer R.E., et al. Undiagnosed abuse in children younger than 3 years with femoral fracture. Am. J. Dis. Child. 1990;144(8):875–878. doi: 10.1001/archpedi.1990.02150320039022. [DOI] [PubMed] [Google Scholar]
- 37.Hinton R.Y., Lincoln A., Crockett M.M., Sponseller P., Smith G. Fractures of the femoral shaft in children: incidence, mechanisms, and sociodemographic risk factors. J. Bone Joint. Surg. Am. 1999;81:500–507. doi: 10.2106/00004623-199904000-00007. [DOI] [PubMed] [Google Scholar]
- 38.Schwend R.M., Werth C., Johnston A. Femur shaft fractures in toddlers and young children: rarely from child abuse. J. Pediatr. Orthop. 2000;20(4):475–481. [PubMed] [Google Scholar]
- 39.Coffey C., Haley K., Hayes J., et al. The risk of child abuse in infants and toddlers with lower extremity injuries. J. Pediatr. Surg. 2005;40(1):120–123. doi: 10.1016/j.jpedsurg.2004.09.003. [DOI] [PubMed] [Google Scholar]
- 40.Skellern C.Y., Wood D.O., Murphy A., et al. Non-accidental fractures in infants: risk of further abuse. J. Paediatr. Child Health. 2000;36(6):590–592. doi: 10.1046/j.1440-1754.2000.00592.x. [DOI] [PubMed] [Google Scholar]
- 41.Baldwin K., Pandya N.K., Wolfgruber H., et al. Femur fractures in the pediatric population: abuse or accidental trauma? Clin. Orthop. Relat. Res. 2011;469(3):798–804. doi: 10.1007/s11999-010-1339-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Hui C., Joughin E., Goldstein S., et al. Femoral fractures in children younger than three years: the role of nonaccidental injury. J. Pediatric Orthoped. 2008;28(3):297–302. doi: 10.1097/BPO.0b013e3181653bf9. [DOI] [PubMed] [Google Scholar]
- 43.Rex C., Kay P. Features of femoral fractures in nonaccidental injury. J. Pediatr. Orthop. 2000;20(3):411–413. [PubMed] [Google Scholar]
- 44.Murphy R., Kelly D.M., Moisan A., et al. Transverse fractures of the femoral shaft are a better predictor of nonaccidental trauma in young children than spiral fractures are. J. Bone Jt. Surg. Am. Vol. 2015;97(2):106–111. doi: 10.2106/JBJS.N.00664. [DOI] [PubMed] [Google Scholar]
- 45.Paine C.W., Fakeye O., Christian C.W., Wood J.N. Prevalence of abuse among young children with rib fractures: a systematic review. Pediatr. Emerg. Care. 2019 Feb;35(2):96–103. doi: 10.1097/PEC.0000000000000911. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Kellerhuis B.E., Jenniskens K., Kusters M.P.T., Schuit E., Hooft L., Moons K.G.M., Reitsma J.B. Expert panel as reference standard procedure in diagnostic accuracy studies: a systematic scoping review and methodological guidance. Diagn. Progn. Res. 2025 May 13;9(1):12. doi: 10.1186/s41512-025-00195-7. PMID: 40355966; PMCID: PMC12070646. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Whiting P.F., Rutjes A.W., Westwood M.E., et al. QUADAS-2: a revised tool for the quality assessment of diagnostic accuracy studies. Ann. Intern. Med. 2011;155(8):529–536. doi: 10.7326/0003-4819-155-8-201110180-00009. [DOI] [PubMed] [Google Scholar]
- 48.Kassin S.M., Gudjonsson G.H. The psychology of confessions: a review of the literature and issues. Psychol. Sci. Publ. Interest. 2004;5(2):33–67. doi: 10.1111/j.1529-1006.2004.00016.x. [DOI] [PubMed] [Google Scholar]
- 49.Rossant C., Brook C. Why admitted cases of AHT make a low quality reference standard: a survey of people accused of AHT in France. Forensic Sci. Int.: Synergy. 2022 Dec 29;6 doi: 10.1016/j.fsisyn.2022.100312. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50.Pike J., et al. Rib fractures in infancy: establishing the mechanisms of cause from the injuries—a literature review. Forensic Sci. Int. 2007;173(2–3):104–112. doi: 10.1258/rsmmsl.47.3.200. [DOI] [PubMed] [Google Scholar]
- 51.Soto Martinez M.E., Love J.C., Crowder C.M., Wiersema J.M., Pinto D.C., Derrick S.M., Gao S., Fleischman J.M., Greeley C., Donaruma-Kwoh M., Bachim A. The first step in an investigation of quantitative ultrasound as a technique for evaluating infant bone strength. J. Forensic Sci. 2021;66:456–469. doi: 10.1111/1556-4029.14605. [DOI] [PubMed] [Google Scholar]
- 52.Expert Panel on Musculoskeletal Imaging. Yu J.S., Krishna N.G., Fox M.G., Blankenbaker D.G., Frick M.A., Jawetz S.T., Li G., Reitman C., Said N., Stensby J.D., Subhas N., Tulchinsky M., Walker E.A., Beaman F.D. ACR appropriateness criteria® osteoporosis and bone mineral density: 2022 update. J. Am. Coll. Radiol. 2022 Nov;19(11S):S417–S432. doi: 10.1016/j.jacr.2022.09.007. [DOI] [PubMed] [Google Scholar]
- 53.Marine M.B., Corea D., Steenburg S.D., Wanner M., Eckert G.J., Jennings S.G., Karmazyn B. Is the new ACR-SPR practice guideline for addition of oblique views of the ribs to the skeletal survey for child abuse justified? AJR Am. J. Roentgenol. 2014 doi: 10.2214/AJR.13.11068. [DOI] [PubMed] [Google Scholar]
- 54.Karmazyn B., Miller E.M., Lay S.E., Massey J.M., Wanner M.R., Marine M.B., Jennings S.G., Ouyang F., Hibbard R.A. Double-read of skeletal surveys in suspected non-accidental trauma: what we learned. Pediatr. Radiol. 2017 May;47(5):584–589. doi: 10.1007/s00247-017-3783-3. [DOI] [PubMed] [Google Scholar]
- 55.Kleinman P.K. Problems in the diagnosis of metaphyseal fractures. Pediatr. Radiol. 2008 Jun;38(Suppl 3):S388–S394. doi: 10.1007/s00247-008-0845-6. PMID: 18470447. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.