BACKGROUND. Child abuse and neglect are leading public health problems with significant morbidity and mortality. Previous studies indicate that physicians often lack knowledge and confidence in addressing child abuse and neglect.
OBJECTIVES. Our goal was to assess the child abuse and neglect curricula in pediatric residency programs as reported by chief residents and to identify levels of preparedness of residents to address child abuse and neglect on graduation. We analyzed variables related to preparedness.
METHODS. A 28-item survey was sent to chief residents of all 203 Accreditation Council for Graduate Medical Education–accredited pediatric residency programs in the United States from 2004–2005. We performed descriptive, bivariable, and multivariable analyses.
RESULTS. The response rate was 71%. Most programs taught didactics on physical and sexual abuse, but only 54% included domestic violence. Ninety-three percent of respondents rated their didactics as useful or very useful. Forty-one percent of programs required mandatory clinical rotations in child abuse and neglect, 57% offered elective rotations, and 25% offered no rotations at all. Respondents rated the levels of preparedness of graduating residents to address child abuse and neglect as: very well (12%), well (54%), somewhat well (28%), or not well (6%). Preparedness was significantly associated with didactic usefulness, number of hours of didactics, total number of inpatient cases of child abuse and neglect seen, percent of residents completing mandatory rotations, number of sexual abuse cases during mandatory rotation, number of physical abuse cases during mandatory rotation, and length of mandatory rotation.
CONCLUSIONS. Mandatory clinical experiences in child abuse and neglect improve the preparedness of graduating residents to identify and evaluate patients for child abuse and neglect. Perhaps residency training in child abuse and neglect should be a required subspecialty rotation with more explicit curricular content than in the current mandates.
Child abuse and neglect is a widespread public health problem with significant morbidity and mortality.1 According to the Third National Incidence Study of Child Abuse and Neglect (the most recent national incidence study), 1553800 children in the United States were abused or neglected in 1993, resulting in demonstrable harm.1 Child abuse and neglect is mandated by all states to be reported, with physicians usually defined as mandatory reporters.2 The reports for child abuse and neglect from physicians represent an important source for investigation.1 However, previous studies indicate that physicians often lack knowledge, training, and confidence in identifying and treating child abuse and neglect,3–7 perhaps rendering their ability to report less effective.
One primary way for physicians to gain knowledge and experience in the diagnosis and management of child abuse and neglect is through pediatric residency training. The Pediatric Residency Review Committee of the Accreditation Council for Graduate Medical Education sets the program requirements for accreditation in pediatric residency education.8 Currently, there is no specific required training in child abuse and neglect as a “required subspecialty experience.”8 Child abuse and neglect is mentioned as a topic to be addressed in “Emergency and Acute Illness Experience” and “Developmental/Behavioral Pediatrics”8 but lacks any further descriptive information.
Several studies have assessed the knowledge, attitude, and beliefs of residents regarding child abuse and neglect. The knowledge base of residents was found to be significantly lacking and in need of improvement.3,9 Specifically, this knowledge base would benefit from an increase in training opportunities.3,4,10,11 Faculty and residents perceive a lack of adequate child abuse and neglect training in residency.4,6,12 Although there have been recent increases in recognition of child abuse as a medical problem, there is no correlation of improved knowledge among younger physicians.5 This suggests that residency training has not addressed new developments in the field of child abuse, which may partially explain the similarity of knowledge among younger and older physicians.5
Our study describes the current state of pediatric residency training in child abuse and neglect, including specific components of didactic and clinical training. We also studied the level of preparedness of residents to identify and evaluate child abuse and neglect on graduation and analyzed the variables that led to improved preparedness. We hypothesized that mandatory training and exposure to larger numbers of maltreated patients would be associated with better preparedness.
A 28-item survey was mailed to the chief residents of all 203 Accreditation Council for Graduate Medical Education–accredited pediatric residency training programs in the United States, including Puerto Rico, during 2004–2005. Respondents answered questions about child abuse training in their residency programs. Each program was assigned a numerical identifier. The assignment list of the numerical identifier corresponding to the name of the actual program was kept confidential. Responders and nonresponders were compared for the comparison data using the numerical identifiers only. Responses were returned and compiled for analysis.
Questions were asked regarding the demographics of the residency program, descriptors and effectiveness of didactic training, descriptors and effectiveness of the types of clinical training, and the level of preparedness of residents to address child abuse and neglect on graduation (very well prepared, well prepared, somewhat well prepared, or not well prepared). In addition, respondents were asked which improvements were needed in child abuse and neglect training in their program. They were asked to select 1 or more of the following responses to answer this question: time scheduled for didactic training, quality of didactic training, number of patients seen, expertise of the child abuse and neglect providers, addition of other components to the training, no improvements needed, or “other.” The complete survey is available (from A.P.N.).
Descriptive, bivariable, and multivariable analyses were performed by using SAS 9.113 and Microsoft Excel.14 Bivariable analyses were performed by using χ2, Cochran-Mantel-Haenszel (CMH) χ2 for ordered categorical variables as indicated in the text. Variables achieving statistical significance (P < .05) on the CMH analysis were entered into a forward-selection logistic-regression model for multivariable analysis using a .2 significance level for entry into the model. Predictors were treated as ordinal variables.
In addition to the analysis described above, the level of preparedness was dichotomized from 4 levels of preparedness (very well, well, somewhat, and not well prepared) to 2 levels of preparedness (“well prepared” by combining very well and well and “not well prepared” by combining somewhat well and not well) for demonstrative purposes. These dichotomized variables were analyzed by using χ2 to assess the relationship between level of preparedness and associated variables.
To assess the generalized applicability of our study, we compared nonresponders to responders. We extracted the following data on the nonresponders and responders from the Fellowship and Residency Electronic Interactive Database Web site of the American Medical Association15: type of program, number of pediatric residents, whether programs include the educational benefit of “training in identifying and reporting of domestic violence/abuse,” and city in which the program is located.16 These variables were analyzed by using the χ2 statistic.
Significance for all statistical tests was set at P < .05. The study was approved for exemption by the Duke University Institutional Review Board Chair.
Of the 203 residency programs, 145 (71%) completed the survey. Seventy percent of the responding programs were self-described as academic, 26% as community, and 4% as military. Thirty-eight states and territories were represented. The city populations in which the residency programs were located were <250000 (21%), 250000 to 500000 (21%), 500000 to 1000000 (25%), and >1000000 (33%). The total numbers of pediatric residents in the responding programs were <15 (8%), 15 to 45 (55%), and >45 (37%).
Nonresponders and responders were analyzed to assess the generalized applicability of the study. There were no statistically significant differences between the responders and nonresponders in regard to type of program (χ2 = 8.5; P = .13), number of pediatric residents (χ2 = 2.2; P = .33), and the size of the city in which the program was located (χ2 = 4.04; P = .26). It is notable that 81% of the responders and 89% of the nonresponders answered a Fellowship and Residency Electronic Interactive Database survey question affirmatively that their programs included the educational benefit of “training in identifying and reporting of domestic violence/abuse”15 with no statistically significant difference between the 2 groups (χ2 = 1.56; P = .21).
Description of Didactic Training
The numbers of didactic hours of training per year were none (2%), 1 to 2 hours (16%), 3 to 6 hours (52%), 7 to 10 hours (14%), and >10 hours (16%). The levels of reported usefulness of didactic training were somewhat useful (7%), useful (50%), and very useful (43%). Nearly all programs taught about physical and sexual abuse in their didactic training, but only 54% of the programs taught about domestic violence (Fig 1). The percentages of residents attending at least half of the didactic training offered per year were <25% (1.4%), 25%–50% (3.6%), 50%–75% (46.4%), and 75%–100% (48.6%).
Description of Child Abuse and Neglect Educators
The didactic sessions were taught by physician experts in child abuse and neglect (87%), general pediatricians without special expertise in child abuse and neglect (31%), emergency medicine physicians (30%), midlevel practitioners in child abuse and neglect (12%), clinical social workers (28%), and sexual assault nurse examiners (17%). Respondents were asked whether they had an “easily identifiable child abuse expert” in their facility, with 84% answering in the affirmative. Note that no specific definition for “expert” was provided. Rather, the question asked for chief residents’ perception of identifiable experts in child abuse and neglect. Child abuse and neglect experts evaluated patients as inpatients in 84% of the programs and outpatients in 81% of the programs.
Description of Clinical Training
The types of child abuse and neglect rotations offered were mandatory (18%), elective (34%), both mandatory and elective (23%), and no rotations (25%). Thus, 59% have no mandatory rotation, and 43% have no elective rotation. The length of rotations offered varied according to the type of rotation, mandatory or elective, with elective rotations providing the longest length of rotation (Fig 2).
Elective rotations provided significantly more sexual abuse and physical abuse cases, although fewer residents completed elective rotations compared with mandatory rotations (Table 1). Inpatient child abuse and neglect consults and court experiences were more likely to be available on elective rotations compared with mandatory rotations (Table 1). Note that of all the programs studied (N = 145), only 35% had 100% completion rates; thus, approximately one third of all programs had all of their residents complete a rotation (Table 1). However, looking at just programs that reported that they offered mandatory rotations (N = 60), 85% of those programs had 100% of their residents completing a rotation. Thus, as would be expected, programs reporting that they offered mandatory rotations have higher rates of 100% completion.
The numbers of inpatient child abuse and neglect cases seen by the end of residency through any pediatric rotation were 0 (1.4%), <5 (26.4%), 5 to 15 (57.2%), and >15 (14.3%). Resources for teaching about child abuse and neglect included goals and objectives (74%), selected reading list (45%), CD-ROMs or slides (62%), textbook library (63%), and radiology review (59%). The following opportunities were additional sources of child abuse and neglect education other than child abuse and neglect rotations: grand rounds (77%), subspecialty rotations (ophthalmology, orthopedics, radiology) (58%), developmental/behavioral rotation (50%), general pediatrics clinics (81%), and PICUs (4%).
Preparedness of Residents to Identify and Evaluate Child Abuse and Neglect on Graduation
Programs rated the level of preparedness of their residents to identify and evaluate child abuse and neglect on graduation and were as follows: very well prepared (12%), well prepared (54%), somewhat well prepared (28%), and not well prepared (6%). In bivariable CMH analyses, the level of preparedness was significantly associated with didactic usefulness (P < .001), number of hours of didactic training (P < .001), number of inpatient cases of child abuse and neglect seen by graduation (P < .001), percent of residents completing mandatory rotations (P < .01), number of sexual abuse cases during mandatory rotation (P = .02), number of physical abuse cases during mandatory rotation (P < .01), and length of mandatory rotation (P = .02). Preparedness was not significantly associated with elective clinical experiences or the number of residents in the training program.
The levels of preparedness were dichotomized for illustrative purposes and were significantly associated with the number of inpatients seen by graduation, number of didactic hours, and didactic usefulness (Table 2). The dichotomized levels of preparedness were not significantly associated with any other variable.
Multivariable logistic regression was performed to analyze the variables associated with the dichotomized levels of preparedness. This regression model of the probability of being well prepared demonstrated that level of preparedness was associated with usefulness of didactics and (marginally) number of inpatients seen (Table 3). Effects not included in the final regression model (P > .2) were number of sexual abuse cases seen in mandatory rotation (P = .65), percent of residents completing mandatory rotation (P = .49), number of physical abuse cases seen in mandatory rotation (P = .68), and length of mandatory rotation (P = .48).
Areas of Improvement in Child Abuse and Neglect Training
Respondents noted the following areas needing improvement in their program’s child abuse and neglect training: time scheduled for training (52%), number of child abuse and neglect patients seen (41%), addition of other components to training (inpatient consults, outpatient clinics, social services experiences, mental health experience, court experience, and/or multidisciplinary team experience) (25%), quality of didactic teaching (15%), expertise of child abuse and neglect providers (14%), and no improvements needed (14%).
Many pediatric residents leave residency with limited clinical training in child abuse and neglect. Twenty-eight percent of all pediatric residents leave residency with exposure to <5 inpatients evaluated for abuse and neglect. Fifty-nine percent of pediatric residency programs offered no mandatory rotation, and 25% offered no rotations at all in child abuse and neglect, whereas 75% offered a mandatory rotation, an elective rotation, or both. Although 57% of the programs reported that they offered elective rotations, 77% of those programs reported that less than one quarter of their residents actually completed the rotation. Elective rotations provided more clinical experiences and exposure to more components of child abuse and neglect training than mandatory rotations. However, the low number of residents actually completing elective rotations likely made these rotations less beneficial than if more residents completed the elective rotations. Although mandatory rotations were more likely to be completed by residents, they were less comprehensive than elective rotations.
Most programs offered didactic training that was described as useful for their residents. Most didactic sessions taught about physical and sexual abuse, with fewer programs teaching about neglect or domestic violence. Neglect is the most common form of child abuse1 and needs further emphasis in didactic child abuse and neglect experiences. The relationship between domestic violence and child abuse and neglect has been well established17 and may be a useful topic in didactic sessions. Our finding that almost 50% of programs did not teach about domestic violence has been supported by earlier studies.17–19 Specific recommendations exist to incorporate domestic violence training in pediatric residency curricula.19
Our study demonstrated that providing high-quality didactic teaching of sufficient duration was associated with improved levels of preparedness. Even programs without clinical experiences in child abuse and neglect may better prepare their residents for child abuse and neglect by emphasizing and promoting quality didactic teaching. The use of nationally recognized lecture series in child abuse and neglect, such as The Visual Diagnosis of Child Physical Abuse from the American Academy of Pediatrics,20 may be useful for programs with limited clinical experience in child abuse and neglect. Specific core content for residency training in child abuse and neglect is available to assist implementation of training curricula.21
Our study also demonstrated that increased clinical experiences in child abuse and neglect during residency were associated with improved levels of preparedness to identify and evaluate patients for child abuse and neglect, although the P value of .07 for number of inpatients represents borderline significance at the P < .05 level in regression modeling. The level of preparedness is likely to be increased by providing more clinical experiences in child abuse and neglect during residency. These clinical experiences may occur in structured mandatory clinical rotations in child abuse and neglect, coverage of more child abuse and neglect inpatients, or exposure to more child abuse and neglect patients across various rotations (emergency department, PICU, general pediatrics clinic). Although it is likely that longer time spent with patients, in addition to increased number of patients seen, is associated with improved preparedness, our study did not measure this variable specifically.
Elective rotations were not significantly associated with improved levels of preparedness in our study. Elective rotations may not be significantly associated with preparedness because of the low number of residents actually completing elective rotations. Although it is likely that the small fraction of residents completing elective rotations were better prepared to address child abuse and neglect than those who did not complete an elective rotation in the same residency program, this particular question was not addressed in our study.
The combination of increased clinical experiences in child abuse and neglect, along with high-quality didactic sessions, is likely the most effective method for increasing levels of preparedness for graduating residents. Respondents identified the need for more clinical experiences in child abuse and neglect, with 41% wanting more patients seen and 52% wanting more time scheduled for training, and only 15% identified the need for improved quality of didactic teaching. This suggests that the biggest opportunity for improvement may be in increasing the quantity of clinical exposure to child abuse and neglect cases. Programs with low numbers of child abuse and neglect patients seen across rotations may benefit most from structured mandatory clinical rotations in child abuse and neglect.
Our findings are consistent with those of previous studies that assessed residency education in child abuse and neglect. The Institute of Medicine described the lack of effective family violence training for health care professionals, including the training of residents in child abuse and neglect.2 In 1988, Dubowitz4 noted that perceived adequacy of residency training in child abuse and neglect was better than adequate (27%), adequate (42%), or less than adequate (30%). This compared with our findings of very well prepared (12%), well prepared (54%), and somewhat to not well prepared (34%). This suggests that there may not be a large difference between the perceived adequacy of training of residents over the past 17 years despite multiple calls for improvements.3,4,6,12
Improvements in knowledge of child abuse and neglect have been previously shown to be correlated with increased exposure to patients, interdisciplinary teams, and time in didactic training.4,6,10,11 Despite correlations of improved knowledge and confidence with child abuse and neglect training experiences, most pediatric residencies in Canada were shown recently to lack mandatory child abuse and neglect training,6 and now we have shown this to also be true in the United States.
Improving the clinical skills of pediatricians to identify and evaluate child abuse and neglect may lead to reduced morbidity and mortality from child abuse. Presenting signs and symptoms of injury resulting from child abuse and neglect are often missed by clinicians.22 Training in early identification may improve outcomes in these cases.
The main limitation of this study is that the data are based on chief residents’ perception of the performance of residents rather than assessing clinical skills of the residents themselves. Chief residents were asked how prepared they perceived their residents to be to address child abuse and neglect on graduation, which may not necessarily reflect actual level of preparedness of the individual residents. Chief residents were also asked to report on average training, which may not reflect individual resident variability and may be less accurate. The perception of preparedness may also be affected by the experiences of the chief residents in their own child abuse and neglect training. It is important to note that the perceived level of preparedness, from either chief residents or the residents themselves, may not necessarily correlate with actual improved clinical skills.
Future studies are needed to determine how to best prepare residents for addressing child abuse and neglect. A direct survey of residents around the time of graduation may be helpful in assessing perceived levels of preparedness. However, more direct measures of the clinical skills of residents in child abuse and neglect would be preferred to truly distinguish perception of preparedness from actual clinical skills. This could be addressed through analysis of recorded standardized patient encounters, oral examinations, or chart reviews.
The American Board of Pediatrics recently voted to make child abuse and neglect a boarded subspecialty and is awaiting final approval.23 If child abuse and neglect does become a boarded subspecialty, it is unclear what may happen regarding child abuse and neglect training experiences. Historically, the assignment of subspecialty status has brought with it mandatory training requirements for recent subspecialties in general pediatrics, such as developmental/behavioral pediatrics and adolescent medicine.8 Both of these subspecialties have specific training requirements in the Residency Review Committee program requirements for general pediatrics residency.8 Although earlier calls for increased child abuse and neglect training have not translated into more widespread changes in training, perhaps the Residency Review Committee can make a more effective impact on general pediatrics residency training in child abuse by establishing more specific requirements such as making this training a “required subspecialty experience.”8 Our data support the notion that high-quality didactic and clinical training experiences in child abuse and neglect should be required to maximize the levels of preparedness to identify and evaluate child abuse and neglect.
We acknowledge Ernestine Briggs, PhD, and the Community Practice Site of the National Center for Child Traumatic Stress (a Substance Abuse and Mental Health Services Administration–funded project) for their contributions; the Center for Child and Family Health-North Carolina for their support; the Biometric Consulting Lab (Department of Biostatistics, University of North Carolina [Chapel Hill]), for statistical guidance; and Drew G. Narayan, MBA, MS, for assistance with data management.
- Accepted March 20, 2006.
- Address correspondence to Aditee Pradhan Narayan, MD, MPH, Center for Child and Family Health-North Carolina, 3518 Westgate Dr, Suite 100, Durham, NC 27707. E-mail:
The authors have indicated they have no financial relationships relevant to this article to disclose.
- ↵Sedlak AJ, Broadhurst DD; Administration on Children, Youth and Families, National Center on Child Abuse and Neglect. Executive Summary of the Third National Incidence Study of Child Abuse and Neglect. Washington, DC: US Department of Health and Human Services Administration for Children and Families; 1996
- ↵Board on Children, Youth, and Families, Committee on the Training Needs of Health Professionals to Respond to Family Violence. Cohn F, Salmon ME, Stobo JD, eds. Confronting Chronic Neglect: The Education and Training of Health Professionals on Family Violence. Washington, DC: National Academy Press; 2002
- ↵Dubowitz H. Child abuse programs and pediatric residency training. Pediatrics.1988;82 (3 pt 2):477–480
- ↵Accreditation Council for Graduate Medical Education. Program Requirements for Residency Education in Pediatrics. Available at: www.acgme.org/acWebsite/downloads/RRC_progReq/320pr106.pdf. Accessed November 16, 2005
- ↵SAS [computer program]. Version 9.1. Cary, NC: SAS Institute, Inc; 2002–2003
- ↵Microsoft Excel [computer program]. Version 2002. Redmond, WA: Microsoft Corporation; 1985–2001
- ↵American Medical Association. Fellowship and Residency Electronic Interactive Database. 2005. Available at: www.ama-assn.org/ama/pub/category/2997.html. Accessed November 10, 2005
- ↵Brinkhoff T. City population, 2005. Available at: www.citypopulation.de. Accessed November 10, 2005
- ↵Wright RJ, Wright RO, Isaac NE. Response to battered mothers in the pediatric emergency department: a call for an interdisciplinary approach to family violence. Pediatrics.1997;99 :186– 192
- ↵American Academy of Pediatrics; C. Henry Kempe National Center on Child Abuse and Neglect. The Visual Diagnosis of Child Physical Abuse [slide and study guide kit]. Elk Grove Village, IL: American Academy of Pediatrics; 1994
- ↵Starling SP, Boos S. Core content for residency training in child abuse and neglect. Child Maltreat.2003;8 :242– 247
- ↵American Board of Pediatrics. News, March 2006. Available at: www.abp.org/abpinfo/news.htm. Accessed April 10, 2006
- Copyright © 2006 by the American Academy of Pediatrics