G Model CHIABU-2752; No. of Pages 8
ARTICLE IN PRESS Child Abuse & Neglect xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Child Abuse & Neglect
Recurrent concerns for child abuse: Repeated consultations by a subspecialty child abuse team Jennifer Martindale a,∗ , Alice Swenson b , Jamye Coffman c , Alice W. Newton d , Daniel M. Lindberg e,f , for the ExSTRA Investigators a
Department of Emergency Medicine, SUNY Downstate, 450 Clarkson Avenue, Box 1228, Brooklyn, NY 11203, USA Children’s Hospital of Wisconsin, 8701 Watertown Plank Road, Milwaukee, WI 53226, USA Cook Children’s Medical Center, 801 7th Avenue, Fort Worth, TX 76104, USA d Massachusetts General Hospital, 55 Fruit Street, Boston, MA 02114, USA e Kempe Center for the Prevention and Treatment of Child Abuse and Neglect, 13123 East 16th Avenue B390, Aurora, CO 80045, USA f The Department of Emergency Medicine, University of Colorado School of Medicine, 13123 East 16th Avenue B390, Aurora, CO 80045, USA b c
a r t i c l e
i n f o
Article history: Received 3 December 2013 Received in revised form 8 February 2014 Accepted 18 March 2014 Available online xxx
Keywords: Abuse Maltreatment Child protection Foster care
a b s t r a c t Physically abused children may be repeatedly reported to child protection services and undergo multiple medical evaluations. Less is known about recurrent evaluations by hospital-based child abuse teams for possible abuse. The objectives of this study were to determine the frequency of repeated consultations by child abuse teams and to describe this cohort in terms of injury pattern, perceived likelihood of abuse, disposition plan, and factors related to repeat consultation. This was a prospectively planned, secondary analysis of data from the Examining Siblings to Recognize Abuse (ExSTRA) research network. Subjects included children younger than 10 years of age who were referred to child abuse subspecialty teams at one of 20 U.S. academic centers. Repeat consultations occurred in 101 (3.5%; 95% CI 2.9–4.2%) of 2890 subjects. The incidence of death was 4% (95% CI 1–9%) in subjects with repeated consults and 3% (95% CI 2–3%) in subjects with single consults. Perceived likelihood of abuse from initial to repeat visit remained low in 33% of subjects, remained high in 24.2% of subjects, went from low to high in 16.5%, and high to low in 26.4% of subjects. Themes identified among the subset of patients suspected of repeated abuse include return to the same environment, failure to comply with a safety plan, and abuse in foster care. Repeated consultation by child abuse specialists occurs for a minority of children. This group of children may be at higher risk of subsequent abuse and may represent an opportunity for quality improvement. © 2014 Elsevier Ltd. All rights reserved.
Child physical abuse is often a chronic, progressive, and fatal disease (Alexander, Crabbe, Sato, Smith, & Bennett, 1990; Deans et al., 2013; Thackeray, 2007). Despite growing awareness, the diagnosis of child abuse is often delayed or missed entirely (Jenny, Hymel, Ritzen, Reinert, & Hay, 1999). Some preexisting injuries are only recognized as having been inflicted when a child returns for care with more severe or obvious abuse (Alexander et al., 1990; Ewing-Cobbs et al., 1998; Ravichandiran et al., 2010; Rubin, Christian, Bilaniuk, Zazyczny, & Durbin, 2003). Missing the diagnosis of child abuse could
∗ Corresponding author. http://dx.doi.org/10.1016/j.chiabu.2014.03.007 0145-2134/© 2014 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Martindale, J., et al. Recurrent concerns for child abuse: Repeated consultations by a subspecialty child abuse team. Child Abuse & Neglect (2014), http://dx.doi.org/10.1016/j.chiabu.2014.03.007
G Model
ARTICLE IN PRESS
CHIABU-2752; No. of Pages 8
J. Martindale et al. / Child Abuse & Neglect xxx (2014) xxx–xxx
2
result in more severe injury or death (King, Kiesel, & Simon, 2006; Oral, Yagmur, Nashelsky, Turkmen, & Kirby, 2008; Ricci, Giantris, Merriam, Hodge, & Doyle, 2003). Despite legal mandates to do so in every state, physicians do not report some cases of suspected abuse to child protective services (CPS), even when they have a reasonable concern for abuse (Flaherty et al., 2008). The standard for reporting to child protection agencies is interpreted variably by generalists and specialists (Laskey, Sheridan, & Hymel, 2007; Levi & Brown, 2005; Levi, Brown, & Erb, 2006; Lindberg, Lindsell, & Shapiro, 2008). Responses of child protection agencies are similarly variable(U.S. Department of Health and Human Services [DHHS], 2011). In the absence of gold-standard diagnostic tests for abuse for most cases (Southall, Plunkett, Banks, Falkov, & Samuels, 1997), CPS is faced with complicated decisions which require balancing the likelihood of recurrent abuse with the goal of family preservation. As a result, some children with inflicted injuries are returned to the same abusive environment in which the injuries occurred. Studies of children who are repeatedly reported to CPS or undergo multiple medical evaluations have been conducted to identify risk factors associated with recidivism and re-abuse (Dakil, Sakai, Lin, & Flores, 2011; Deans et al., 2013; Fluke, 2008; Kohl, Jonson-Reid, & Drake, 2009). However, there are no published data about recidivism for children seen by a hospital-based, subspecialty child abuse team. Multidisciplinary, hospital-based child abuse teams have been established to assist clinicians and CPS workers caring for children with injuries indicative of abuse (Block, 1998; Block & Palusci, 2006). Because child abuse team consultation implies at least some concern for abuse, children who return after an initial consultation may represent a missed opportunity for abuse prevention. Purpose The primary objective of this observational study was to determine the frequency of repeated consultations by child abuse teams among children evaluated for possible physical abuse. The secondary objective was to describe demographics, mortality, injuries, level of suspicion for abuse, dispositions, and typographical themes associated with repeated consultation in this cohort of individuals. Method Study design This is a prospectively planned secondary analysis of the Examining Siblings To Recognize Abuse (ExSTRA) research network, an observational study of 20 hospital-based, subspecialty child abuse teams in the United States conducted between January 2010 and April 2011 (Lindberg, Shapiro, Laskey, Pallin, Blood, & Berger, 2012). Each participating center and the datacoordinating center obtained approval for the parent study from their local Institutional Review Board (IRB). This secondary analysis of data, purged of all individual identifiers, was determined by each IRB to be exempt from IRB review as human individuals research. Inclusion criteria and data collection Child abuse teams collected data from all children less than 10 years old who underwent subspecialty evaluation for concern for physical abuse. Although the parent study of the ExSTRA research network focused on siblings and other contact children, this analysis deals only with index children and does not include data from siblings or other contacts of children evaluated with concern for abuse. Investigators reported whether their team had previously evaluated each index child or any of their contacts. Investigators recorded data for the initial consult retrospectively for individuals whose initial visit occurred before the start of the study enrollment period and prospectively for individuals with multiple visits during the 15-month enrollment period. Abstracted information was limited to that which is acquired in the normal course of clinical care. Participants were asked to record the disposition of the individual and to rate the likelihood of child physical abuse based on a previously published 7-point ordinal scale (Lindberg et al., 2008; see Table 1). Participants included the disposition of each individual in one of Table 1 Rating scale for perceived abuse likelihood. Number 1 2 3 4 5 6 7
Summary statement Definitely not inflicted injury No concern for inflicted injury Mild concern for inflicted injury Intermediate concern for inflicted injury Very concerning for inflicted injury Substantial evidence of inflicted injury Definite inflicted injury
Please cite this article in press as: Martindale, J., et al. Recurrent concerns for child abuse: Repeated consultations by a subspecialty child abuse team. Child Abuse & Neglect (2014), http://dx.doi.org/10.1016/j.chiabu.2014.03.007
G Model CHIABU-2752; No. of Pages 8
ARTICLE IN PRESS J. Martindale et al. / Child Abuse & Neglect xxx (2014) xxx–xxx
3
Fig. 1. Subject flow. Of the 2890 subjects included in the ExSTRA study, 161 underwent more than one consultation. This study focused on the 102 consultations (including 1 subject with 2 repeated consultations) obtained for recurrent concerns for abuse in the index child.
the following mutually exclusive categories: original home-no restrictions, original home-with restrictions, relative’s home, unrelated home, deceased, and unknown. Data analysis We compared demographics between children with single consultations and children with repeated consultations using the chi-squared test for dichotomous variables and t-tests for continuous variables. For analysis of the change in perceived likelihood of abuse, a low likelihood was considered to be ≤4 and a high level of concern to be ≥5 on the 7-point scale. The ExSTRA research network did not record dates of hospital admission or initial consultation. Minimum follow-up time was calculated by determining the difference between the time when a individual was initially entered into the data collection system and the end of the enrollment period. Qualitative analysis A qualitative analysis of repeated child abuse evaluations was performed in recognition of the complexity of child abuse evaluation and decision-making. The study authors took an inductive approach to develop a list of possible relevant themes as they emerged while reviewing individual case records (Bradley, Curry, & Devers, 2007). These themes were intended to describe factors associated with repeated abuse and to categorize those children in whom abuse may not have been identified despite repeated consultation. Each study author independently reviewed each case to increase the breadth and depth of coding individual cases. Summary findings were shared among the authors and records were iteratively reviewed to refine definitions of existing themes and identify additional themes until thematic saturation was reached. A theme was ultimately determined to apply to a case if it was identified by 3 of 5 independent reviewers. Themes were not mutually exclusive and some cases had multiple themes identified by a majority of investigators. Results This study included 2890 children enrolled in the ExSTRA study (see Fig. 1). One individual was removed from the sample because incomplete information made it unclear if the visit was part of an initial or repeat consultation. Mean minimum follow-up time was 189 days (standard deviation ± 144 days). A total of 102 (3.5%) consultations were for recurrent concerns for physical abuse. The initial visit for 30 of these consultations occurred during the study enrollment period. One individual underwent three separate consultations for abuse. Repeat visits occurred from two days to 72 months (median duration between consultations = 6 months, IQR 3–16 months). Demographic information appears in Table 2. No statistically significant demographic differences were noted between children evaluated once and those evaluated on multiple occasions.
Please cite this article in press as: Martindale, J., et al. Recurrent concerns for child abuse: Repeated consultations by a subspecialty child abuse team. Child Abuse & Neglect (2014), http://dx.doi.org/10.1016/j.chiabu.2014.03.007
G Model
ARTICLE IN PRESS
CHIABU-2752; No. of Pages 8
J. Martindale et al. / Child Abuse & Neglect xxx (2014) xxx–xxx
4
Table 2 Baseline characteristics of subjects with single and repeated consultations. Repeated consultations N Male (%) Racea White Black Other Unknown/not reported Ethnicity Hispanic Non-Hispanic Unknown/not reported Median age in months (IQR) a
Single
101a 53 (52%)
2728 1597 (58%)
63 (62%) 26 (26%) 9 (9%) 3 (3%)
1723 (63%) 611 (22%) 249 (9%) 145 (5%)
29 (29%) 64 (63%) 8 (8%) Initial consult 15 (6.5–29.0)
620 (22%) 1946 (71%) 162 (5%) Repeat consult 25 (12.7–48.5)
10 (4.0–30.1)
101 subjects underwent a total of 102 consultations (1 subject with 2 repeated consultations).
Table 3 Comparison of injury patterns. Injury N (%)
Initial N = 102
Cutaneous Bruises Burns Fracture Thoracic/Abdominal TBI Skull fracture RH Death
31 (31) 23 (23) 2 (2) 15 (15) 0 (0) 13 (13) 13 (13) 1 (1) n/a
Repeat N = 102
Single Visit N = 2728
65 (64) 49 (48) 11 (11) 15 (15) 2 (2) 6 (6) 5 (5) 2 (2) 4 (4)
1503 (55) 1005 (37) 210 (8) 1144 (42) 92 (3) 568 (21) 443 (16) 246 (9) 69 (3)
OR Initial vs. Single Consult (95% CI) 0.36 (0.23–0.56) 0.50 (0.30–0.82) 0.24 (0.04–0.10) 0.24 (0.13–0.43) 0 (0–1.37) 0.55 (0.29–1.03) 0.75 (0.40–1.40) 0.10 (0.01–0.66) n/a
Injuries Cutaneous injuries were the most common type of injury identified regardless of the number of presentations (Table 3). Fractures occurred in a higher percentage of individuals in the single consultation subgroup (42%, 95% CI [40, 44]) than in those individuals on initial (15%, 95% CI [8, 23]) and repeated (15%, 95% CI [8, 23]) visits. The incidence of death was 4% (95% CI [1, 9]) in individuals who had been evaluated previously and 3% (95% CI [2, 3]) in those with single visits. Although the ExSTRA research network included only individuals who were evaluated for concerns of physical abuse, investigators identified 28 individuals that had previously been evaluated by their hospital-based child abuse teams for other concerns, including neglect (n = 14) and sexual abuse (n = 14). Perceived likelihood of abuse For individuals with multiple visits, median level of perceived abuse likelihood was 4 (intermediate) on initial visit and 5 (very concerning) on repeat visit. Perceived likelihood of abuse remained low in 33% (95% CI [23, 44]) of individuals, remained high in 24.2% (95% CI [16, 34]) of individuals, went from low to high in 16.5% (95% CI [10, 26]), and high to low in 26% (95% CI [18, 37]) of individuals (see Table 4). Table 4 Perceived likelihood of abuse ratings on initial and repeat consultation. Level of Concern on Repeat Consult
Level of concern on initial consult
1 2 3 4 5 6 7
1
2
3
4
5
6
7
1 1 0 1 1 0 1
0 3 6 2 0 2 3
0 3 3 3 1 2 4
1 2 2 2 3 4 3
0 2 0 1 7 1 4
0 0 0 2 3 3 0
1 6 2 1 0 0 4
Each cell contains the number of subjects with repeat consultations for the given perceived level of concern for abuse on the initial consultation (rows) and repeat consultation (columns). Darker shading represents cases we have coded as having a low level of concern on both initial and subsequent consultations. Lighter shading represents cases with a high level of concern for abuse on initial and subsequent consultation.
Please cite this article in press as: Martindale, J., et al. Recurrent concerns for child abuse: Repeated consultations by a subspecialty child abuse team. Child Abuse & Neglect (2014), http://dx.doi.org/10.1016/j.chiabu.2014.03.007
G Model
ARTICLE IN PRESS
CHIABU-2752; No. of Pages 8
J. Martindale et al. / Child Abuse & Neglect xxx (2014) xxx–xxx
5
Table 5 Comparison of disposition resulting from consultation.
Deceased Original home: no restrictions Original home: with restrictions Relative home Unrelated home Unknown
Initial consult
Repeat consult
OR initial vs. repeat consult (95% CI)
Single consult
n/a 38 12 12 16 11
4 31 13 17 29 8
n/a 1.43 (0.76–2.67) 0.91 (0.37–2.28) 0.69 (0.29–1.64) 0.49 (0.23–1.02) 0.49 (0.52–4.23)
78 1169 350 436 479 216
OR initial vs. single consult (95% CI)
OR repeat vs. single consult (95% CI)
0.83 (0.54–1.28) 0.94 (0.48–1.78) 0.73 (0.37–1.38) 0.91 (0.51–1.60) 1.45 (0.72–2.89)
1.38 (0.42–4.04) 0.66 (0.41–1.04) 0.99 (0.52–1.85) 1.05 (0.60–1.83) 1.87 (1.17–2.96) 0.99 (0.44–2.14)
Disposition Individuals were more likely to be sent to an unrelated home (OR = 1.87; 95% CI [1.17, 2.96]) after a repeat consultation than those individuals who were evaluated only once (see Table 5). There were no statistically significant differences in disposition after initial and repeated consultations. Themes We identified seven themes that occurred in multiple cases when children underwent more than one evaluation for abuse. Ninety-four cases were determined to have at least one of the identified themes. The other cases included five without sufficient information to determine whether themes applied and three in which no theme was identified by a majority of authors. Child returned to the same environment. The most common factor related to repeated child abuse consultation was a child that was abused after returning to the same environment from which the initial concern for abuse arose. This finding was noted in 36 cases. The initial level of perceived abuse likelihood in 19 of these cases (53%) was 5 (very concerning) or higher. Example 1. A 36-month-old female was seen in a clinic for bruising, abrasions, and a chipped tooth (level of perceived abuse likelihood: 4) before being returned home without restrictions. Four days later, while reportedly left unattended by the mother’s boyfriend, the child was found facedown in bathwater. She was pulseless and apneic when emergency medical services arrived. The level of concern on the subsequent visit was 5. Safety plan not followed. Recurrent maltreatment occurred in seven individuals for whom a safety plan had been established but had not been diligently followed. Abuse in these individuals occurred during an unsupervised visit with the perpetrator or while living in a home from which a child was presumed to have been removed. Example 2. Child abuse specialists initially evaluated a 13-month-old boy for patterned facial injury and he was removed from his mother’s care and placed with his maternal grandmother. Six months later, a second child abuse consult was made to evaluate cutaneous bruising; Care was transferred then to the maternal grandfather. At approximately 20 months of age, this individual received medical evaluation for left-sided hemiparesis attributed to subdural hemorrhage and ischemic stroke one week after reported fall. At this time, the child’s injuries were not initially thought to be the result of abuse and the child abuse team was not re-consulted. A third child abuse consultation was requested and performed when the individual arrived in the emergency department, pulseless and unresponsive. Autopsy revealed acute subdural hemorrhage, cerebral edema, optic nerve sheath hemorrhage, cutaneous injury to the upper lip, and mesenteric tears. It was discovered after this individual’s death that the paternal grandfather returned the individual to the mother’s care without informing CPS. Abuse occurred in a different environment. Recurrent maltreatment was thought to occur in eight children who had been removed from one abusive setting and placed in another. Four of these children were evaluated for injuries that occurred in a foster home. Example 3. A male infant was placed in a foster home after evaluation for a femur fracture. At 28 months, he was evaluated for multiple burns and more than ten bruises. Repeat skeletal survey revealed multiple bilateral rib fractures. Injuries were thought to have occurred in the foster home. He was then placed in another foster home. Concern for abuse remained low. In 25 cases, the perceived likelihood of abuse on both initial and repeat visits was determined to be low. In most of these cases, a low level of concern was assigned when mechanism matched injury pattern. We did not identify factors in these individuals that might have decreased the threshold to undertake evaluation. Example 4. A 28-month old girl was initially evaluated for a radial head fracture (initial perceived level of abuse likelihood: 3). Two months later she was evaluated for abrasions on her back that were thought to be consistent with abrasions from a non-abusive scrape from concrete (level of perceived abuse likelihood: 2). No concern for abuse on repeat evaluation. After an initial visit that raised a high level of concern for abuse in five children, child abuse teams determined that there was no concern for repeated abuse during a later evaluation. Please cite this article in press as: Martindale, J., et al. Recurrent concerns for child abuse: Repeated consultations by a subspecialty child abuse team. Child Abuse & Neglect (2014), http://dx.doi.org/10.1016/j.chiabu.2014.03.007
G Model CHIABU-2752; No. of Pages 8
ARTICLE IN PRESS J. Martindale et al. / Child Abuse & Neglect xxx (2014) xxx–xxx
6
Example 5. A 25-month-old female was evaluated for bruising to the right ear, scalp, and cheek and a left proximal humerus fracture. The initial perceived likelihood for abuse was seven and the child was sent to an unrelated home. She was evaluated 14 months later for persistent pain at the site of a distal ulnar fracture attributed to a witnessed fall at daycare. While the repeat evaluation was initiated because of concern by one physician that the child’s splint had been prematurely removed, the child’s primary care physician had given instructions that the splint could be removed. There was no concern for inflicted injury on this second visit. Complication of initial injury. A second child abuse consultation was obtained for two children who had radiologic findings consistent with the natural progression of subdural hematomas. Example 6. A 6.5-month-old boy was evaluated for traumatic brain injury and found to have a complex skull fracture, subdural hemorrhage with evidence of parenchymal shear injury and multiple healing rib fractures. Perceived likelihood of abuse was rated as seven and the patient was placed in a relative’s home. The child abuse team was consulted again when a new area of discrete subdural blood was identified within the resolving subdural hemorrhage on a scheduled follow-up MRI. This radiologic finding was thought not to represent a new traumatic injury. Initial evaluation not for physical abuse. Initial child abuse consultation was requested for reasons other than suspected physical abuse (neglect, alleged sexual abuse) in 28 of these cases. Example 7. A 13-month-old female was initially evaluated for possible sexual assault when blood-stained underwear was found by the mother. She suspected her boyfriend at the time. Ten months later, the patient was reportedly found by a different boyfriend in the same home to be pinned underneath a dollhouse. The patient was found to have a buckle fracture of the distal femoral metaphysics and bruises on the buttock, knee and lumbar-sacral region. The patient returned to the same home with restrictions. Discussion This article contains, to our knowledge, the first description of children who are repeatedly evaluated by a child abuse team. We found that 3.5% of children who were evaluated for possible physical abuse had more than one evaluation by a child abuse specialty team. This finding does not suggest that 3.5% of cases represent missed opportunities to prevent recurrent maltreatment. The perceived likelihood of abuse remained low or decreased after repeated evaluations in 59% of children for whom we had data about perceived abuse likelihood. The incidence of repeated consultations in this study is lower than repeated reports to CPS. This finding is likely related to the lower number of injuries for which medical evaluation is sought and the limited time period during which individuals were followed. Children in this study who returned for care were most commonly found to have cutaneous injuries on initial evaluation. Because accidental bruising is common, it is often thought to require only minor force. However, certain types of bruising such as facial bruising, bruising in infants who are not yet independently mobile, abdominal bruising, or bruising to the ear may represent red flags that imply a violent home environment. In general, individuals who returned were less likely to have severe injury (retinal hemorrhages, fractures) than those seen only once. More serious injuries may have led to more aggressive interventions to protect children from re-abuse. Alternatively, children with an initial report of abuse may have had closer monitoring for signs of abuse and may have had recurrent referrals with more subtle findings than children who had not previously been referred. Several types of potential failures and compliance issues within the systems built to ensure child safety were identified. These potential failures to protect children occurred either within the hospital or once the child had been discharged with CPS oversight or custody. The most common theme among those individuals who underwent repeat consultation was repeated abuse after returning to the environment where abuse was first suspected (36 cases). Recidivism has been previously described in caregivers that continue to care for abused children (Alexander et al., 1990; Dakil et al., 2011; Ellaway et al., 2004). Recurrent maltreatment in children remaining in parental custody ranges from 17% to 35% (Dakil et al., 2011; Fluke, 2008). In seven of these 36 cases, the child abuse team’s initial perceived likelihood of inflicted injury was low, or was not recorded. These cases may reflect misattribution of injuries to accidental trauma by child abuse specialists. In 16 of 36 cases, the child abuse team expressed a high, perceived likelihood of abuse. These cases might represent opportunities for CPS systems to improve risk assessment or collaboration with child abuse teams. Repeated abuse also occurred in the context of a safety plan or intervention that was not closely followed. In several cases, injuries were observed immediately after visitation with parents. In other cases, children were later discovered to be cared for by, and residing with, initial perpetrators rather than the home and caregivers designated by CPS. Repeated abuse occurred in the setting of a new foster home in four children. Children in the foster care system are at higher risk for physical abuse (DHHS, 2011). Return to an abusive environment, failed compliance with a safety plan, and abuse in foster care are themes this study identified as being associated with recurrent abuse, but these data, obtained with the benefit of hindsight, do not identify any single actor or agency solely responsible for missed opportunities to prevent recurrent abuse. Our study does not suggest any single, easy intervention. Implementation of aggressive safety planning for all abuse consultations with minor injuries would be implausible. Placement of children in foster care is not without cost to the child, family, and society; recurrent maltreatment also occurs in children placed outside of the home. Instead, we conclude that close and formal review of Please cite this article in press as: Martindale, J., et al. Recurrent concerns for child abuse: Repeated consultations by a subspecialty child abuse team. Child Abuse & Neglect (2014), http://dx.doi.org/10.1016/j.chiabu.2014.03.007
G Model CHIABU-2752; No. of Pages 8
ARTICLE IN PRESS J. Martindale et al. / Child Abuse & Neglect xxx (2014) xxx–xxx
7
children who are evaluated on multiple occasions by a child abuse team and in the end, are suspected of having repeated abuse, may be an opportunity for child abuse specialists to identify areas of system improvement and compliance. Study limitations This study has important limitations. Although our qualitative methodology may be useful for hypothesis generation, we are not able to determine the presence of each theme for the large denominator of individuals who did not return for subsequent evaluation. We are therefore unable to determine the test characteristics (sensitivity, predictive value) of each identified theme. Some data regarding initial consultation were obtained retrospectively and may therefore be individual to recall bias. For those individuals in whom an initial evaluation took place before ExSTRA enrollment, data collection relied on the memory of specialists participating in the ExSTRA study. The 3.5% of children who underwent repeat child abuse consultation is comprised only of those children who were evaluated more than once by the same child abuse team at a participating hospital. Some children categorized as having undergone single child abuse evaluation may have been evaluated by specialists at a non-participating institution. We are also limited by our method of counting repeat evaluations over time. An optimal approach would be to prospectively enroll a group of children referred with an initial concern for abuse and follow them for a long and pre-specified period. Because we were most interested in identifying factors associated with repeat evaluations, and to avoid logistical and ethical challenges with this approach, we chose to examine a cohort that had previously been enrolled, and we chose to include children whose initial evaluation occurred prior to the enrollment period. One major theme that was identified was that of children whose subsequent evaluation for physical abuse followed an initial consultation that was for neglect, sexual abuse, or another maltreatment type. Our results likely underestimate the significance of this phenomenon because the parent dataset included only individuals referred with concerns of physical abuse, rather than examining children referred for any maltreatment type. Our estimate of the minimal follow-up time is certainly an underestimate because the date of the initial consultation was not abstracted; Data was entered for some children after an unmeasured delay. Finally, children who underwent repeat evaluation after the enrollment period would not be counted. Implications and future research This study finds that child abuse teams are consulted to evaluate an important minority of individuals on more than one occasion. This population may reflect missed opportunities to effectively intervene when a suspicion for child abuse had already been declared. This study may generate hypotheses that direct prospective investigations into secondary prevention of child abuse and inform future efforts to improve communication between child abuse specialists and child protection services. Acknowledgements The Examining Siblings To Recognize Abuse (ExSTRA) investigators are: Deb Bretl, APNP (Children’s Hospital Wisconsin, Wauwatosa, WI), Nancy Harper, MD (Driscoll Children’s Hospital, Corpus Cristi, TX), Katherine Deye, MD (Children’s National Medical Center, Washington, DC), Antoinette L. Laskey, MD and Tara Harris, MD (Riley Hospital for Children, Indianapolis, IN), Yolanda Duralde, MD (Mary Bridge Children’s Health Center, Tacoma, WA), Marcella Donaruma-Kwoh, MD (Texas Children’s Hospital, Houston, TX), Daryl Steiner, DO (Akron Children’s Hospital, Akron, OH), Ken Feldman, MD (Seattle Children’s Hospital, Seattle, WA), Kimberly Schwartz, MD (University of Massachusetts Medical Center, Worcester, MA), Robert A. Shapiro, MD and Mary Greiner, MD (Cincinnati Children’s Hospital Medical Center, Cincinnati, OH), Ivone Kim, MD (Children’s Hospital Pittsburgh or University of Pittsburgh Medical Center), Kent Hymel, MD (Dartmouth-Hitchcock Medical Center, Lebanon, NH), Suzanne Haney, MD (Children’s Hospital & Medical Center, Omaha, NE), Alicia Pekarsky, MD (SUNY Upstate Medical University, Syracuse, NY), Andrea Asnes, MD (Yale-New Haven Children’s Hospital, New Haven, CT), Paul McPherson, MD (Akron Children’s Hospital, Youngstown, OH), Neha Mehta, MD (Sunrise Children’s Hospital, Las Vegas, NV), and Gwendolyn Gladstone, MD (Exeter Pediatric Associates, Exeter, NH). References Alexander, R., Crabbe, L., Sato, Y., Smith, W., & Bennett, T. (1990). Serial abuse in children who are shaken. The American Journal of Diseases of Children, 144, 58–60. Block, R. (1998). What doctors who perform child abuse medical evaluations want the rest of the team to know. Journal of the Oklahoma State Medical Association, 91(8), 457. Block, R. W., & Palusci, V. J. (2006). Child abuse pediatrics: A new pediatric subspecialty. Journal of Pediatrics, 148, 711–712. Bradley, E. H., Curry, L. A., & Devers, K. J. (2007). Qualitative data analysis for health services research: Developing taxonomy, themes, and theory. Health Services Research, 42, 1758–1772. Dakil, S. R., Sakai, C., Lin, H., & Flores, G. (2011). Recidivism in the child protection system: Identifying children at greatest risk of reabuse among those remaining in the home. Archives of Pediatrics and Adolescent Medicine, 165, 1006–1012.
Please cite this article in press as: Martindale, J., et al. Recurrent concerns for child abuse: Repeated consultations by a subspecialty child abuse team. Child Abuse & Neglect (2014), http://dx.doi.org/10.1016/j.chiabu.2014.03.007
G Model CHIABU-2752; No. of Pages 8
8
ARTICLE IN PRESS J. Martindale et al. / Child Abuse & Neglect xxx (2014) xxx–xxx
Deans, K. J., Thackeray, J., Askegard-Giesmann, J. R., Earley, E., Groner, J. I., & Minneci, P. C. (2013). Mortality increases with recurrent episodes of nonaccidental trauma in children. Journal of Trauma and Acute Care Surgery, 75, 161–165. Ellaway, B. A., Payne, E. H., Rolfe, K., Dunstan, F. D., Kemp, A. M., Butler, I., & Sibert, J. R. (2004). Are abused babies protected from further abuse? Archives of Disease in Childhood, 89, 845–846. Ewing-Cobbs, L., Kramer, L., Prasad, M., Canales, D. N., Louis, P. T., Fletcher, J. M., Vollero, H., Landry, S. H., & Cheung, K. (1998). Neuroimaging, physical, and developmental findings after inflicted and noninflicted traumatic brain injury in young children. Pediatrics, 102(2 (Pt. 1)), 300–307. Flaherty, E. G., Sege, R. D., Griffith, J., Price, L. L., Wasserman, R., Slora, E., & Binns, H. J. (2008). From suspicion of physical child abuse to reporting: Primary care clinician decision-making. Pediatrics, 122, 611–619. Fluke, J. D. (2008). Child protective services rereporting and recurrence – Context and considerations regarding research. Child Abuse & Neglect, 32, 749–751. Jenny, C., Hymel, K. P., Ritzen, A., Reinert, S. E., & Hay, T. C. (1999). Analysis of missed cases of abusive head trauma. JAMA, 281, 621–626. King, W. K., Kiesel, E. L., & Simon, H. K. (2006). Child abuse fatalities: Are we missing opportunities for intervention? Pediatric Emergency Care, 22, 211–214. Kohl, P. L., Jonson-Reid, M., & Drake, B. (2009). Time to leave substantiation behind: Findings from a national probability study. Child Maltreatment, 14, 17–26. Laskey, A. L., Sheridan, M. J., & Hymel, K. P. (2007). Physicians’ initial forensic impressions of hypothetical cases of pediatric traumatic brain injury. Child Abuse & Neglect, 31, 329–342. Levi, B. H., & Brown, G. (2005). Reasonable suspicion: A study of Pennsylvania pediatricians regarding child abuse. Pediatrics, 116, e5–e12. Levi, B. H., Brown, G., & Erb, C. (2006). Reasonable suspicion: A pilot study of pediatric residents. Child Abuse & Neglect, 30, 345–356. Lindberg, D. M., Lindsell, C. J., & Shapiro, R. A. (2008). Variability in expert assessments of child physical abuse likelihood. Pediatrics, 121, e945–e953. Lindberg, D. M., Shapiro, R. A., Laskey, A. L., Pallin, D. J., Blood, E. A., & Berger, R. P. (2012). Prevalence of abusive injuries in siblings and household contacts of physically abused children. Pediatrics, 130, 193–201. Oral, R., Yagmur, F., Nashelsky, M., Turkmen, M., & Kirby, P. (2008). Fatal abusive head trauma cases: Consequence of medical staff missing milder forms of physical abuse. Pediatric Emergency Care, 24, 816–821. Ravichandiran, N., Schuh, S., Bejuk, M., Al-Harthy, N., Shouldice, M., Au, H., & Boutis, K. (2010). Delayed identification of pediatric abuse-related fractures. Pediatrics, 125, 60. Ricci, L., Giantris, A., Merriam, P., Hodge, S., & Doyle, T. (2003). Abusive head trauma in Maine infants: Medical, child protective, and law enforcement analysis. Child Abuse & Neglect, 27, 271–283. Rubin, D. M., Christian, C. W., Bilaniuk, L. T., Zazyczny, K. A., & Durbin, D. R. (2003). Occult head injury in high-risk abused children. Pediatrics, 111, 1382–1386. Southall, D. P., Plunkett, M. C., Banks, M. W., Falkov, A. F., & Samuels, M. P. (1997). Covert video recordings of life-threatening child abuse: Lessons for child protection. Pediatrics, 100, 735–760. Thackeray, J. D. (2007). Frena tears and abusive head injury: A cautionary tale. Pediatric Emergency Care, 23, 735–737. U.S. Department of Health and Human Services, Administration for Children and Families, Children’s Bureau. (2011). Child Maltreatment 2011. Washington, DC: Author.
Please cite this article in press as: Martindale, J., et al. Recurrent concerns for child abuse: Repeated consultations by a subspecialty child abuse team. Child Abuse & Neglect (2014), http://dx.doi.org/10.1016/j.chiabu.2014.03.007
ARTICLE IN PRESS Child Abuse & Neglect xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Child Abuse & Neglect
Recurrent concerns for child abuse: Repeated consultations by a subspecialty child abuse team Jennifer Martindale a,∗ , Alice Swenson b , Jamye Coffman c , Alice W. Newton d , Daniel M. Lindberg e,f , for the ExSTRA Investigators a
Department of Emergency Medicine, SUNY Downstate, 450 Clarkson Avenue, Box 1228, Brooklyn, NY 11203, USA Children’s Hospital of Wisconsin, 8701 Watertown Plank Road, Milwaukee, WI 53226, USA Cook Children’s Medical Center, 801 7th Avenue, Fort Worth, TX 76104, USA d Massachusetts General Hospital, 55 Fruit Street, Boston, MA 02114, USA e Kempe Center for the Prevention and Treatment of Child Abuse and Neglect, 13123 East 16th Avenue B390, Aurora, CO 80045, USA f The Department of Emergency Medicine, University of Colorado School of Medicine, 13123 East 16th Avenue B390, Aurora, CO 80045, USA b c
a r t i c l e
i n f o
Article history: Received 3 December 2013 Received in revised form 8 February 2014 Accepted 18 March 2014 Available online xxx
Keywords: Abuse Maltreatment Child protection Foster care
a b s t r a c t Physically abused children may be repeatedly reported to child protection services and undergo multiple medical evaluations. Less is known about recurrent evaluations by hospital-based child abuse teams for possible abuse. The objectives of this study were to determine the frequency of repeated consultations by child abuse teams and to describe this cohort in terms of injury pattern, perceived likelihood of abuse, disposition plan, and factors related to repeat consultation. This was a prospectively planned, secondary analysis of data from the Examining Siblings to Recognize Abuse (ExSTRA) research network. Subjects included children younger than 10 years of age who were referred to child abuse subspecialty teams at one of 20 U.S. academic centers. Repeat consultations occurred in 101 (3.5%; 95% CI 2.9–4.2%) of 2890 subjects. The incidence of death was 4% (95% CI 1–9%) in subjects with repeated consults and 3% (95% CI 2–3%) in subjects with single consults. Perceived likelihood of abuse from initial to repeat visit remained low in 33% of subjects, remained high in 24.2% of subjects, went from low to high in 16.5%, and high to low in 26.4% of subjects. Themes identified among the subset of patients suspected of repeated abuse include return to the same environment, failure to comply with a safety plan, and abuse in foster care. Repeated consultation by child abuse specialists occurs for a minority of children. This group of children may be at higher risk of subsequent abuse and may represent an opportunity for quality improvement. © 2014 Elsevier Ltd. All rights reserved.
Child physical abuse is often a chronic, progressive, and fatal disease (Alexander, Crabbe, Sato, Smith, & Bennett, 1990; Deans et al., 2013; Thackeray, 2007). Despite growing awareness, the diagnosis of child abuse is often delayed or missed entirely (Jenny, Hymel, Ritzen, Reinert, & Hay, 1999). Some preexisting injuries are only recognized as having been inflicted when a child returns for care with more severe or obvious abuse (Alexander et al., 1990; Ewing-Cobbs et al., 1998; Ravichandiran et al., 2010; Rubin, Christian, Bilaniuk, Zazyczny, & Durbin, 2003). Missing the diagnosis of child abuse could
∗ Corresponding author. http://dx.doi.org/10.1016/j.chiabu.2014.03.007 0145-2134/© 2014 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Martindale, J., et al. Recurrent concerns for child abuse: Repeated consultations by a subspecialty child abuse team. Child Abuse & Neglect (2014), http://dx.doi.org/10.1016/j.chiabu.2014.03.007
G Model
ARTICLE IN PRESS
CHIABU-2752; No. of Pages 8
J. Martindale et al. / Child Abuse & Neglect xxx (2014) xxx–xxx
2
result in more severe injury or death (King, Kiesel, & Simon, 2006; Oral, Yagmur, Nashelsky, Turkmen, & Kirby, 2008; Ricci, Giantris, Merriam, Hodge, & Doyle, 2003). Despite legal mandates to do so in every state, physicians do not report some cases of suspected abuse to child protective services (CPS), even when they have a reasonable concern for abuse (Flaherty et al., 2008). The standard for reporting to child protection agencies is interpreted variably by generalists and specialists (Laskey, Sheridan, & Hymel, 2007; Levi & Brown, 2005; Levi, Brown, & Erb, 2006; Lindberg, Lindsell, & Shapiro, 2008). Responses of child protection agencies are similarly variable(U.S. Department of Health and Human Services [DHHS], 2011). In the absence of gold-standard diagnostic tests for abuse for most cases (Southall, Plunkett, Banks, Falkov, & Samuels, 1997), CPS is faced with complicated decisions which require balancing the likelihood of recurrent abuse with the goal of family preservation. As a result, some children with inflicted injuries are returned to the same abusive environment in which the injuries occurred. Studies of children who are repeatedly reported to CPS or undergo multiple medical evaluations have been conducted to identify risk factors associated with recidivism and re-abuse (Dakil, Sakai, Lin, & Flores, 2011; Deans et al., 2013; Fluke, 2008; Kohl, Jonson-Reid, & Drake, 2009). However, there are no published data about recidivism for children seen by a hospital-based, subspecialty child abuse team. Multidisciplinary, hospital-based child abuse teams have been established to assist clinicians and CPS workers caring for children with injuries indicative of abuse (Block, 1998; Block & Palusci, 2006). Because child abuse team consultation implies at least some concern for abuse, children who return after an initial consultation may represent a missed opportunity for abuse prevention. Purpose The primary objective of this observational study was to determine the frequency of repeated consultations by child abuse teams among children evaluated for possible physical abuse. The secondary objective was to describe demographics, mortality, injuries, level of suspicion for abuse, dispositions, and typographical themes associated with repeated consultation in this cohort of individuals. Method Study design This is a prospectively planned secondary analysis of the Examining Siblings To Recognize Abuse (ExSTRA) research network, an observational study of 20 hospital-based, subspecialty child abuse teams in the United States conducted between January 2010 and April 2011 (Lindberg, Shapiro, Laskey, Pallin, Blood, & Berger, 2012). Each participating center and the datacoordinating center obtained approval for the parent study from their local Institutional Review Board (IRB). This secondary analysis of data, purged of all individual identifiers, was determined by each IRB to be exempt from IRB review as human individuals research. Inclusion criteria and data collection Child abuse teams collected data from all children less than 10 years old who underwent subspecialty evaluation for concern for physical abuse. Although the parent study of the ExSTRA research network focused on siblings and other contact children, this analysis deals only with index children and does not include data from siblings or other contacts of children evaluated with concern for abuse. Investigators reported whether their team had previously evaluated each index child or any of their contacts. Investigators recorded data for the initial consult retrospectively for individuals whose initial visit occurred before the start of the study enrollment period and prospectively for individuals with multiple visits during the 15-month enrollment period. Abstracted information was limited to that which is acquired in the normal course of clinical care. Participants were asked to record the disposition of the individual and to rate the likelihood of child physical abuse based on a previously published 7-point ordinal scale (Lindberg et al., 2008; see Table 1). Participants included the disposition of each individual in one of Table 1 Rating scale for perceived abuse likelihood. Number 1 2 3 4 5 6 7
Summary statement Definitely not inflicted injury No concern for inflicted injury Mild concern for inflicted injury Intermediate concern for inflicted injury Very concerning for inflicted injury Substantial evidence of inflicted injury Definite inflicted injury
Please cite this article in press as: Martindale, J., et al. Recurrent concerns for child abuse: Repeated consultations by a subspecialty child abuse team. Child Abuse & Neglect (2014), http://dx.doi.org/10.1016/j.chiabu.2014.03.007
G Model CHIABU-2752; No. of Pages 8
ARTICLE IN PRESS J. Martindale et al. / Child Abuse & Neglect xxx (2014) xxx–xxx
3
Fig. 1. Subject flow. Of the 2890 subjects included in the ExSTRA study, 161 underwent more than one consultation. This study focused on the 102 consultations (including 1 subject with 2 repeated consultations) obtained for recurrent concerns for abuse in the index child.
the following mutually exclusive categories: original home-no restrictions, original home-with restrictions, relative’s home, unrelated home, deceased, and unknown. Data analysis We compared demographics between children with single consultations and children with repeated consultations using the chi-squared test for dichotomous variables and t-tests for continuous variables. For analysis of the change in perceived likelihood of abuse, a low likelihood was considered to be ≤4 and a high level of concern to be ≥5 on the 7-point scale. The ExSTRA research network did not record dates of hospital admission or initial consultation. Minimum follow-up time was calculated by determining the difference between the time when a individual was initially entered into the data collection system and the end of the enrollment period. Qualitative analysis A qualitative analysis of repeated child abuse evaluations was performed in recognition of the complexity of child abuse evaluation and decision-making. The study authors took an inductive approach to develop a list of possible relevant themes as they emerged while reviewing individual case records (Bradley, Curry, & Devers, 2007). These themes were intended to describe factors associated with repeated abuse and to categorize those children in whom abuse may not have been identified despite repeated consultation. Each study author independently reviewed each case to increase the breadth and depth of coding individual cases. Summary findings were shared among the authors and records were iteratively reviewed to refine definitions of existing themes and identify additional themes until thematic saturation was reached. A theme was ultimately determined to apply to a case if it was identified by 3 of 5 independent reviewers. Themes were not mutually exclusive and some cases had multiple themes identified by a majority of investigators. Results This study included 2890 children enrolled in the ExSTRA study (see Fig. 1). One individual was removed from the sample because incomplete information made it unclear if the visit was part of an initial or repeat consultation. Mean minimum follow-up time was 189 days (standard deviation ± 144 days). A total of 102 (3.5%) consultations were for recurrent concerns for physical abuse. The initial visit for 30 of these consultations occurred during the study enrollment period. One individual underwent three separate consultations for abuse. Repeat visits occurred from two days to 72 months (median duration between consultations = 6 months, IQR 3–16 months). Demographic information appears in Table 2. No statistically significant demographic differences were noted between children evaluated once and those evaluated on multiple occasions.
Please cite this article in press as: Martindale, J., et al. Recurrent concerns for child abuse: Repeated consultations by a subspecialty child abuse team. Child Abuse & Neglect (2014), http://dx.doi.org/10.1016/j.chiabu.2014.03.007
G Model
ARTICLE IN PRESS
CHIABU-2752; No. of Pages 8
J. Martindale et al. / Child Abuse & Neglect xxx (2014) xxx–xxx
4
Table 2 Baseline characteristics of subjects with single and repeated consultations. Repeated consultations N Male (%) Racea White Black Other Unknown/not reported Ethnicity Hispanic Non-Hispanic Unknown/not reported Median age in months (IQR) a
Single
101a 53 (52%)
2728 1597 (58%)
63 (62%) 26 (26%) 9 (9%) 3 (3%)
1723 (63%) 611 (22%) 249 (9%) 145 (5%)
29 (29%) 64 (63%) 8 (8%) Initial consult 15 (6.5–29.0)
620 (22%) 1946 (71%) 162 (5%) Repeat consult 25 (12.7–48.5)
10 (4.0–30.1)
101 subjects underwent a total of 102 consultations (1 subject with 2 repeated consultations).
Table 3 Comparison of injury patterns. Injury N (%)
Initial N = 102
Cutaneous Bruises Burns Fracture Thoracic/Abdominal TBI Skull fracture RH Death
31 (31) 23 (23) 2 (2) 15 (15) 0 (0) 13 (13) 13 (13) 1 (1) n/a
Repeat N = 102
Single Visit N = 2728
65 (64) 49 (48) 11 (11) 15 (15) 2 (2) 6 (6) 5 (5) 2 (2) 4 (4)
1503 (55) 1005 (37) 210 (8) 1144 (42) 92 (3) 568 (21) 443 (16) 246 (9) 69 (3)
OR Initial vs. Single Consult (95% CI) 0.36 (0.23–0.56) 0.50 (0.30–0.82) 0.24 (0.04–0.10) 0.24 (0.13–0.43) 0 (0–1.37) 0.55 (0.29–1.03) 0.75 (0.40–1.40) 0.10 (0.01–0.66) n/a
Injuries Cutaneous injuries were the most common type of injury identified regardless of the number of presentations (Table 3). Fractures occurred in a higher percentage of individuals in the single consultation subgroup (42%, 95% CI [40, 44]) than in those individuals on initial (15%, 95% CI [8, 23]) and repeated (15%, 95% CI [8, 23]) visits. The incidence of death was 4% (95% CI [1, 9]) in individuals who had been evaluated previously and 3% (95% CI [2, 3]) in those with single visits. Although the ExSTRA research network included only individuals who were evaluated for concerns of physical abuse, investigators identified 28 individuals that had previously been evaluated by their hospital-based child abuse teams for other concerns, including neglect (n = 14) and sexual abuse (n = 14). Perceived likelihood of abuse For individuals with multiple visits, median level of perceived abuse likelihood was 4 (intermediate) on initial visit and 5 (very concerning) on repeat visit. Perceived likelihood of abuse remained low in 33% (95% CI [23, 44]) of individuals, remained high in 24.2% (95% CI [16, 34]) of individuals, went from low to high in 16.5% (95% CI [10, 26]), and high to low in 26% (95% CI [18, 37]) of individuals (see Table 4). Table 4 Perceived likelihood of abuse ratings on initial and repeat consultation. Level of Concern on Repeat Consult
Level of concern on initial consult
1 2 3 4 5 6 7
1
2
3
4
5
6
7
1 1 0 1 1 0 1
0 3 6 2 0 2 3
0 3 3 3 1 2 4
1 2 2 2 3 4 3
0 2 0 1 7 1 4
0 0 0 2 3 3 0
1 6 2 1 0 0 4
Each cell contains the number of subjects with repeat consultations for the given perceived level of concern for abuse on the initial consultation (rows) and repeat consultation (columns). Darker shading represents cases we have coded as having a low level of concern on both initial and subsequent consultations. Lighter shading represents cases with a high level of concern for abuse on initial and subsequent consultation.
Please cite this article in press as: Martindale, J., et al. Recurrent concerns for child abuse: Repeated consultations by a subspecialty child abuse team. Child Abuse & Neglect (2014), http://dx.doi.org/10.1016/j.chiabu.2014.03.007
G Model
ARTICLE IN PRESS
CHIABU-2752; No. of Pages 8
J. Martindale et al. / Child Abuse & Neglect xxx (2014) xxx–xxx
5
Table 5 Comparison of disposition resulting from consultation.
Deceased Original home: no restrictions Original home: with restrictions Relative home Unrelated home Unknown
Initial consult
Repeat consult
OR initial vs. repeat consult (95% CI)
Single consult
n/a 38 12 12 16 11
4 31 13 17 29 8
n/a 1.43 (0.76–2.67) 0.91 (0.37–2.28) 0.69 (0.29–1.64) 0.49 (0.23–1.02) 0.49 (0.52–4.23)
78 1169 350 436 479 216
OR initial vs. single consult (95% CI)
OR repeat vs. single consult (95% CI)
0.83 (0.54–1.28) 0.94 (0.48–1.78) 0.73 (0.37–1.38) 0.91 (0.51–1.60) 1.45 (0.72–2.89)
1.38 (0.42–4.04) 0.66 (0.41–1.04) 0.99 (0.52–1.85) 1.05 (0.60–1.83) 1.87 (1.17–2.96) 0.99 (0.44–2.14)
Disposition Individuals were more likely to be sent to an unrelated home (OR = 1.87; 95% CI [1.17, 2.96]) after a repeat consultation than those individuals who were evaluated only once (see Table 5). There were no statistically significant differences in disposition after initial and repeated consultations. Themes We identified seven themes that occurred in multiple cases when children underwent more than one evaluation for abuse. Ninety-four cases were determined to have at least one of the identified themes. The other cases included five without sufficient information to determine whether themes applied and three in which no theme was identified by a majority of authors. Child returned to the same environment. The most common factor related to repeated child abuse consultation was a child that was abused after returning to the same environment from which the initial concern for abuse arose. This finding was noted in 36 cases. The initial level of perceived abuse likelihood in 19 of these cases (53%) was 5 (very concerning) or higher. Example 1. A 36-month-old female was seen in a clinic for bruising, abrasions, and a chipped tooth (level of perceived abuse likelihood: 4) before being returned home without restrictions. Four days later, while reportedly left unattended by the mother’s boyfriend, the child was found facedown in bathwater. She was pulseless and apneic when emergency medical services arrived. The level of concern on the subsequent visit was 5. Safety plan not followed. Recurrent maltreatment occurred in seven individuals for whom a safety plan had been established but had not been diligently followed. Abuse in these individuals occurred during an unsupervised visit with the perpetrator or while living in a home from which a child was presumed to have been removed. Example 2. Child abuse specialists initially evaluated a 13-month-old boy for patterned facial injury and he was removed from his mother’s care and placed with his maternal grandmother. Six months later, a second child abuse consult was made to evaluate cutaneous bruising; Care was transferred then to the maternal grandfather. At approximately 20 months of age, this individual received medical evaluation for left-sided hemiparesis attributed to subdural hemorrhage and ischemic stroke one week after reported fall. At this time, the child’s injuries were not initially thought to be the result of abuse and the child abuse team was not re-consulted. A third child abuse consultation was requested and performed when the individual arrived in the emergency department, pulseless and unresponsive. Autopsy revealed acute subdural hemorrhage, cerebral edema, optic nerve sheath hemorrhage, cutaneous injury to the upper lip, and mesenteric tears. It was discovered after this individual’s death that the paternal grandfather returned the individual to the mother’s care without informing CPS. Abuse occurred in a different environment. Recurrent maltreatment was thought to occur in eight children who had been removed from one abusive setting and placed in another. Four of these children were evaluated for injuries that occurred in a foster home. Example 3. A male infant was placed in a foster home after evaluation for a femur fracture. At 28 months, he was evaluated for multiple burns and more than ten bruises. Repeat skeletal survey revealed multiple bilateral rib fractures. Injuries were thought to have occurred in the foster home. He was then placed in another foster home. Concern for abuse remained low. In 25 cases, the perceived likelihood of abuse on both initial and repeat visits was determined to be low. In most of these cases, a low level of concern was assigned when mechanism matched injury pattern. We did not identify factors in these individuals that might have decreased the threshold to undertake evaluation. Example 4. A 28-month old girl was initially evaluated for a radial head fracture (initial perceived level of abuse likelihood: 3). Two months later she was evaluated for abrasions on her back that were thought to be consistent with abrasions from a non-abusive scrape from concrete (level of perceived abuse likelihood: 2). No concern for abuse on repeat evaluation. After an initial visit that raised a high level of concern for abuse in five children, child abuse teams determined that there was no concern for repeated abuse during a later evaluation. Please cite this article in press as: Martindale, J., et al. Recurrent concerns for child abuse: Repeated consultations by a subspecialty child abuse team. Child Abuse & Neglect (2014), http://dx.doi.org/10.1016/j.chiabu.2014.03.007
G Model CHIABU-2752; No. of Pages 8
ARTICLE IN PRESS J. Martindale et al. / Child Abuse & Neglect xxx (2014) xxx–xxx
6
Example 5. A 25-month-old female was evaluated for bruising to the right ear, scalp, and cheek and a left proximal humerus fracture. The initial perceived likelihood for abuse was seven and the child was sent to an unrelated home. She was evaluated 14 months later for persistent pain at the site of a distal ulnar fracture attributed to a witnessed fall at daycare. While the repeat evaluation was initiated because of concern by one physician that the child’s splint had been prematurely removed, the child’s primary care physician had given instructions that the splint could be removed. There was no concern for inflicted injury on this second visit. Complication of initial injury. A second child abuse consultation was obtained for two children who had radiologic findings consistent with the natural progression of subdural hematomas. Example 6. A 6.5-month-old boy was evaluated for traumatic brain injury and found to have a complex skull fracture, subdural hemorrhage with evidence of parenchymal shear injury and multiple healing rib fractures. Perceived likelihood of abuse was rated as seven and the patient was placed in a relative’s home. The child abuse team was consulted again when a new area of discrete subdural blood was identified within the resolving subdural hemorrhage on a scheduled follow-up MRI. This radiologic finding was thought not to represent a new traumatic injury. Initial evaluation not for physical abuse. Initial child abuse consultation was requested for reasons other than suspected physical abuse (neglect, alleged sexual abuse) in 28 of these cases. Example 7. A 13-month-old female was initially evaluated for possible sexual assault when blood-stained underwear was found by the mother. She suspected her boyfriend at the time. Ten months later, the patient was reportedly found by a different boyfriend in the same home to be pinned underneath a dollhouse. The patient was found to have a buckle fracture of the distal femoral metaphysics and bruises on the buttock, knee and lumbar-sacral region. The patient returned to the same home with restrictions. Discussion This article contains, to our knowledge, the first description of children who are repeatedly evaluated by a child abuse team. We found that 3.5% of children who were evaluated for possible physical abuse had more than one evaluation by a child abuse specialty team. This finding does not suggest that 3.5% of cases represent missed opportunities to prevent recurrent maltreatment. The perceived likelihood of abuse remained low or decreased after repeated evaluations in 59% of children for whom we had data about perceived abuse likelihood. The incidence of repeated consultations in this study is lower than repeated reports to CPS. This finding is likely related to the lower number of injuries for which medical evaluation is sought and the limited time period during which individuals were followed. Children in this study who returned for care were most commonly found to have cutaneous injuries on initial evaluation. Because accidental bruising is common, it is often thought to require only minor force. However, certain types of bruising such as facial bruising, bruising in infants who are not yet independently mobile, abdominal bruising, or bruising to the ear may represent red flags that imply a violent home environment. In general, individuals who returned were less likely to have severe injury (retinal hemorrhages, fractures) than those seen only once. More serious injuries may have led to more aggressive interventions to protect children from re-abuse. Alternatively, children with an initial report of abuse may have had closer monitoring for signs of abuse and may have had recurrent referrals with more subtle findings than children who had not previously been referred. Several types of potential failures and compliance issues within the systems built to ensure child safety were identified. These potential failures to protect children occurred either within the hospital or once the child had been discharged with CPS oversight or custody. The most common theme among those individuals who underwent repeat consultation was repeated abuse after returning to the environment where abuse was first suspected (36 cases). Recidivism has been previously described in caregivers that continue to care for abused children (Alexander et al., 1990; Dakil et al., 2011; Ellaway et al., 2004). Recurrent maltreatment in children remaining in parental custody ranges from 17% to 35% (Dakil et al., 2011; Fluke, 2008). In seven of these 36 cases, the child abuse team’s initial perceived likelihood of inflicted injury was low, or was not recorded. These cases may reflect misattribution of injuries to accidental trauma by child abuse specialists. In 16 of 36 cases, the child abuse team expressed a high, perceived likelihood of abuse. These cases might represent opportunities for CPS systems to improve risk assessment or collaboration with child abuse teams. Repeated abuse also occurred in the context of a safety plan or intervention that was not closely followed. In several cases, injuries were observed immediately after visitation with parents. In other cases, children were later discovered to be cared for by, and residing with, initial perpetrators rather than the home and caregivers designated by CPS. Repeated abuse occurred in the setting of a new foster home in four children. Children in the foster care system are at higher risk for physical abuse (DHHS, 2011). Return to an abusive environment, failed compliance with a safety plan, and abuse in foster care are themes this study identified as being associated with recurrent abuse, but these data, obtained with the benefit of hindsight, do not identify any single actor or agency solely responsible for missed opportunities to prevent recurrent abuse. Our study does not suggest any single, easy intervention. Implementation of aggressive safety planning for all abuse consultations with minor injuries would be implausible. Placement of children in foster care is not without cost to the child, family, and society; recurrent maltreatment also occurs in children placed outside of the home. Instead, we conclude that close and formal review of Please cite this article in press as: Martindale, J., et al. Recurrent concerns for child abuse: Repeated consultations by a subspecialty child abuse team. Child Abuse & Neglect (2014), http://dx.doi.org/10.1016/j.chiabu.2014.03.007
G Model CHIABU-2752; No. of Pages 8
ARTICLE IN PRESS J. Martindale et al. / Child Abuse & Neglect xxx (2014) xxx–xxx
7
children who are evaluated on multiple occasions by a child abuse team and in the end, are suspected of having repeated abuse, may be an opportunity for child abuse specialists to identify areas of system improvement and compliance. Study limitations This study has important limitations. Although our qualitative methodology may be useful for hypothesis generation, we are not able to determine the presence of each theme for the large denominator of individuals who did not return for subsequent evaluation. We are therefore unable to determine the test characteristics (sensitivity, predictive value) of each identified theme. Some data regarding initial consultation were obtained retrospectively and may therefore be individual to recall bias. For those individuals in whom an initial evaluation took place before ExSTRA enrollment, data collection relied on the memory of specialists participating in the ExSTRA study. The 3.5% of children who underwent repeat child abuse consultation is comprised only of those children who were evaluated more than once by the same child abuse team at a participating hospital. Some children categorized as having undergone single child abuse evaluation may have been evaluated by specialists at a non-participating institution. We are also limited by our method of counting repeat evaluations over time. An optimal approach would be to prospectively enroll a group of children referred with an initial concern for abuse and follow them for a long and pre-specified period. Because we were most interested in identifying factors associated with repeat evaluations, and to avoid logistical and ethical challenges with this approach, we chose to examine a cohort that had previously been enrolled, and we chose to include children whose initial evaluation occurred prior to the enrollment period. One major theme that was identified was that of children whose subsequent evaluation for physical abuse followed an initial consultation that was for neglect, sexual abuse, or another maltreatment type. Our results likely underestimate the significance of this phenomenon because the parent dataset included only individuals referred with concerns of physical abuse, rather than examining children referred for any maltreatment type. Our estimate of the minimal follow-up time is certainly an underestimate because the date of the initial consultation was not abstracted; Data was entered for some children after an unmeasured delay. Finally, children who underwent repeat evaluation after the enrollment period would not be counted. Implications and future research This study finds that child abuse teams are consulted to evaluate an important minority of individuals on more than one occasion. This population may reflect missed opportunities to effectively intervene when a suspicion for child abuse had already been declared. This study may generate hypotheses that direct prospective investigations into secondary prevention of child abuse and inform future efforts to improve communication between child abuse specialists and child protection services. Acknowledgements The Examining Siblings To Recognize Abuse (ExSTRA) investigators are: Deb Bretl, APNP (Children’s Hospital Wisconsin, Wauwatosa, WI), Nancy Harper, MD (Driscoll Children’s Hospital, Corpus Cristi, TX), Katherine Deye, MD (Children’s National Medical Center, Washington, DC), Antoinette L. Laskey, MD and Tara Harris, MD (Riley Hospital for Children, Indianapolis, IN), Yolanda Duralde, MD (Mary Bridge Children’s Health Center, Tacoma, WA), Marcella Donaruma-Kwoh, MD (Texas Children’s Hospital, Houston, TX), Daryl Steiner, DO (Akron Children’s Hospital, Akron, OH), Ken Feldman, MD (Seattle Children’s Hospital, Seattle, WA), Kimberly Schwartz, MD (University of Massachusetts Medical Center, Worcester, MA), Robert A. Shapiro, MD and Mary Greiner, MD (Cincinnati Children’s Hospital Medical Center, Cincinnati, OH), Ivone Kim, MD (Children’s Hospital Pittsburgh or University of Pittsburgh Medical Center), Kent Hymel, MD (Dartmouth-Hitchcock Medical Center, Lebanon, NH), Suzanne Haney, MD (Children’s Hospital & Medical Center, Omaha, NE), Alicia Pekarsky, MD (SUNY Upstate Medical University, Syracuse, NY), Andrea Asnes, MD (Yale-New Haven Children’s Hospital, New Haven, CT), Paul McPherson, MD (Akron Children’s Hospital, Youngstown, OH), Neha Mehta, MD (Sunrise Children’s Hospital, Las Vegas, NV), and Gwendolyn Gladstone, MD (Exeter Pediatric Associates, Exeter, NH). References Alexander, R., Crabbe, L., Sato, Y., Smith, W., & Bennett, T. (1990). Serial abuse in children who are shaken. The American Journal of Diseases of Children, 144, 58–60. Block, R. (1998). What doctors who perform child abuse medical evaluations want the rest of the team to know. Journal of the Oklahoma State Medical Association, 91(8), 457. Block, R. W., & Palusci, V. J. (2006). Child abuse pediatrics: A new pediatric subspecialty. Journal of Pediatrics, 148, 711–712. Bradley, E. H., Curry, L. A., & Devers, K. J. (2007). Qualitative data analysis for health services research: Developing taxonomy, themes, and theory. Health Services Research, 42, 1758–1772. Dakil, S. R., Sakai, C., Lin, H., & Flores, G. (2011). Recidivism in the child protection system: Identifying children at greatest risk of reabuse among those remaining in the home. Archives of Pediatrics and Adolescent Medicine, 165, 1006–1012.
Please cite this article in press as: Martindale, J., et al. Recurrent concerns for child abuse: Repeated consultations by a subspecialty child abuse team. Child Abuse & Neglect (2014), http://dx.doi.org/10.1016/j.chiabu.2014.03.007
G Model CHIABU-2752; No. of Pages 8
8
ARTICLE IN PRESS J. Martindale et al. / Child Abuse & Neglect xxx (2014) xxx–xxx
Deans, K. J., Thackeray, J., Askegard-Giesmann, J. R., Earley, E., Groner, J. I., & Minneci, P. C. (2013). Mortality increases with recurrent episodes of nonaccidental trauma in children. Journal of Trauma and Acute Care Surgery, 75, 161–165. Ellaway, B. A., Payne, E. H., Rolfe, K., Dunstan, F. D., Kemp, A. M., Butler, I., & Sibert, J. R. (2004). Are abused babies protected from further abuse? Archives of Disease in Childhood, 89, 845–846. Ewing-Cobbs, L., Kramer, L., Prasad, M., Canales, D. N., Louis, P. T., Fletcher, J. M., Vollero, H., Landry, S. H., & Cheung, K. (1998). Neuroimaging, physical, and developmental findings after inflicted and noninflicted traumatic brain injury in young children. Pediatrics, 102(2 (Pt. 1)), 300–307. Flaherty, E. G., Sege, R. D., Griffith, J., Price, L. L., Wasserman, R., Slora, E., & Binns, H. J. (2008). From suspicion of physical child abuse to reporting: Primary care clinician decision-making. Pediatrics, 122, 611–619. Fluke, J. D. (2008). Child protective services rereporting and recurrence – Context and considerations regarding research. Child Abuse & Neglect, 32, 749–751. Jenny, C., Hymel, K. P., Ritzen, A., Reinert, S. E., & Hay, T. C. (1999). Analysis of missed cases of abusive head trauma. JAMA, 281, 621–626. King, W. K., Kiesel, E. L., & Simon, H. K. (2006). Child abuse fatalities: Are we missing opportunities for intervention? Pediatric Emergency Care, 22, 211–214. Kohl, P. L., Jonson-Reid, M., & Drake, B. (2009). Time to leave substantiation behind: Findings from a national probability study. Child Maltreatment, 14, 17–26. Laskey, A. L., Sheridan, M. J., & Hymel, K. P. (2007). Physicians’ initial forensic impressions of hypothetical cases of pediatric traumatic brain injury. Child Abuse & Neglect, 31, 329–342. Levi, B. H., & Brown, G. (2005). Reasonable suspicion: A study of Pennsylvania pediatricians regarding child abuse. Pediatrics, 116, e5–e12. Levi, B. H., Brown, G., & Erb, C. (2006). Reasonable suspicion: A pilot study of pediatric residents. Child Abuse & Neglect, 30, 345–356. Lindberg, D. M., Lindsell, C. J., & Shapiro, R. A. (2008). Variability in expert assessments of child physical abuse likelihood. Pediatrics, 121, e945–e953. Lindberg, D. M., Shapiro, R. A., Laskey, A. L., Pallin, D. J., Blood, E. A., & Berger, R. P. (2012). Prevalence of abusive injuries in siblings and household contacts of physically abused children. Pediatrics, 130, 193–201. Oral, R., Yagmur, F., Nashelsky, M., Turkmen, M., & Kirby, P. (2008). Fatal abusive head trauma cases: Consequence of medical staff missing milder forms of physical abuse. Pediatric Emergency Care, 24, 816–821. Ravichandiran, N., Schuh, S., Bejuk, M., Al-Harthy, N., Shouldice, M., Au, H., & Boutis, K. (2010). Delayed identification of pediatric abuse-related fractures. Pediatrics, 125, 60. Ricci, L., Giantris, A., Merriam, P., Hodge, S., & Doyle, T. (2003). Abusive head trauma in Maine infants: Medical, child protective, and law enforcement analysis. Child Abuse & Neglect, 27, 271–283. Rubin, D. M., Christian, C. W., Bilaniuk, L. T., Zazyczny, K. A., & Durbin, D. R. (2003). Occult head injury in high-risk abused children. Pediatrics, 111, 1382–1386. Southall, D. P., Plunkett, M. C., Banks, M. W., Falkov, A. F., & Samuels, M. P. (1997). Covert video recordings of life-threatening child abuse: Lessons for child protection. Pediatrics, 100, 735–760. Thackeray, J. D. (2007). Frena tears and abusive head injury: A cautionary tale. Pediatric Emergency Care, 23, 735–737. U.S. Department of Health and Human Services, Administration for Children and Families, Children’s Bureau. (2011). Child Maltreatment 2011. Washington, DC: Author.
Please cite this article in press as: Martindale, J., et al. Recurrent concerns for child abuse: Repeated consultations by a subspecialty child abuse team. Child Abuse & Neglect (2014), http://dx.doi.org/10.1016/j.chiabu.2014.03.007
