An interventional study of Application of a validated Clinical Decision Rule (CDR) as an AHT screening tool in Pediatric Abusive Head Trauma, sponsored by Milton S. Hershey Medical Center. Completed at 8 sites in United States. Open to participants aged Up to 3 Years. Per ClinicalTrials.gov, last updated 2020-12-16.
Sponsored by Milton S. Hershey Medical Center · Not applicable, Interventional, and Screening
To increase the accuracy of doctors' decisions to launch or forgo child abuse evaluations in their young, acutely head-injured patients, investigators have derived and validated a clinical decision rule (CDR) that detects abusive head trauma (AHT) with 96% sensitivity in pediatric intensive care unit (PICU) settings. This "CDR Implementation Trial" across eight PICU sites will assess the CDR's actual impact on AHT screening accuracy, identify factors associated with maximal physician acceptance and application of this novel AHT screening tool, and assess the sustainability of active CDR implementation strategies.
Investigators' long-term goal is to increase the accuracy of doctors' decisions to launch or forgo child abuse evaluations in their young, acutely head-injured patients. To this end, PediBIRN investigators have derived and validated a 4-variable clinical decision rule (CDR) that detects abusive head trauma (AHT) with 96% sensitivity in PICU settings. Applied at PICU admission, the CDR categorizes young, acutely head-injured patients as higher risk vs. lower risk, and recommends thorough abuse evaluations for all higher risk patients.
The "CDR Implementation Trial" across eight PICUs will assess the CDR's actual impact on AHT screening accuracy. The stratified cluster randomized trial design will facilitate direct comparison of child abuse evaluations at four, randomly selected, control sites to four matched intervention sites, where investigators will deploy active, multifaceted, implementation strategies designed to promote CDR acceptability and application. These strategies will include physician training with onsite visits, monthly "booster training emails," access to an "AHT probability calculator," audit and site-specific feedback, and local "information sharing sessions" designed to address local barriers to CDR acceptance and application.
PediBIRN investigators will conduct the CDR Implementation Trial with three Specific Aims. Aim 1 is to assess the CDR's actual impact on AHT screening accuracy. Investigators hypothesize that deployment of CDR implementation strategies at the four intervention sites will be associated with higher percentages of higher risk patients evaluated thoroughly for abuse, and lower percentages of lower risk patients evaluated (even partially) for abuse. Aim 2 is to identify factors that impact CDR application in PICU settings. Investigators hypothesize that PICUs with higher patient volumes, providers with child abuse expertise, and providers with more intense exposure to CDR implementation strategies will be predictive of higher percentages of higher risk patients thoroughly evaluated for abuse, whereas patients of minority race or ethnicity will be predictive of higher percentages of lower risk patients evaluated for abuse. Investigators' third Exploratory Aim is to measure the sustained impacts of CDR implementation strategies. Investigators hypothesize that CDR utilization at intervention sites will be sustained twelve months after CDR implementation strategies have been discontinued.
Based on strong Preliminary Studies, investigators predict that CDR adoption as an AHT screening tool will increase AHT detection; reduce overall abuse evaluations and their associated risks; reduce unwarranted variation in current AHT screening practices; minimize the adverse impacts of doctors' inherent biases, uncertainty, and practice disparities; reduce AHT-associated acute health care costs in PICU settings; and save the lives of children who will be reinjured and killed if their AHT is missed or unrecognized.
253 studies on the registry are indexed under Craniocerebral Trauma; 45 are open to participants now.
This study's enrollment of 420 is above the median of 60 across 137 interventional studies indexed under Craniocerebral Trauma.
Browse Craniocerebral Trauma studies →Milton S. Hershey Medical Center is the lead sponsor of 480 studies on the registry; 61 are open to participants now.
Of its 56 completed or terminated interventional studies of FDA-regulated products, 42 (75%) have results posted.
Counted across the registry records on this site, refreshed daily.
Exclusion Criteria:
At the four intervention sites, investigators will deploy active, multifaceted, implementation strategies designed to promote CDR acceptability and application as an AHT screening tool. These strategies will include physician training with onsite visits, monthly "booster training emails," access to an "AHT probability calculator," audit and site-specific feedback, and local "information sharing sessions" designed to address local barriers to CDR acceptance and application.
Other: Application of a validated Clinical Decision Rule (CDR) as an AHT screening tool
At the four matched control sites, physicians will engage in "AHT screening as usual."
The Clinical Decision Rule (CDR) for AHT reads as follows: Every acutely head-injured infant or young child hospitalized for intensive care presenting with any one or more of these four variables should be considered "high risk" and thoroughly evaluated for abuse: (1) any clinically significant respiratory compromise at the scene of injury, during transport, in the Emergency Department, or prior to admission; (2) Any bruising involving the child's ear(s), neck, or torso; (3) Any subdural hemorrhage(s) or fluid collection(s) that are bilateral OR involve the interhemispheric space; (4) Any skull fracture(s) other than an isolated, nondiastatic, linear, parietal, skull fracture.
The Number of Higher Risk Patients Evaluated Thoroughly for Abuse at Intervention vs. Control Sites
This outcome measure facilitates a comparison of the percentage of patients that the clinical decision rule stratified as higher risk who were evaluated thoroughly for abuse (with both skeletal survey and retinal exam) at intervention vs. control sites. We hypothesized that thorough evaluations of higher risk patients would be significantly higher at intervention sites.
Time frame: To be measured 32 months after the start of the clinical trial
The Number of Lower Risk Patients Evaluated Even Partially for Abuse at Intervention vs. Control Sites
This outcome measure facilitates a comparison of the percentage of patients that the clinical decision rule stratified as lower risk who were nevertheless evaluated at least partially for abuse (with skeletal survey and/or retinal examination) at intervention vs. control sites. We hypothesized that (partial or complete) abuse evaluations of lower risk patients would be significantly lower at intervention sites.
Time frame: To be measured 32 months after the start of the clinical trial
Estimated Rates (Percentages) of Missed AHT at Intervention vs. Control Sites
This outcome measures and compares estimated rates (percentages) of missed AHT (among all patients with AHT) at intervention vs. control sites. Using secondary outcome measures, it was calculated as \[estimated cases of missed AHT\] / \[estimated cases of missed AHT + patients with corroborating findings of abuse\]. We hypothesized that the estimated rate of missed AHT would be significantly lower at intervention sites. This outcome measure is best interpreted in the following contexts: (1) Applied accurately and consistently, the clinical decision rule's potential sensitivity for AHT is 96% (see references). That is, it should "miss" (categorize as lower risk) only 4% of AHT patients, and (2) We estimate that intervention and control site physicians "missed" 15% and 11% of their AHT patients, respectively, in prior PediBIRN studies (see the Post-Hoc Outcome "The Estimated Rate of Missed AHT at Intervention vs. Control Sites in Prior PediBIRN Studies").
Time frame: To be measured 32 months after the start of the clinical trial
The Number of Patients Evaluated at Least Partially for Abuse at Intervention vs. Control Sites
This outcome measure facilitates a comparison of the percentage of patients evaluated at least partially for abuse (with skeletal survey and/or retinal examination) at intervention vs. control sites. Thus, it facilitates a broad-based comparison of AHT evaluation practices at intervention vs. control sites.
Time frame: To be measured 32 months after the start of the clinical trial
The Number of Patients With Corroborating Findings of Abuse at Intervention vs. Control Sites (Aka Overall Diagnostic Yield)
This outcome measure facilitates a comparison of the percentage of patients whose completed skeletal surveys and/or retinal exams revealed findings considered moderately or highly specific for abuse at intervention vs. control sites. Thus, it is also a measure of the overall diagnostic yield of patients' completed skeletal surveys and retinal examinations.
Time frame: To be measured 32 months after the start of the clinical trial
The Number of Potential Cases of Missed AHT at Intervention vs. Control Sites
This outcome measure facilitates a comparison of the percentage of eligible patients who might be potential cases of missed AHT (that is, patients lacking skeletal survey and/or retinal exam, whose abuse evaluation is therefore incomplete) at intervention vs. control sites.
Time frame: To be measured 32 months after the start of the clinical trial
The Number of Estimated Patients With Missed AHT at Intervention vs. Control Sites
This outcome measure facilitates a comparison of the estimated percentage of patients with missed AHT (among potential cases of missed AHT) at intervention vs. control sites. It was calculated as \[potential cases of missed AHT\] x \[their mean estimate of abuse probability\]. The patient-specific estimates of abuse probability used to calculate the mean estimates were accessed by applying the 4-variable rule as a clinical prediction tool (rather than a directive decision rule).
Time frame: To be measured 32 months after the start of the clinical trial
Estimated Prevalence of AHT at Intervention vs. Control Sites
This outcome measure facilitates a comparison of the estimated prevalence of AHT (among all eligible patients) at intervention vs. control sites. It was calculated as \[patients with corroborating findings of abuse + estimated cases of missed AHT\] / \[all eligible patients in each arm of the trial\].
Time frame: To be measured 32 months after the start of the clinical trial
The Change (From Prior PediBIRN Studies to the Current Clinical Trial) in the Number of Higher Risk Patients Evaluated Thoroughly for Abuse at Intervention Sites
This outcome measure facilitates a comparison of the percentage of higher risk patients evaluated thoroughly for abuse (with skeletal survey AND retinal examination) at intervention sites in prior strictly observational PediBIRN studies vs. the current cluster randomized trial. It was calculated using precisely equivalent methods and data captured prospectively between 2010 and 2013 in comparable patient cohorts at the same four intervention sites.
Time frame: To be measured 32 months after the start of the clinical trial
The Change (From Prior PediBIRN Studies to the Current Clinical Trial) in the Number of Potential Cases of Missed AHT at Intervention Sites
This outcome measure facilitates a comparison of the percentage of potential cases of missed AHT (among all eligible patients) at intervention sites in prior strictly observational PediBIRN studies vs. the current cluster randomized trial. It was calculated using precisely equivalent methods and data captured prospectively between 2010 and 2013 in comparable patient cohorts at the same four intervention sites.
Time frame: To be measured 32 months after the start of the clinical trial
The Change (From Prior PediBIRN Studies to the Current Clinical Trial) in the Estimated Rate (Percentage) of Missed AHT at Intervention Sites
This outcome measure facilitates a comparison of the estimated rate (percentage) of missed AHT (among all patients with AHT) at intervention sites in prior strictly observational PediBIRN studies vs. the current cluster randomized trial. It was calculated using precisely equivalent methods and data captured prospectively between 2010 and 2013 in comparable patient cohorts at the same four intervention sites.
Time frame: To be measured 32 months after the start of the clinical trial
The Number of AHT Patients (Among All Patients With AHT) That the CDR Would Have Stratified as Higher Risk if the CDR Had Been Applied Accurately and Consistently (Aka the Clinical Decision Rule's Potential AHT Screening Sensitivity)
This outcome measure facilitates estimation of the clinical decision rule's potential AHT screening sensitivity IF it had been applied accurately and consistently across all eight participating sites. It was calculated based on the following assumptions: (1) All higher risk patients were evaluated thoroughly for abuse with skeletal survey AND retinal exam by an ophthalmologist; Therefore, all cases of AHT among higher risk patients were recognized, and (2) Abuse evaluations were deferred in all lower risk patients; Therefore, all cases of AHT among lower risk patients were missed or unrecognized.
Time frame: To be measured 32 months after the start of the clinical trial
Participating PICUs were stratified based on projected patient volumes, matched into pairs, and allocated randomly to intervention (n=4) or control (n=4) conditions. Eligible participants were acutely head injured patients \<3 years admitted for intensive care (excluding MVA victims and patients with pre-existing brain abnormalities). Prospective data capture at all 8 participating PICUs began 1 August 2017 and ended 31 March 2020. For patients, the study was strictly observational.
| Milestone | Intervention Sites | Control Sites |
|---|---|---|
| Started | 183 | 237 |
| Completed | 183 | 237 |
| Not completed | 0 | 0 |
This outcome measure facilitates a comparison of the percentage of patients that the clinical decision rule stratified as higher risk who were evaluated thoroughly for abuse (with both skeletal survey and retinal exam) at intervention vs. control sites. We hypothesized that thorough evaluations of higher risk patients would be significantly higher at intervention sites.
| Participants | Intervention Sites | Control Sites |
|---|---|---|
| The Number of Higher Risk Patients Evaluated Thoroughly for Abuse at Intervention vs. Control Sites | 128 | 128 |
This outcome measure facilitates a comparison of the percentage of patients that the clinical decision rule stratified as lower risk who were nevertheless evaluated at least partially for abuse (with skeletal survey and/or retinal examination) at intervention vs. control sites. We hypothesized that (partial or complete) abuse evaluations of lower risk patients would be significantly lower at intervention sites.
| Participants | Intervention Sites | Control Sites |
|---|---|---|
| The Number of Lower Risk Patients Evaluated Even Partially for Abuse at Intervention vs. Control Sites | 17 | 37 |
This outcome measures and compares estimated rates (percentages) of missed AHT (among all patients with AHT) at intervention vs. control sites. Using secondary outcome measures, it was calculated as \[estimated cases of missed AHT\] / \[estimated cases of missed AHT + patients with corroborating findings of abuse\]. We hypothesized that the estimated rate of missed AHT would be significantly lower at intervention sites. This outcome measure is best interpreted in the following contexts: (1) Applied accurately and consistently, the clinical decision rule's potential sensitivity for AHT is 96% (see references). That is, it should "miss" (categorize as lower risk) only 4% of AHT patients, and (2) We estimate that intervention and control site physicians "missed" 15% and 11% of their AHT patients, respectively, in prior PediBIRN studies (see the Post-Hoc Outcome "The Estimated Rate of Missed AHT at Intervention vs. Control Sites in Prior PediBIRN Studies").
| Participants | Intervention Sites | Control Sites |
|---|---|---|
| Estimated Rates (Percentages) of Missed AHT at Intervention vs. Control Sites | 6 | 14 |
This outcome measure facilitates a comparison of the percentage of patients evaluated at least partially for abuse (with skeletal survey and/or retinal examination) at intervention vs. control sites. Thus, it facilitates a broad-based comparison of AHT evaluation practices at intervention vs. control sites.
| Participants | Intervention Sites | Control Sites |
|---|---|---|
| The Number of Patients Evaluated at Least Partially for Abuse at Intervention vs. Control Sites | 163 | 189 |
This outcome measure facilitates a comparison of the percentage of patients whose completed skeletal surveys and/or retinal exams revealed findings considered moderately or highly specific for abuse at intervention vs. control sites. Thus, it is also a measure of the overall diagnostic yield of patients' completed skeletal surveys and retinal examinations.
| Participants | Intervention Sites | Control Sites |
|---|---|---|
| The Number of Patients With Corroborating Findings of Abuse at Intervention vs. Control Sites (Aka Overall Diagnostic Yield) | 75 | 93 |
This outcome measure facilitates a comparison of the percentage of eligible patients who might be potential cases of missed AHT (that is, patients lacking skeletal survey and/or retinal exam, whose abuse evaluation is therefore incomplete) at intervention vs. control sites.
| Participants | Intervention Sites | Control Sites |
|---|---|---|
| The Number of Potential Cases of Missed AHT at Intervention vs. Control Sites | 38 | 75 |
This outcome measure facilitates a comparison of the estimated percentage of patients with missed AHT (among potential cases of missed AHT) at intervention vs. control sites. It was calculated as \[potential cases of missed AHT\] x \[their mean estimate of abuse probability\]. The patient-specific estimates of abuse probability used to calculate the mean estimates were accessed by applying the 4-variable rule as a clinical prediction tool (rather than a directive decision rule).
| Participants | Intervention Sites | Control Sites |
|---|---|---|
| The Number of Estimated Patients With Missed AHT at Intervention vs. Control Sites | 6 | 14 |
This outcome measure facilitates a comparison of the estimated prevalence of AHT (among all eligible patients) at intervention vs. control sites. It was calculated as \[patients with corroborating findings of abuse + estimated cases of missed AHT\] / \[all eligible patients in each arm of the trial\].
| Participants | Intervention Sites | Control Sites |
|---|---|---|
| Estimated Prevalence of AHT at Intervention vs. Control Sites | 81 | 107 |
This outcome measure facilitates a comparison of the percentage of higher risk patients evaluated thoroughly for abuse (with skeletal survey AND retinal examination) at intervention sites in prior strictly observational PediBIRN studies vs. the current cluster randomized trial. It was calculated using precisely equivalent methods and data captured prospectively between 2010 and 2013 in comparable patient cohorts at the same four intervention sites.
| Participants | Intervention Sites at Baseline | Intervention Sites During the Cluster Randomized Trial |
|---|---|---|
| The Change (From Prior PediBIRN Studies to the Current Clinical Trial) in the Number of Higher Risk Patients Evaluated Thoroughly for Abuse at Intervention Sites | 89 | 128 |
This outcome measure facilitates a comparison of the percentage of potential cases of missed AHT (among all eligible patients) at intervention sites in prior strictly observational PediBIRN studies vs. the current cluster randomized trial. It was calculated using precisely equivalent methods and data captured prospectively between 2010 and 2013 in comparable patient cohorts at the same four intervention sites.
| Participants | Intervention Sites at Baseline | Intervention Sites During the Cluster Randomized Trial |
|---|---|---|
| The Change (From Prior PediBIRN Studies to the Current Clinical Trial) in the Number of Potential Cases of Missed AHT at Intervention Sites | 68 | 38 |
This outcome measure facilitates a comparison of the estimated rate (percentage) of missed AHT (among all patients with AHT) at intervention sites in prior strictly observational PediBIRN studies vs. the current cluster randomized trial. It was calculated using precisely equivalent methods and data captured prospectively between 2010 and 2013 in comparable patient cohorts at the same four intervention sites.
| Participants | Intervention Sites at Baseline | Intervention Sites During the Cluster Randomized Trial |
|---|---|---|
| The Change (From Prior PediBIRN Studies to the Current Clinical Trial) in the Estimated Rate (Percentage) of Missed AHT at Intervention Sites | 11 | 6 |
This outcome measure facilitates estimation of the clinical decision rule's potential AHT screening sensitivity IF it had been applied accurately and consistently across all eight participating sites. It was calculated based on the following assumptions: (1) All higher risk patients were evaluated thoroughly for abuse with skeletal survey AND retinal exam by an ophthalmologist; Therefore, all cases of AHT among higher risk patients were recognized, and (2) Abuse evaluations were deferred in all lower risk patients; Therefore, all cases of AHT among lower risk patients were missed or unrecognized.
| Participants | All Eight Participating Sites |
|---|---|
| The Number of AHT Patients (Among All Patients With AHT) That the CDR Would Have Stratified as Higher Risk if the CDR Had Been Applied Accurately and Consistently (Aka the Clinical Decision Rule's Potential AHT Screening Sensitivity) | 181 |
This outcome facilitates comparison of the estimated rates (percentages) of missed AHT (among all patients with AHT) at intervention vs. control sites in prior strictly observational PediBIRN studies. It was calculated using precisely equivalent methods and data captured in comparable patient cohorts at the same eight sites between 2010 and 2013. The results provide context for interpreting estimated rates of missed AHT during the subsequent cluster randomized trial.
| Participants | Intervention Sites | Control Sites |
|---|---|---|
| The Estimated Rate (Percentage) of Missed AHT at Intervention vs. Control Sites in Prior PediBIRN Studies | 11 | 7 |
This outcome measure facilitates comparison of the percentage of potential cases of missed AHT (among all eligible patients) in prior strictly observational PediBIRN studies at intervention vs. control sites. It was calculated using precisely equivalent methods and data captured in comparable patient cohorts at the same eight sites between 2010 and 2013. The results provide context for interpreting estimated rates of missed AHT during the subsequent cluster randomized trial.
| Participants | Intervention Sites | Control Sites |
|---|---|---|
| The Number of Potential Cases of Missed AHT at Intervention vs. Control Sites in Prior PediBIRN Studies | 68 | 48 |
Collected over 32 months. Non-serious events are listed at a 0% frequency threshold.
| Group | Deaths | Serious | Other |
|---|---|---|---|
| Intervention Sites | 0/183 (0%) | 0/183 (0%) | 0/183 (0%) |
| Control Sites | 0/237 (0%) | 0/237 (0%) | 0/237 (0%) |
| Age, Continuous(Months) | Intervention Sites | Control Sites | Total |
|---|---|---|---|
| Mean | 7.7 ± 7.2 | 8.7 ± 8.8 | 8.3 ± 8.1 |
| Sex: Female, Male(Participants) | Intervention Sites | Control Sites | Total |
|---|---|---|---|
| Female | 68 | 87 | 155 |
| Male | 115 | 150 | 265 |
| Ethnicity (NIH/OMB)(Participants) | Intervention Sites | Control Sites | Total |
|---|---|---|---|
| Hispanic or Latino | 21 | 88 | 109 |
| Not Hispanic or Latino | 142 | 127 | 269 |
| Unknown or Not Reported | 20 | 22 | 42 |
| Race (NIH/OMB)(Participants) | Intervention Sites | Control Sites | Total |
|---|---|---|---|
| American Indian or Alaska Native | 1 | 1 | 2 |
| Asian | 0 | 4 | 4 |
| Native Hawaiian or Other Pacific Islander | 3 | 0 | 3 |
| Black or African American | 43 | 37 | 80 |
| White | 99 | 174 | 273 |
| More than one race | 12 | 7 | 19 |
| Unknown or Not Reported | 25 | 14 | 39 |
| Region of Enrollment(participants) | Intervention Sites | Control Sites | Total |
|---|---|---|---|
| United States | 183 | 237 | 420 |
| Higher Risk(participants) | Intervention Sites | Control Sites | Total |
|---|---|---|---|
| Number | 158 | 175 | 333 |
| Lower Risk(participants) | Intervention Sites | Control Sites | Total |
|---|---|---|---|
| Number | 25 | 62 | 87 |
Documents are hosted by the registry — open the source record to download them.
Plan to share: No
This study is completed, as verified in Nov 2020. You cannot join it, but the record below documents what was studied.
Get an email when the registry record changes — status, dates, results — or when someone posts here.
Sign in to followQuestions and observations about this study, from anyone following it. Not medical advice, and not a channel to the study team — their contact details are on the registry record.
Sign in to join the discussion. Reading takes no account; posting does. You choose a display name, and a pseudonym is the default.
Nothing here yet. If you are running this trial, taking part in it, or weighing whether to, this is the place to say so.
Milton S. Hershey Medical Center