Final Research Plan

Prevention of Child Maltreatment: Primary Care Interventions

May 19, 2022

Recommendations made by the USPSTF are independent of the U.S. government. They should not be construed as an official position of the Agency for Healthcare Research and Quality or the U.S. Department of Health and Human Services.

  1. For children without obvious signs and symptoms of abuse or neglect, do primary care–feasible or referable preventive interventions reduce exposure to abuse or neglect; improve behavioral, developmental, emotional, physical, or mental health and well-being; or reduce mortality? Does the effectiveness of interventions differ by populations of interest (e.g., defined by child or caregiver characteristics such as age, developmental stage of the child, sex, gender identity, race and ethnicity, sociodemographic characteristics [rural/urban location, place of residence, or family income or wealth], or special healthcare needs)?
  2. What are the harms from interventions intended to prevent child maltreatment? Do the harms of interventions differ by populations of interest (e.g., defined by child or caregiver characteristics such as age, developmental stage of the child, sex, gender identity, race and ethnicity, sociodemographic characteristics [rural/urban location, place of residence, or family income or wealth], or special healthcare needs)?

Contextual Questions will not be systematically reviewed and are not shown in the Analytic Framework.

  1. What are current practices for a) identifying children at risk of maltreatment, b) referring children or families to prevention programs, c) reporting children or families to Child Protective Services, and d) diagnosing child maltreatment outcomes? Do current practices in identification, referral, reporting, and diagnosis of outcomes of child maltreatment differ by race or ethnicity of the child or caregiver? If evidence exists of practice differences, what factors might explain these differences?
  2. What are the validity and reliability of risk assessment tools to identify children and adolescents who are at risk of child maltreatment? Does the reported validity and reliability (of risk assessment tools) differ by race and ethnicity? If yes, what might explain these differences? Is there evidence that these tools alter or increase inequity?
  3. What are the effects of primary care–feasible or referable preventive interventions that report on child maltreatment outcomes on social determinants of health? Do primary care–feasible or referable preventive interventions that report on child maltreatment outcomes examine the association between social determinants of health and child maltreatment outcomes?

The Research Approach identifies the study characteristics and criteria that the Evidence-based Practice Center will use to search for publications and to determine whether identified studies should be included or excluded from the Evidence Review. Criteria are overarching as well as specific to each of the Key Questions.

Category Included Excluded
Population Children or adolescents (younger than age 18 years) with no known exposure to maltreatment or specific signs or symptoms of current or past maltreatment Symptomatic children and adolescents undergoing diagnostic evaluations for conditions related to abuse or neglect (e.g., those presenting with a broken bone or other signs of physical abuse or neglect, or trauma symptoms associated with domestic violence exposure), children with known exposure to child maltreatment and perpetrators of maltreatment, and children of caregivers who perpetrated maltreatment toward them
Interventions Primary care–based programs or services; services that could result from a referral by a primary care provider

Services may include home visiting programs, respite care, parent education programs, and family support and family strengthening programs

Services may be implemented by nonclinicians

(Interventions may be directed at the caregiver and may or may not include components directed at the child)
Communitywide interventions such as public awareness campaigns or public service announcements only, without specific interventions linked to clinical settings
Comparisons Usual care, delayed intervention, or active interventions that allow for the assessment of the independent contribution of primary care–relevant preventive intervention (e.g., clinical interventions + media campaigns vs. media campaigns) Comparators that do not allow for the assessment of the independent contribution of the effect of primary care–feasible or referable preventive interventions (e.g., clinical interventions + media campaigns vs. usual care)
Outcomes KQ 1: Direct or proxy measures of abuse or neglect (required):
  • Physical abuse, sexual abuse, or emotional abuse perpetrated by a parent or caretaker against a child (not parent reported)
  • Physical (e.g., failure to thrive), emotional, dental/medical (for needed dental, medical, or mental health treatment), or educational neglect
  • Child Protective Services reports
  • Removal of the child from the home
  • Injuries such as broken bones, bruises, burns, and other injuries with a high specificity for abuse
  • Emergency department visits
  • Hospitalizations

Behavioral, developmental, emotional, mental, or physical health and well-being*:

  • Quality of life or functional status measures (using validated instruments)
  • Internalizing behaviors: depression or anxiety
  • Externalizing behaviors: disruptive, aggressive, or delinquent behavior
  • Child development (including school readiness and academic performance): social-emotional (e.g., attachment problems, peer relationships, or community involvement), developmental delays (language or cognitive)
  • Incidence of reactive attachment disorder
  • Incidence of disinhibited social engagement disorder
  • Incidence of acute stress disorder
  • Incidence of posttraumatic stress disorder
  • Incidence of traumatic stress symptoms: attachment, self-regulation, under- or overcontrolling behaviors (e.g., irritable/angry outbursts, self-destructive behavior, or food hoarding), executive functioning, self-concept, hypervigilance, exaggerated startle response, dissociation, concentration problems, somatic problems (e.g., headaches or gastrointestinal problems), sleep disturbance, or nightmares
  • Unintended pregnancy, sexually transmitted diseases, or termination of pregnancy
  • Suicidality and self-injurious behaviors

KQ 2: Any harms that result as an effect of interventions (e.g., stigma, labeling, legal risks, risks of further harm to the child, dissolution of families, or worsening of inequities) or worsening of outcomes listed in KQ 1

KQ 1: Outcomes not otherwise specified, studies without direct or proxy measures of abuse of neglect, and parent-reported measures of exposure to abuse or neglect

KQ 2: None specified

Clinical settings Pediatrics, primary care, family medicine, school-based clinics, or other settings where primary care services are offered; services that could result from an assessment by a clinician (including delivery hospitals, in-home settings, and nonspecialist settings) Not a primary care–feasible or referable setting, populations or services/interventions not applicable to U.S. practice
Geographic setting Research conducted in the United States or in populations similar to U.S. populations with services and interventions applicable to U.S. practice (countries categorized as “very high” on the United Nations Human Development Index, as defined by the United Nations Development Programme) Research not relevant to the United States in countries categorized as less than “very high” on the Human Development Index
Study designs KQ 1: RCTs

KQ 2: RCTs, controlled clinical trials, cohorts with controls, and case-control studies

Systematic reviews will be hand searched for additional eligible studies

KQ 1: Systematic reviews, nonrandomized cohort trials, case-control, case series, and case studies

KQ 2: Systematic reviews, case series, and case studies

Timing Any timing No exclusion based on timing
Languages Full text published in English Non-English language
Publication type Original research and systematic reviews Editorials, commentaries, and narrative reviews

Abbreviations: KQ=key question; RCT=randomized, controlled trial.

*These outcomes will be evaluated in studies that also report at least one child maltreatment outcome. Studies that do not report at least one child maltreatment outcome are ineligible for this review.

The draft Research Plan was posted for public comment on the U.S. Preventive Services Task Force (USPSTF) website from February 17, 2022, to March 16, 2022. In response to comments, the USPSTF added greater specificity to the analytic framework and key questions 1 and 2 by adding “developmental health” as an outcome. The USPSTF also added the developmental stage of the child to populations of interest for key questions 1 and 2; specified that sociodemographic characteristics include rural/urban location, place of residence, and family income or wealth; and replaced “special needs” with “special healthcare needs.” The USPSTF added “worsening of inequities” as a potential harm in key question 2. The USPSTF clarified the language around the requirement that studies must report direct or proxy measures of child maltreatment to be eligible for analyses of other relevant child outcomes. Several reviewers requested greater specificity on the type of intervention or screening instrument and wanted more information on context or applicability of the study; the USPSTF will address these issues during abstraction and synthesis.

Reviewers suggested including social determinants of health as outcomes. The USPSTF added a contextual question to explore this issue. Although the inclusion of other intermediate outcomes such as parent-reported punishment and discipline measures was considered, direct and proxy measures of child maltreatment outcomes are included and expected to provider stronger links between interventions to prevent child maltreatment and child maltreatment outcomes. For this reason, the USPSTF is not including parent-reported punishment and discipline measures. Reviewers suggested including observational studies for benefits. Randomized, controlled trials offer the highest rigor for benefits. Observational studies have inherently higher potential risk of bias than randomized, clinical trials; as a result, they will be rated lower for quality and are not likely to expand the yield of robust evidence for decision making.