Commentary: Child maltreatment and physical health: a call to action.

The preceding section provides new and important information about the link between exposure to one or more types of child maltreatment and impairment that moves the field beyond the association with mental health problems. Despite differences in samples, measurement, and design, the message is consistent across these studies: child maltreatment is associated with a broad range of adverse physical health outcomes, as well as behaviors that increase risk for such outcomes. Using data from the Chicago Longitudinal Study (CLS), Topitzes, Mersky, and Reynolds (2010) examined the relationship between official reports of child maltreatment collected prospectively and self-report survey data of tobacco smoking by young adults, taking into account a broad range of mediators. In this sample of socially disadvantaged minority participants, official reports of child maltreatment predicted smoking in young adulthood in both genders. This relationship was mediated by several emotional, social, and cognitive factors in adolescence and young adulthood, as well as by school mobility. The study by Clark, Thatcher, and Martin (2010) surveyed a combined clinical and community sample of adolescents about a range of traumas, substance-use disorders and health-related symptoms, and administered a baseline comprehensive physical examination in addition to laboratory studies. At each assessment, health-related symptoms showed a strong association with trauma severity, a relationship mediated by anxiety. The authors concluded that some physical and laboratory findings in adolescence suggested stress-related responses. Lanier, Jonson-Reid, Stahlschmidt, Drake, and Constantino (2010) compared the risk of hospital-based care for asthma, cardio-respiratory illness and infections among maltreated and nonmaltreated groups of low-income children. Using longitudinal administrative data from state information systems and controlling for individual, family, and community factors, they found that children reported for maltreatment were at higher risk of hospital care for each of these conditions. Furthermore, recurrent reports of maltreatment were associated with a greater number of hospital care episodes for these illnesses. The meta-analysis by Irish, Kobayashi, and Delahanty (2010), examined the relationship between exposure to child sexual abuse and physical health outcomes taking into account sampling, measurement, and study design characteristics. A history of sexual abuse in childhood was associated with more physical health symptoms compared with control groups for the majority of outcomes. With one exception, these differences were larger for clinical than community groups. The overall coherence in findings across these studies is particularly notable, given the differences in approaches to measurement of child maltreatment: Clark et al. (2010) used a structured trauma interview to measure subtypes of abuse but did not include neglect; Irish et al. (2010) focused exclusively on studies of sexual abuse assessed by one or more of self-report questionnaire, interview, or chart review; and the remaining two investigations relied on official reports of maltreatment, although the CLS records did not include cases of emotional abuse. Some might conclude that the approach to measurement of child maltreatment does not matter much—the majority of studies show a link between exposure and impairment; in the case of these four studies, physical health outcomes. Nothing could be further from the truth—especially when