Page images

TABLE. Number and percentage of infants aged ≤1 week who were victims of substantiated maltreatment,* by type of maltreatment and source of report — National Child Abuse and Neglect Data System, United States, October 2005-September 2006†

[blocks in formation]


*Defined as maltreatment by a parent or other caregiver deemed to have occurred after thorough investigation by a qualified staff member from a chic protective services agency with jurisdiction over the geographic area in which the maltreatment took place. Additional information available a

Data from five states (Alaska, Maryland, North Dakota, Pennsylvania, and Vermont) were not available for analysis.

Includes deprivation of necessities and medical neglect.

Includes infants who were victims of more than one type of maltreatment.

(19,574 [21.4%]), social services personnel (13,740 [15.1%]), parents/other relatives (8,058 [8.8%]), and friends/neighbors (2,927 [3.2%]).

Reported by: ML Brodowski, MSW, MPH, CM Nolan, MSW, Office on Child Abuse and Neglect, JA Gaudiosi, DBA, Data Team, Admin for Children and Families. YYYuan, PhD, L Zikratova, MS, MJ Oritz, MA, MM Aveni, Walter R. McDonald and Associates, Inc., Sacramento, California. RT Leeb, PhD, TR Simon, PhD, WR Hammond, PhD, Div of Violence Prevention, National Center for Injury Prevention and Control, CDC.

Editorial Note: The findings in this report indicate that, in fiscal year 2006, 23.2 children per 1,000 population aged <1 year experienced substantiated nonfatal maltreatment in the United States. Among these infants, neglect was the maltreatment category most commonly cited, experienced by 68.5% of victims. Among infant victims aged <1 year who experienced substantiated maltreatment, 32.7% were aged ≤1 week, and 30.6% were aged <4 days. Neglect also was the maltreatment category most often cited among children aged <1 week.

This report is the first published national analysis of substantiated nonfatal maltreatment of infants, using NCANDS data. Although the results demonstrate a concentration of maltreatment and neglect at age ≤1 week, NCANDS data cannot be used to determine the etiology of the infant maltreatment and neglect because NCANDS

reports are limited to broad categories and do not provide specific information about diagnoses or the circumstances o the maltreatment. The concentration of reports of neglect in the first few days of life and the preponderance of reports from medical professionals during the same period suggest that neglect often was identified at birth. One hypothesis for the concentration of maltreatment and neglect reports in the first few days of life is that the majority of reports resulted from maternal or newborn drug tests. Although tracking of prenatal substance exposure and hospital postnata toxicology-screening practices vary among states and within states, positive maternal or neonatal drug test results routinely are reported to CPS agencies as child neglect (3). Additional research is needed to clearly define the causes of substantiated neglect and maltreatment among newborns and to determine the best strategies for intervention.

The percentage of substantiated reports categorized as physical abuse among infants aged ≤1 week (13.2%) is similar to the percentage among maltreated children of all ages (16%) (1). Physical abuse is defined by CDC and NCANDS as the intentional use of physical force by a parent or caregiver against a child that results in, or has the potential to result in, physical injury. Physical abuse includes beating, kicking, biting, burning, shaking, or otherwise harming a child. Although the act is intentional.

the consequence might be intentional or unintentional -(i.e., resulting from overdiscipline or physical punishment) (1,4). One type of physical abuse, shaken baby syndrome/ abusive head trauma (SBS/AHT) (5), is a cause of severe physical injury and death in infants, occurring in 21.0– 32.2 infants aged <1 year per 100,000 population. More detailed study of contextual information is needed to determine the causes of physical abuse in infants reported to NCANDS and to develop additional prevention strategies. Few studies have examined rates and risk factors for maltreatment in infants aged <1 year, and risk for nonfatal maltreatment among infants has not been examined previously at the national level in the United States. A study by the Public Health Agency of Canada provided nationallevel data for that country (excluding the province of Quebec) and reported incidence in 2003 of substantiated nonfatal maltreatment among infants aged <1 year of 27.3 per 1,000 population for females and 29.1 for males,** similar to the rates described in this report. Also similar to this study, the Canadian study found that neglect was the most common form of substantiated maltreatment for children aged <3 years; the Canadian study did not determine the most common form of maltreatment among infants aged <1 year.

The findings in this report are subject to at least two other limitations, in addition to the lack of specific information about maltreatment circumstances. First, underreporting or delayed reporting might influence the findings. Both mandated reporters and the public might lack sufficient knowledge or training that supports reporting possible child maltreatment (6,7). To assist health-care professionals in better reporting child maltreatment, CDC developed uniform definitions and recommended data elements to promote and improve consistency of child maltreatment reporting and serve as a technical reference for the collection of data (4). Second, data collection and reporting practices vary among states, and data from certain states were not available for analysis.

CDC supports a range of research, early intervention, and prevention programs at the national, state, and local levels. These efforts include a focus on developing child-maltreatment tracking programs in state health departments and promotion of positive parenting and prevention of child maltreatment through a framework of safe, stable, and nurturing relationships between children and caregivers. Similarly, ACF supports a range of prevention and intervention programs, including programs to identify and serve

[blocks in formation]

substance-exposed newborns and reduce variation in the policies and procedures related to prenatal substance exposure. Reframing neglect as a series of missed opportunities for prevention and emphasizing safe, stable, and nurturing relationships can highlight opportunities for prevention that might otherwise be missed. For example, approximately 84% of pregnant women in the United States receive some prenatal care, and approximately 99% of infants are born in medical settings (8), these setting provide an opportunity for medical professionals to detect and manage early risk for maltreatment (e.g., maternal substance abuse) that can impair or interfere with child-caregiver relationships.

Serious injury resulting from physical abuse of infants can be decreased by efforts focusing on reduction of SBS/ AHT through in-hospital programs aimed at parents of newborns. These programs have produced a substantial reduction in reported SBS/AHT in localized areas (9), and CDC is supporting research to evaluate the replicability of these results in diverse settings. In addition, home-visitation and parent-training programs (10), particularly those that 1) begin during pregnancy, 2) provide social support to parents, and 3) teach parents about developmentally appropriate infant behavior and age-appropriate disciplinary communication skills, have been determined to reduce risk for child maltreatment.


1. US Department of Health and Human Services, Administration on Children, Youth and Families. Child maltreatment 2006. Washington, DC: US Department of Health and Human Services; 2008. In press. 2. CDC. Variation in homicide during infancy-United States, 1989– 1998. MMWR 2002;51:187-9.

3. Birchfield M, Scully J, Handler A. Perinatal screening for illicit drugs: policies in hospitals in a large metropolitan area. J Perinatol 1995;15: 208-14.

4. Leeb RT, Paulozzi L, Melanson C, Simon T, Arias I. Child maltreatment surveillance: uniform definitions for public health and recommended data elements, version 1.0. Atlanta, GA: US Department of Health and Human Services, CDC; 2008.

5. American Academy of Pediatrics, Committee on Child Abuse and Neglect. Shaken baby syndrome: rotational cranial injuries—technical report. Pediatrics 2001;108:206–10.

6. Levi BH, Brown G. Reasonable suspicion: a study of Pennsylvania pediatricians regarding child abuse. Pediatrics 2005;116:e5-12. 7. Markenson D, Tunik M, Cooper A, et al. A national assessment of knowledge, attitudes, and confidence of prehospital providers in the assessment and management of child maltreatment. Pediatrics 2007;119:e103-8.

8. Martin JA, Hamilton BE, Sutton PD, et al. Births: final data for 2005. Natl Vital Stat Rep 2007;56(6).

9. Dias MS, Smith K, DeGuehery K, Mazur P, Li V, Shaffer ML. Preventing abusive head trauma among infants and young children: a hospital-based, parent education program. Pediatrics 2005;115: e470-7.

10. Bilukha O, Hahn RA, Crosby A, Task Force on Community Preventive Services. The effectiveness of early childhood home visitation in preventing violence: a systematic review. Am J Prev Med 2005;28(2 suppl 1):11–39.

[merged small][ocr errors]

Clostridium difficile is a well-known cause of hospitalacquired infectious diarrhea and is associated with increased health-care costs, prolonged hospitalizations, and increased patient morbidity. Previous antimicrobial use, especially use of clindamycin or ciprofloxacin, is the primary risk factor for development of C. difficile-associated diarrhea (CDAD) because it disrupts normal bowel flora and promotes C. difficile overgrowth (1). Historically, CDAD has been associated with elderly hospital in-patients or longterm-care facility (LTCF) residents. Since 2000, a strain of C. difficile that has been identified as North American pulsed-field type 1 (NAP1) and produces an extra toxin (binary toxin) and increased amounts of toxins A and B has caused increased morbidity and mortality among hospitalized patients (2,3). During 2005, related strains caused severe disease in generally healthy persons in the community at a rate of 7.6 cases per 100,000 population, suggesting that traditional risk factors for C. difficile might not always be factors in development of community-associated CDAD (CA-CDAD) (4). Cases of CA-CDAD are not nationally reportable, and population-based data at a statewide level have not been reported previously. In 2006, the Connecticut Department of Public Health (DPH) implemented a statewide surveillance system to assess the burden of CA-CDAD and to determine the descriptive epidemiology, trends, and risk factors for this disease. This report describes that surveillance system and summarizes results from the first year of surveillance. The findings indicated the presence of occasionally severe CDAD among healthy persons living in the community, including persons with no established risk factors for infection. Clinicians should consider a diagnosis of CA-CDAD in outpatients with severe diarrhea, even in the absence of established risk factors. In addition, continued surveillance is needed to determine trends in occurrence and whether more toxigenic strains are having an increasing impact in the community and in the hospital setting.

On January 1, 2006, CA-CDAD was added to the list of conditions reportable by Connecticut health-care providers. A case of CA-CDAD was defined as a positive C. difficile toxin assay for a person with gastrointestinal symptoms and no known previous overnight hospitalizations or LTCF stays during the 3 months preceding specimen collection, collected from an outpatient or within 48 hours of hospital admission (5). DPH staff members contacted hospital infection-control practitioners at Connecticut's 32 acute

care hospitals by telephone, informed them about the new reporting requirements, and asked them to review positive laboratory results to identify cases. Laboratories were no required to report to DPH. Physicians were informed by special mailing. In May 2006, all hospitals were sent a let ter summarizing initial findings and reminding physician and infection-control practitioners about the reporting requirements. In addition, hospitals that did not initiall report cases were recontacted by telephone and reminded of the reporting requirements. DPH staff members contacted treating physicians to confirm case status and collect patient information, including demographics. symptoms, select medical history, and possible risk factors. When necessary, DPH staff members reviewed medical records or conducted patient interviews. However, systematic patient interviews to verify absence of a recent stay in a health-care setting were not conducted.

Incidence rates were calculated using the number of confirmed cases reported among Connecticut residents and 2005 U.S. Census state population estimates. Differences in proportions and tests for trend by age group were evaluated using the chi-square test and chi-square test for trend: multivariate logistic regression analysis was conducted. A separate 3-month pilot study was conducted during 2006 by FoodNet,* Emerging Infections Program sites,† and CDC to collect specimens from patients with CA-CDAD for culture for C. difficile and to characterize the isolates by toxinotyping and detection of binary toxin and deletions in the tcdC gene (6). As part of this study, in Connecticut. all toxin-positive stool specimens from confirmed CACDAD patients at three hospital laboratories were collected and cultured.

A total of 456 possible cases, determined on the basis of tests conducted on outpatients or within 2 days of hospitalization, were reported during 2006; 241 (53%) were subsequently confirmed as meeting the case definition. Of the 215 cases that were not confirmed, 159 (74%) occurred in persons who had an LTCF stay or hospitalization during the preceding 3 months, 50 (23%) occurred in per

*The Foodborne Diseases Active Surveillance Network (FoodNet) is the principal foodborne and enteric disease surveillance component of CDC's Emerging Infections Program (EIP). FoodNet is a set of population-based surveillance projects for laboratory-confirmed disease collaboratively undertaken by CDC. 10 EIP sites (Connecticut, Georgia, Maryland, Minnesota, New Mexico, Oregon. Tennessee, and selected counties in California, Colorado, and New York), the U.S. Department of Agriculture, and the Food and Drug Administration. * The CDC Emerging Infections Program supports population-based surveillance in 10 sites in the United States. Each site is based in a state health department, often with a local academic center partner, working in collaboration with local health departments, public health laboratories, clinical laboratories, infection-contro practitioners, health-care providers, and hospitals to assess the public health impact of emerging infections and evaluate methods for their prevention and control.

sons for whom insufficient medical information was avails able to enable confirmation; and six (<1%) were in persons who were asymptomatic The overall annual 2006 incidence of CA-CDAD was 6.9 cases per 100,000 population, with similar rates found in most counties. Incidence among those aged ≥5 years increased with age; females had nearly twice the incidence of males. Rates were higher during the spring and summer months than during the fall and winter months (Table 1).

A total of 28 (88%) of 32 acute-care hospitals reported at least one case of CA-CDAD (range: 1-26 cases). Among the 241 cases, 110 (46%) were in patients who required hospitalization for CA-CDAD, mainly for diagnosis and treatment of dehydration or colitis; 13 (12%) were in patients who required an intensive-care unit stay, two (2%) were in patients who had both toxic megacolon and a colectomy, and two (2%) were in patients who died of complications related to C. difficile infection. The median length of stay among hospitalized patients was 4 days (range: 1– 39 days).

Among all patients for whom follow-up information was available, 29% had an inpatient health-care exposure (defined as overnight hospitalization or LTCF stay during the >3 to 12 months preceding illness or day surgery during the 12 months preceding illness), 67% had an underlying medical condition, and 68% had taken an antimicrobial during the 3 months preceding symptom onset (Table 2). When CA-CDAD patients requiring hospitalization were compared with those managed as outpatients, independent predictors of hospitalization by multivariate analysis included age of ≥65 years (p = 0.001), fever (p = 0.001), and inpatient health-care exposure during the >3 to 12 months preceding illness (p = 0.04). A total of 59 (25%) patients had no underlying conditions and no inpatient health-care exposures during the 12 months preceding illness. Compared with all other patients, this group was younger (63% versus 23% were aged <45 years [p<0.0001]), less likely to be hospitalized for their CA-CDAD illness (36% versus 52% [p<0.04]), and more likely to report bloody diarrhea (37% versus 19%

TABLE 1. Number, percentage, and rate* of community-associated Clostridium difficile-associated disease cases,† by selected characteristics

[ocr errors]

Connecticut, 2006

[blocks in formation]

† A case was defined as a positive C. difficile toxin assay for a person with gastrointestinal symptoms and no known previous overnight hospitalizations or long-term care facility stays during the 3 months preceding specimen collection, collected from an outpatient or within 48 hours of hospital admission. § Relative risk.

¶ Confidence interval.

Chi-square test for linear trend for 5-14 years age group and older.

[blocks in formation]

* A case was defined as a positive C. difficile toxin assay for a person with gastrointestinal symptoms and no known previous overnight hospitalizations or long-term-care facility stays during the 3 months preceding specimen collection, collected from an outpatient or within 48 hours of hospital admission.

† Median number of stools per day was six (range: 1-30). § Documented as a temperature >100.4°F (>38.0°C).

¶ Among those with reported antibiotic use; 33 patients used antibiotics but were unable to specify type.

** Includes cephalosporins, tetracycline, macrolides, and metronidazole.

[p=0.01]). In addition, 35 (59%) patients received an antimicrobial during the 3 months preceding symptom onset, 21 (36%) took no antimicrobial, and three (5%) patients had no information on antimicrobial use available.

Twelve C. difficile isolates were recovered from toxinpositive stool specimens and were characterized at CDC. Eight (67%) had binary toxin genes similar to the epidemic NAP1 strain, and three (25%) were identified as NAP1.

Coinfection with a second pathogen appeared to be rare. A review of the FoodNet enteric pathogen surveillance database in Connecticut indicated that five (2%) of the 241 patients with CA-CDAD also had a stool-culture positive result for another reportable enteric pathogen from a specimen collected on the same day or within 1 day of the toxin-positive C. difficile sample: Salmonella (one patient), Campylobacter (three), and Escherichia coli O157:H7 (one). Reported by: T Rabatsky-Ehr, MPH, K Purviance, MPH, D Mlynarski, MPH, P Mshar, MPH, J Hadler, MD, Epidemiology and Emerging Infections Program, Connecticut Dept of Public Health. L Sosa, MD, EIS Officer, CDC.

Editorial Note: The findings in this report demonstrate that CA-CDAD is an important and geographically widespread health problem among Connecticut outpatients, a

population previously thought to be at low risk for thi disease. Although interest in CA-CDAD has grown in recent years, this report describes the first attempt: define population-based incidence of this disease at the stat level. The CA-CDAD incidence in Connecticut in 200 (6.9 per 100,000 population) was similar to that found Philadelphia in 2005 (7.6 per 100,000 population) using a similar case definition. Both of these rates were conside ably lower than that found in the United Kingdom (UK in 2004 (22.0 per 100,000 population), despite the fact the UK study used a more restrictive case definition in which persons with hospitalization during the 12 month preceding illness onset were excluded (4,7).

The findings in this report highlight the importance c increasing age (with the attendant underlying health problems and increased use of the health-care system) and antibiotic exposure in the development of CDAD. However one fourth of all CA-CDAD cases were in persons whe lacked established predisposing risk factors for CDAD including advanced age, an underlying health condition. and a health-care exposure during the 12 months preceding illness. Moreover, similar to what was observed in th community studies conducted in Philadelphia and the UK 32% of patients had no recent exposure to antimicrobia's Approximately 9% of all cases were in patients who had none of these factors. These findings emphasize the need for continued study of this disease to identify additional risk factors for exposure to C. difficile and for developmer: of disease.

The ability of C. difficile to form spores is thought to be a key feature in enabling the bacteria to persist in patients and the physical environment for long periods, thereby facilitating its transmission. C. difficile is transmitted through the fecal-oral route. Postulated risk factors for acquiring C. difficile in the community include contact with a contaminated health-care environment, contact with persons who are infected with and shedding C. difficile (person-to-person transmission), and ingestion of contaminated food.

Studies have shown C. difficile to be a pathogen or colenizer of calves, pigs, and humans (8,9). The recent detec tion of the NAP1 strain of C. difficile in retail ground bee is cause for concern (9). This hyper-toxin-producing strain has been reported as a cause of serious outbreaks of healthcare-associated disease in humans in North America and Europe (10) and was found among a small subset of specimens from CA-CDAD cases in Connecticut. Further studies are needed to determine whether C. difficile is transmitted via the food chain and the relative importance of such transmission in human CDAD.

« PreviousContinue »