Objective: To investigate the characteristics and trends of injury among hospitalized child abuse children younger than 18 years between 1997 and 2009. Methods: We selected hospitalized child abuse cases from the Taiwan's National Health Insurance (NHI) Database in accordance with ICD-9-CM 995.5x or E967.x. Results: There were 1,212 victims of hospitalized child abuse during the research period, including 735 boys and 477 girls. The victims were most frequently inflicted by the father or stepfather, and the most common injury was intracranial damage; besides, the victims most commonly sought treatment in medical centers. Infants younger than 1 year accounted for the highest percentage of hospitalized victims. Although there were no apparent trends in the overall hospitalization rate of the child abuse victims over the 13-year period, the rate of abuse among girls increased gradually over the years. Conclusions: The medical staffin hospitals should provide appropriate treatment to the victims and contact relevant organizations to intervene; moreover, government agencies should formulate effective control measures to develop a safe growth environment for children.
Keywords: child abuse; injury; National Health Insurance Research Database; Taiwan, hospitalization rate, medical use
In recent years, child abuse had grown to become a severe social problem globally (World Health Organization, 2006). A United Nations Children's Fund (UNICEF) report revealed that in industrialized countries, 3,500 children abuse deaths less than 15 years every year (UNICEF, 2003). In Taiwan, according to the domestic violence statistics reported by the Ministry of the Interior, there were 16,989 child and juvenile protection cases in 2008 (Domestic Violence and Sexual Assault Prevention Committee, Ministry of Interior, ROC, 2010). In addition, previous studies pointed out that child younger than 4 years were the most vulnerable to death resulting from abuse (Chang et al., 2004; Forjuoh, 2000; World Health Organization, 2002).
In Taiwan, few studies focused on child maltreatment by using a nationwide population-based database (Chiang, Huang, Feng, & Lu, 2012; Chun, Chiu, Pai, & Chien, 2010). Chiang's research showed that the incidence of hospitalization coded as "because of child maltreatment" differed by age groups, and infants had the highest and the largest increase in incidence during 1996-2007 (Chiang et al., 2012); however, it had no attention paid to the medical use of hospitalized child abuse. Thus, this study examined data from the National Health Insurance (NHI) database to investigate the characteristics (including demographic, injury type, and medical use) and trends of injury among hospitalized child abuse children younger than 18 years between 1997 and 2009.
This study used data from Taiwan's NHI database collected from 1997 to 2009. The use of the data was approved by the Professional Peer Reviewer Committee. Because the patients remained anonymous, this study did not infringe on patient privacy and was therefore in compliance with the Declaration of Helsinki. Taiwan had an NHI acceptance rate of more than 99%. The NHI database was the empirical information representative of Taiwan (National Health Research Institutes, 2011); as such, it contained information about hospitalized patients and medical professionals in various medical institutions.
The ICD-9-CM (International Classification of Diseases, 9th revision, Clinical Modification) N-Code (nature of injury code) is a classification system for the diagnosis of diseases and injuries, and E-Code (external causes of injury code) is a classification system for the causes of injuries. The NHI database in Taiwan employed an ICD-9-CM coding system, which offered one N-Code for primary diagnosis and four N-Codes for secondary diagnosis. To code injury causes, the database provided an E-Code for primary diagnosis and an E-Code for secondary diagnosis.
Definition of the Study Subjects
This study defined abused children as those who were younger than 18 years of age with an ICD-9-CM N-Code of 995.50 (child maltreatment syndrome-child abuse), 995.51 (child emotional/psychological abuse), 995.52 (child neglect), 995.53 (child sexual abuse), 995.54 (child physical abuse), 995.55 (shaken infant syndrome), 995.59 (other child abuse and neglect) or an E-Code of E967.0 (child abuse perpetrated by father, stepfather, or boyfriend), E967.1 (child abuse perpetrated by other specified person), E967.2 (child abuse perpetrated by mother, stepmother or girlfriend), E967.3 (child abuse perpetrated by spouse or partner), E967.4 (child abuse perpetrated by a child), E967.5 (child abuse perpetrated by a sibling), E967.6 (child abuse perpetrated by a grandparent), E967.7 (child abuse perpetrated by other relative), E967.8 (child abuse perpetrated by nonrelated caregiver), E967.9 (child abuse perpetrated by unspecified person).
Regarding to NHI hospitalization data, because the NHI requires medical facilities to file claims on a monthly basis, the same case may be separated into several data entries in between 2 months and across the year. Likewise, a single patient may appear as several entries because of hospital transfers. Therefore, all entries with the same scrambled ID, N-Code and E-Code was regarded as the same episodes.
For the statistical analysis, SPSS 18.0 was used to apply the chi-square test, Fisher's exact test and an independent samples t test to elucidate gender differences in demography, injury type, and medical use. This study used the Joinpoint Regression Program 3.5 (National Cancer Institute, 2011) to examine changing trends in the hospitalization rates for child abuse from 1997 to 2009, for which p , .05 was set as the significance threshold.
Demography and Medical Use of Victims of Hospitalized Child Abuse
From 1997 to 2009, 1,212 children were hospitalized because of abuse of these children, 75 died. The ratio of hospitalized boys to girls was 1.54:1.00; however, the hospitalized girls were 1.14 times more likely than the hospitalized boys to die. The average age of the hospitalized victims in boys was higher than that in girls. In addition, age-based grouping analysis showed that infants younger than 1 year accounted for the highest percentage of hospitalized children, followed by children/juveniles aged 12-18 years (Table 1).
The 1,212 children hospitalized because of abuse from 1997 to 2009 had an average length of hospital stay of 11.63 days, and each patient had an average cost of NT$99,851.74. In terms of gender, girls experienced longer stays and higher costs than boys. Most victims of hospitalized child abuse went to medical centers, next to regional hospitals. In addition, more than half of the abused children received surgery or medical treatment (661 cases, 54.54%; Table 1).
Injury Types and Causes of Victims of Hospitalized Child Abuse
Among the 1,212 hospitalized children who were abuse victims during research period, 107 patients (8.83%) suffered two or more abuse events. On average, each hospitalized abused child had 2.12 injury sites. The most common trauma was intracranial injury, followed by child maltreatment syndrome (Table 2). Among the 383 patients primary diagnosed with child maltreatment syndrome, 70 (18.28%) also had the secondary diagnosis of shaken infant syndrome.
An analysis of gender revealed differences in the injury types sustained by boys and girls. Aside from burns and poisoning, which were of the same proportions for boys and girls, all the remaining injury types showed higher proportions in boys than in girls (Table 2). A review of age showed the following differences in injury types (primary diagnosis) among different age groups: infants younger than 1 year had much higher proportions of child maltreatment syndrome and skull fracture than children of other age groups; lower limb fracture was mostly seen among children aged 1-2 years; internal injury of chest, abdomen, and pelvis accounted for the highest among children aged 3-5 years; the highest proportions of injury among children aged 6-11 years was of burn; and children/juveniles aged 12-18 years suffered much higher likelihoods of upper limb fracture, intracranial injury, superficial injury, and open wounds of the head, neck, trunk, and limbs (Table 2).
We investigated the injury types sustained by the 75 dead victims of hospitalized child abuse and concluded that child maltreatment syndrome, intracranial injury, and skull fracture were the top three injury types leading to death (Table 2).
An analysis of injury causes (E-Code) showed that of the 928 hospitalized child abuse for whom causes were provided, child abuse perpetrated by unspecified person was the most common (292 cases), next is child abuse perpetrated by others (213 cases), and then child abuse perpetrated by father, stepfather, or boyfriend (206 cases). However, child abuse perpetrated by mother, stepmother, or girlfriend was only 28 cases (Table 3).
Trends in the Hospitalization Rates of Victims of Hospitalized Child Abuse
Overall, there was no significant change during the 13-year period (1.5097 per 100,000 population in 1997 and 1.8724 in 2009; annual percent change [APC] 5 2.08; p 5 .085). However, once gender was taken into account, it was revealed that the hospitalization rate of girls increased over the years (APC 5 5.17; p 5 .008), whereas the hospitalization rate of boys exhibited little change (APC 5 20.22; p 5 .882; Figure 1).
Demography and Medical Use
Our analysis of gender showed that most victims of hospitalized child abuse were boys. This result was in agreement with other studies (Bullock, Koval, Moen, Carney, & Spratt, 2009; Chang et al., 2004; DiScala, Sege, Li, & Reece, 2000); moreover, a Canadian report on the National Statistics of Child Abuse indicated that boys were more likely to suffer physical maltreatment among 8-11 years, but girls were more likely to be sexually abused than boys regardless of age (Jack, Munn, Cheng, & MacMillan, 2006). Similar patterns were apparent in this study: boys aged 8-11 suffered physical abuse than girls at these ages (62.3% vs. 37.7%), and there were more girls than boys among the sexually abused children (66.6% vs. 33.3%).
Many domestic and international studies had shown that infants younger than 1 year accounted for a significant or very high percentage of abused children (29%-62%; Bullock et al., 2009; Chang et al., 2004; DiScala et al., 2000; Forjuoh, 2000), which was also found in our study (33.09% for infants younger than 1 year of age). These were mainly because infants had no capacity for action (e.g., they are completely dependent on the caregivers) and were powerless to resist or escape from maltreatment. Furthermore, their fragile bodies made them more vulnerable to severe and potentially life-threatening trauma (Chang et al., 2004; Forjuoh, 2000).
The review of hospitalization length in this study revealed that abused children had an average hospital stay of 11.63 days, slightly longer than the corresponding numbers in other countries (i.e., 9.30 days [DiScala et al., 2000] and 8.20 days [Rovi, Chen, & Johnson, 2004]). Although most abused children in the United States are covered by Medicaid (Bullock et al., 2009; Rovi et al., 2004), Taiwan has an NHI system. Without the pressure of high medical costs, abuse victims in Taiwan could stay in hospitals slightly longer than those in countries with other health care payment systems. In addition, the severity of the abuse or the delay in seeking treatment might also influence factors of length of stays, and they should be discussed in further studies.
The analysis of hospitalization costs in this study revealed that in Taiwan, each abused child had an average cost of NT$99,851.74 or US$3,120 (based on an exchange rate of US$1 to NT$32). In contrast, each abused child in the United States had a hospitalization cost of US$15,000 to US$25,000 (Bullock et al., 2009; Libby, Sills, Thurston, & Orton, 2003; Rovi et al., 2004), 4.81-8.01 times that of the cost in Taiwan. This divergence may reflect the differences in the two countries' medical insurance and payment systems. Furthermore, many foreign studies have shown that abused children are likely to accrue both longer hospital stays and higher medical costs than their nonabused peers (Bullock et al., 2009; DiScala et al., 2000; Libby et al., 2003; Rovi et al., 2004). The possible cause of this divergence may be the injury severity and medical history (such as congenital diseases) of the abused children as well as the possibility that abused children might require further observation and assessment, thus extending their hospital stays (Rovi et al., 2004).
Injury Types and Cause
Our examination of child abuse-related injury types showed that abused children were most likely to suffer intracranial injury (39.19%). This finding was in agreement with many studies conducted in the United States that have found intracranial injury (Bullock et al., 2009; Chang et al., 2004; DiScala et al., 2000; Libby et al., 2003); fractures of the skull, neck, and trunk (Bullock et al., 2009; Chang et al., 2004); shaken infant syndrome (Rovi et al., 2004); and thoracic and abdominal trauma (DiScala et al., 2000) to be common injuries among child abuse victims.
Examining the injury types that lead to abusive death, this study follows many previous reports in showing that severe head trauma was the leading cause of abuse-related death in children (Berkowitz, 1995; Chang et al., 2004; Lee, 2003; Rubin, Christian, Bilaniuk, Zazyczny, & Durbin, 2003; World Health Organization, 2006). This result might be related to the immature brain tissues of infants and toddlers, which were prone to severe and irreversible damage from violent shaking or impact (Chang et al., 2004; Li, 2006).
This study and many previous reports agreed that the perpetrators of abuse were mostly the children's caretakers. The abuser was most likely to be the father (stepfather), male guardian, or the mother's male cohabitant (Bullock et al., 2009; Chang et al., 2004; Lee, 2003), although the abuser might also be the mother (stepmother) or the father's female cohabitant (Chang et al., 2004; Lee, 2003). Multiple factors contribute to long-term neglect, abuse, and even abandonment of children by parents. These factors include the child crying frequently or being rebellious, a traditional concept of "spare the rod, spoil the child" or the misconception held by certain families that demanding children ruin the parents (Lee, 2003; Li, 2006).
Trends in the Hospitalization Rates
The statistics from the Child and Juvenile Protection Bulletin in Taiwan revealed an apparent upward trend in the number of cases in which children and juveniles were abused or neglected (from 2,238 to 17,476 cases during the period 2003-2009; Domestic Violence and Sexual Assault Prevention Committee, Ministry of Interior, ROC, 2010). Hence, government departments should formulate measures for preventing child abuse to increase the general public's awareness about the issues of child protection. Importantly, birth statistics showed that the sex ratio of boys to girls increased from 1.087 to 1.095 during the period from 1997 to 2009 in Taiwan (Bureau of Health Promotion, Department of Health, Executive Yuan, ROC, 2010), illustrating the ongoing increase in the number of male infants compared to female infants. Nevertheless, the data used in this study showed that the hospitalization rate for girls increased significantly over the study period, whereas the hospitalization rate for boys exhibited little change. It needs further research to explore the reason for the increasing hospitalization rate for girls.
This study examined NHI data, which excluded some types of information that might had been informative (i.e., the victims' family structure and caretaker characteristics) and therefore limited the extent of our investigation. In addition, the research data only provided hospitalization profiles for abused children who had medical records from 1997 to 2009, thus excluding those who were not hospitalized. Finally, the definition of "total hospital cost" did not include charges for other related services (e.g., psychological counseling), potentially underestimating case numbers and/or medical costs.
Infants younger than 1 year of age are a group at high risk for child abuse (33.09%). The injury types of the abused children were predominantly intracranial trauma (39.19%) and child maltreatment syndrome (31.60%). The most common perpetrator of abuse was the father or stepfather (48.70%). Abused patients were most often admitted to medical centers (41.83%). The average hospitalization length was 11.63 days, and the average hospitalization cost was NT$99,851.74. According to statistics reported by domestic violence notification units of the Ministry of the Interior in Taiwan, only 37.08% of the cases were disclosed by medical institutions (i.e., health centers, clinics, or hospitals); these cases might represent only the tip of the iceberg. Medical professionals have the first line of contact with abused children and therefore occupy an ideal position to notify authorities of maltreatment. If medical workers can remain vigilant, develop enhanced sensitivity to suspected cases of child abuse and appropriately handle abuse cases, they can provide a safe environment for abused children.
Berkowitz, C. D. (1995). Pediatric abuse. New patterns of injury. Emergency Medicine Clinics of North America, 13, 321-341.
Bullock, D. P., Koval, K. J., Moen, K. Y., Carney, B. T., & Spratt, K. F. (2009). Hospitalized cases of child abuse in America: Who, what, when, and where. Journal of Pediatric Orthopedics, 29, 231-237.
Bureau of Health Promotion, Department of Health, Executive Yuan, ROC (Taiwan). (2010). The imbalance of sex ratio. Retrieved from http://www.health99.doh.gov.tw/Hot_News/h_ NewsDetailN.aspx?TopIcNo=5679
Chang, D. C., Knight, V., Ziegfeld, S., Haider, A., Warfield, D., & Paidas, C. (2004). The tip of the iceberg for child abuse: The critical roles of the pediatric trauma service and its registry. The Journal of Trauma, 57, 1189-1198.
Chiang, W. L., Huang, Y. T., Feng, J. Y., & Lu, T. H. (2012). Incidence of hospitalization due to child maltreatment in Taiwan, 1996-2007: A nationwide population-based study. Child Abuse & Neglect, 36, 135-141.
Chun, C. H., Chiu, Y. L., Pai, L., & Chien, W. C. (2010). Epidemiology of inpatients due to domestic violence in Taiwan, 2006-2007. Asian Journal of Domestic Violence and Sex Offense, 6, 65-85.
DiScala, C., Sege, R., Li, G., & Reece, R. M. (2000). Child abuse and unintentional injuries: A 10-year retrospective. Archives of Pediatrics & Adolescent Medicine, 154, 16-22.
Domestic Violence and Sexual Assault Prevention Committee, Ministry of Interior, ROC (Taiwan). (2010). Statistical report of domestic violence type. Retrieved from http://dspc.moi.gov.tw/ct .asp?xItem=2315&ctNode=776&mp=1
Forjuoh, S. N. (2000). Child maltreatment related injuries: Incidence, hospital charges, and correlates of hospitalization. Child Abuse & Neglect, 24, 1019-1025.
Jack, S., Munn, C., Cheng, C., & MacMillan, H. L. (2006). Child maltreatment in Canada: Overview paper. Ottawa, Canada: Public Health Agency of Canada.
Lee, A. (2003). Management of child abuse in Hong Kong: results of a territory-wide inter-hospital prospective surveillance study. Hong Kong Medical Journal, 9, 6-9.
Li, C. C. (2006). Identify child physical abuse. Medical Journal Today, 33, 610-619.
Libby, A. M., Sills, M. R., Thurston, N. K, & Orton, H. D. (2003). Costs of childhood physical abuse: Comparing inflicted and unintentional traumatic brain injuries. Pediatrics, 112, 58-65.
National Cancer Institute. (2011). Surveillance Research-Joinpoint Regression Program (Latest Release: Version 3.5). Retrieved from http://surveillance.cancer.gov/joinpoint/
National Health Research Institutes, ROC (Taiwan). (2011). National Health Insurance Research Database-Introduction of service. Retrieved from http://w3.nhri.org.tw/nhird//brief_01.htm
Rovi, S., Chen, P. H., Johnson, M. S. (2004). The economic burden of hospitalizations associated with child abuse and neglect. American Journal of Public Health, 94, 586-590.
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.
The United Nations Children's Fund. (2003). A league table of child maltreatment deaths in rich nations. Innocenti Report Card, No. 5. Florence, Italy: UNICEF Innocenti Research Center.
World Health Organization. (2002). World report on violence and health: Summary. Retrieved from http://www.who.int/violence_injury_prevention/violence/world_report/en/summary_en.pdf
World Health Organization. (2006). Preventing child maltreatment: A guide to taking action and generating evidence. Retrieved from http://www.phac-aspc.gc.ca/ncfv-cnivf/pdfs/nfnts-prev -maltr_e.pdf
Acknowledgments. This study was based in part on data from the NHI Research Database provided by the Bureau of NHI, Department of Health and managed by the National Health Research Institutes. The interpretation and conclusions contained herein did not represent those of the Bureau of NHI, Department of Health or National Health Research Institutes.
Wu-Chien Chien, PhD
Chi-Hsiang Chung, PhD
National Defense Medical Center, Taiwan
Lu Pai, PhD
Taiwan Injury Prevention & Safety Promotion Association
Senyeong Kao, PhD
Yu-Luan Chiu, MS
Chien-Hua Cheng, MS
National Defense Medical Center, Taiwan
Correspondence regarding this article should be directed to Wu-Chien Chien, PhD, School of Public Health 4210R, No. 161, Section 6, Min-Chuan East Road, Neihu District, Taipei City, 11490, Taiwan, Republic of China. E-mail: email@example.com…