Factors Affecting Pediatricians' Reporting of Suspected Child Maltreatment

Factors Affecting Pediatricians’ Reporting of Suspected Child Maltreatment Veronica L. Gunn, MD, MPH; Gerald B. Hickson, MD; William O. Cooper, MD, MPH Objective.—To identify factors associated with pediatricians’ decision not to report suspected child maltreatment. Design.—A survey was distributed to a random sample of pediatricians in a single state. Participants were asked if they had ever suspected child abuse or neglect but did not report. In addition, all were asked to list all the considerations that pediatricians incorporate into their decisions not to report. Results.—One hundred ninety-five pediatricians completed the survey (56% of those eligible). Twenty-eight percent of respondents stated that they had considered reporting an incident of suspected child maltreatment but had chosen not to. Providers who had chosen not to report were more likely to be men (P 5 .006), to have been in practice longer (P 5 .001), to have reported more cases (P 5 .001), to have been deposed (P 5 .001) or to have testified (P 5 .01) in child maltreatment cases, and to have been threatened with lawsuit (P 5 .02) than were pediatricians who had never declined to report. Multivariate logistic regression demonstrated that male gender (odds ratio [OR] 2.18; 95% confidence interval [CI] 1.05–4.49), years in practice (OR 1.23; 95% CI 1.05–1.44), and experience reporting (OR 1.28; 95% CI 1.02–1.60) were all independently associated with decisions not to report. Respondents who had declined to report were more likely to cite lack of knowledge about reporting laws and process (P 5 .05) and poor experiences with child service agencies (P 5 .03) as reasons for not reporting than were their counterparts who had never declined to report suspected maltreatment. Conclusions.—Many barriers exist to reporting suspected maltreatment. Specific educational interventions may be helpful in eliminating barriers to reporting. KEY WORDS: abuse; child maltreatment; neglect; reporting Ambulatory Pediatrics 2005;5:96 101 I than 10% of total maltreatment reports to child protection agencies.1–3 The personal costs to children who are abused and neglected are substantial, and the financial costs of child maltreatment are estimated at $94 billion per year.16 Thus, addressing potential deterrents to physician reporting is important to protect children and to reduce a significant health care burden. Although pediatricians in the United States have a prominent role in the care of children, their reporting practices are still not fully understood.11,17 This study was designed to identify why pediatricians might suspect child abuse or neglect but choose not to report. On the basis of literature of the reporting practices of other medical and nonmedical professionals,4,11–15,17–29 we hypothesized that pediatricians’ reporting practices are influenced by physician characteristics and perceived barriers in the mechanism for reporting suspected child maltreatment. n 2001, more than 900 000 children were victims of child maltreatment; an estimated 1300 children died secondary to abuse or neglect.1 Children rarely report abuse themselves; child victims are the source of fewer than 1% of reports to child protection agencies regarding child sexual, physical, and emotional abuse and neglect.1–4 As a result, abused and neglected children are usually identified by police, health care providers, teachers, and other adults who report suspected maltreatment to child protection authorities. In every state, certain individuals with regular access to children, including physicians, are mandated by law to report any reasonable suspicion of child maltreatment.5 Despite evidence that reporting suspected child abuse or neglect provides protection against subsequent abuse for both affected and at-risk children,6–10 some physicians may suspect child maltreatment but decline to report for a variety of reasons.11–14 In addition, although abused children may have a number of health care encounters before diagnosis,15 medical personnel are the source for fewer METHODS We conducted a cross-sectional survey of physician child maltreatment reporting practices, targeting pediatricians holding an active medical license in a single state. From the Department of Pediatrics (Drs Gunn and Cooper) and the Center for Patient and Professional Advocacy (Dr Hickson), Vanderbilt University Medical Center, Nashville, Tenn. Address correspondence to Veronica L. Gunn, MD, MPH, Department of Pediatrics, Vanderbilt University School of Medicine, 5028 MCE, 1215 21st Ave South, Nashville, TN 37232-8555 (e-mail: veronica.gunn@vanderbilt.edu). Received for publication June 21, 2004; accepted August 10, 2004. AMBULATORY PEDIATRICS Copyright q 2005 by Ambulatory Pediatric Association Environment Child abuse and neglect reporting laws and protections from civil liability for individuals who file reports differ by state. To eliminate the potential effect of differences in reporting laws between states, this study was conducted in a southeastern state with over 1.4 million children 96 Volume 5, Number 2 March–April 2005 AMBULATORY PEDIATRICS Pediatricians’ Reporting of Suspected Child Maltreatment younger than 18 years of age. In 2001, more than 38 500 children in the state were the subject of reports of suspected child abuse and neglect.30 Mandated reporters in the state include ‘‘any person having knowledge of any child who is suffering from injury, disability, physical or mental condition which is of such a nature as to reasonably indicate that it was caused by brutality, abuse or neglect.’’31 Even though the statute provides criminal penalties for failure to meet reporting obligations, no health care providers have been prosecuted for failing to report. In contrast, civil lawsuits have been filed against mandated reporters because they reported suspected maltreatment.32,33 The state subsequently adopted legislation providing absolute immunity for medical providers who report child maltreatment in the course of their work.34 Subjects We randomly selected 500 physicians for participation in the study from a total of 1112 licensed pediatricians in the state’s Department of Health master file of licensed physicians. The sample size was determined by the number needed to detect a 15% difference in reporting practice with alpha of .05, power of 80, and using an estimated return rate of 55%. In the spring of 2003, participants received a 25-question survey, a prepaid self-addressed return envelope, and a cover letter from one of us (V.L.G.) that stressed that participation was voluntary and that all responses would be confidential. Those who did not respond within 3 weeks were sent the survey again; a third survey was sent to nonrespondents after another 3 weeks. Survey Instrument We developed the survey with the assistance of an expert in survey design and used a variety of question formats, including open-ended responses, yes-or-no questions, and Likert scales. Demographic information was collected about respondents, including gender, age, race (white, black, Hispanic or Latino, Asian, Native American, other), year the respondent first started practicing medicine, practice location (urban, nonurban), and practice type (primary care, hospital based, school clinic, public health). The survey, which took approximately 15 minutes to complete in pilot studies, included an open-ended section for comments about respondents’ reporting experiences. We asked providers questions about their familiarity with child abuse and neglect reporting and immunity laws, previous experiences reporting, and perceived barriers to reporting suspected maltreatment. All participants were asked if they had ever suspected that a child was abused or neglected but chose not to report that incidence of suspected maltreatment. Those who answered affirmatively were asked to list all the factors that contributed to their decision not to report. To further elucidate perceived barriers to reporting, we asked respondents to list reasons why any medical provider might not report suspected child abuse or neglect. To minimize respondent pressure to provide socially desirable responses, this question was 97 phrased as ‘‘Please list as many reasons as possible why you think OTHER medical providers may be reluctant to report suspected child abuse or neglect.’’ This technique to increase the reliability of physician responses about reporting practice has been previously described.13 Analysis Survey data were entered into a commercially available database (Microsoft Access, Microsoft Corporation, Redmond, Wash) with data-entry parameters to minimize entry errors. Two of us (V.L.G., W.O.C.) reviewed the data for accuracy, and quantitative analysis was performed with STATA 7.0 (STATA Corporation, College Station, Tex). Each of us independently reviewed qualitative data to identify common themes appearing across the surveys by a process called manifest content analysis.35 We then independently assigned codes for each identified theme; those without complete agreement were selected for further discussion. Interrater agreement on initial categorization and coding of qualitative responses was 91.6%. We discussed differences on the remaining coding decisions until consensus was achieved. This method of analysis of qualitative data has been previously used in child maltreatment research.36 Bivariate and multivariate logistic regression were used to model dichotomous dependent variables. The Institutional Review Board of Vanderbilt University Medical Center approved this study before study implementation. RESULTS Quantitative Analysis Of the 500 physicians in the original sample, 16 were deemed ineligible because they had moved or had retired. Of the remaining 484 physicians, 263 returned surveys. Of these, 66 were ineligible because they had not provided medical care to children under the age of 18 years within the previous 5 years (n 5 37) or because they had never treated a child who they believed had experienced abuse or neglect (n 5 29). Of the 197 eligible respondents, 195 completed the survey and 2 declined. The estimated response rate was 56% based on the method of the Council of American Survey Research Organizations, which calculates the completion rate as the number of eligible respondents divided by the number of eligible respondents plus the estimated number of eligible nonrespondents.37 Nonrespondents did not differ significantly from respondents in years licensed (nonrespondents [mean SD: 15.4, 11.0], respondents [14.7, 11.4]) or gender (nonrespondents: 58% men, respondents: 54% men). In addition, respondents were similar to other pediatricians in the state regarding average age (respondents: 44.5 years, other pediatricians in state: 47 years), gender (respondents: 54% men, other pediatricians in the state: 57% men), and ethnicity (respondents: 88% white, other pediatricians in state: 84% white) (B. L. Markham, personal communication, April 2004). Respondent demographics, practice characteristics, previous child abuse reporting experiences, and characteristics of nonreporters are listed in Table 1. The mean age 98 Gunn et al AMBULATORY PEDIATRICS Table 1. Characteristics and Reporting Experience of Physicians* Characteristic (n) Gender Male (105) Female (89) % Physicians Who Suspected Maltreatment but Did Not Report P† ,.01 36 18 Race White (169) Nonwhite (22) NS 30 18 Practice type Primary care (129) Hospital based (56) Public health and school based (9) 27 29 33 Years in practice 1–5 (59) 6–10 (29) 11–15 (15) 16–20 (19) 21–25 (17) 26–30 (16) 31–35 (7) $36 (9) 17 0 21 32 59 69 14 67 Practice location Urban (91) Nonurban (103) 25 31 Guidelines for reporting in practice No (90) Yes (102) 30 27 Number of cases reported 1–2 (33) 3–5 (48) 6–10 (42) 11–20 (22) $21 (25) 15 19 31 41 48 Deposed in maltreatment case No (99) Yes (94) 17 39 Testified in maltreatment case No (115) Yes (79) 21 38 Threatened with lawsuit No (168) Yes (25) 25 48 Sued for reporting maltreatment No (186) Yes (6) 27 33 NS ,.001‡ NS NS .001‡ .001 .01 Table 2. Characteristics Associated With Not Reporting Suspected Child Maltreatment* Characteristic Odds Ratio 95% Confidence Interval Male gender More years in practice Greater number of cases reported 2.18 1.23 1.28 1.05–4.49 1.05–1.44 1.02–1.60 *Multivariate logistic regression models included respondents’ gender, decision to report suspected maltreatment, number of years in practice, number of cases reported, experience with depositions and testifying in child maltreatment cases, and experience being threatened with civil suit for reporting. ported that their practice location had specific guidelines for reporting suspected child maltreatment. Ninety-six percent of respondents had reported suspected child maltreatment at least 1 time since completing residency training. Nearly half (49%) had been deposed, and 41% had testified in court in a child maltreatment case. Thirteen percent of respondents reported that they had been threatened with a lawsuit for reporting suspected abuse, and 3% indicated that they had been sued for reporting suspected child abuse. When asked ‘‘Have you ever considered reporting suspected child abuse or neglect, but chose not to do so,’’ 28% of respondents indicated that they had, at one point in time, considered reporting a case of suspected child maltreatment but had decided not to report. Logistic regression analysis of the independent variables revealed that providers who chose not to report were more likely to be men, to have been in practice longer, to have reported more cases, to have had more experience giving depositions or testifying in child maltreatment cases, and to have been threatened with lawsuit (Table 1). In multivariate logistic regression models, male gender, more years in practice, and greater experience reporting remained significantly associated with not reporting suspected abuse (Table 2). Qualitative Analysis .02 NS *Some respondents did not answer every question. †P values determined by bivariate logistic regression with decision not to report as the main outcome variable. This P value reflects the difference between those who have ever chosen not to report and those who have ‘‘always’’ reported. ‡P value refers to trend across group. of respondents was 44.5 years (range 29–83), just over half (54%) were men, and the majority (88%) self-identified as white. Of the total, 66% practiced in a primary care setting and 29% were hospital based. Nearly half (47%) practiced in urban areas, and over half (53%) re- One hundred fifty-three respondents offered 363 reasons why other physicians might be reluctant to report suspected child abuse and neglect. Providers most often expressed concern about the potential consequences of reporting for the provider or child. For example, responses in this category included phrases such as ‘‘fear of alienating the family or precipitating a crisis which could result in harm to the child,’’ ‘‘physically threatened by a family member,’’ and ‘‘personal safety, lawsuits, and concerns about child’s safety if intervention is not adequate.’’ Others cited hassle and time involved in reporting as deterrents, for example, ‘‘I think time is a major factor—knowing that you will have to take time away from your practice to testify’’ and ‘‘unwilling to take time, disrupt schedule and deal with lawyers.’’ Providers also expressed fear of being wrong or of wrongly accusing a family: ‘‘afraid to be wrong or to wrongly accuse someone,’’ ‘‘fear of damaging a family’s social status if it turns out that abuse AMBULATORY PEDIATRICS Pediatricians’ Reporting of Suspected Child Maltreatment has not really occurred.’’ Other factors cited included uncertainty that maltreatment occurred (eg, ‘‘lack of confidence in diagnosis’’), denial that maltreatment could occur (eg, ‘‘can’t believe this family could be abusive—the ‘nice family syndrome’’’), and avoidance of angry parents. For those who indicated that they had chosen not to report, we asked them to list all factors that influenced their decision not to report. Of those respondents, only 17% stated that they did not report because further assessment led them to change their beliefs. Conversely, when asked why other physicians might not report suspected maltreatment, none of the 153 respondents indicated that their colleagues might not report because further information led them to change their beliefs. Respondents who had chosen not to report suspected child abuse or neglect were more likely to cite lack of knowledge about reporting laws and process (eg, ‘‘not knowing who to report to and the mechanism in place to do so’’) (P 5 .05) and previous poor experiences with child service agencies (eg, ‘‘frustration with inaction and difficulty contacting appropriate people’’) (P 5 .03) as reasons for not reporting than were pediatricians who had always reported. DISCUSSION Twenty-eight percent of pediatricians surveyed considered reporting suspected maltreatment in specific instances but chose not to do so. Male physicians, those in practice longer, and those with more reporting experience were more likely to have not reported suspected maltreatment. The most commonly expressed reasons for not reporting included concerns about consequences of reporting for the provider, the child, or for the family. The rate of nonreporting in our study is higher than some published rates of nonreporting by US physicians (8%) and educators (11%)11,18 but lower than physician nonreporting in a case-based survey of Australian physicians (43%).12 Although we used a similar metric to define nonreporting as other studies,11,12,18 to our knowledge this survey represents the first study of US pediatricians to use a method of eliciting physician behaviors that minimized respondent pressure to provide socially desirable responses. Many surveys use a case-based approach where professionals are asked which clinical vignettes they would report4,12,17–23; others ask subjects directly to describe their own reporting behaviors.11,12,18,24,25 In addition to asking about personal reporting experiences, our survey asked respondents to list why other pediatricians might not report suspected child maltreatment in order to acknowledge the moral demands and potential legal implications of not reporting. Furthermore, only 2 other US studies of physician reporting have surveyed pediatricians,11,17 and these studies had smaller samples. This study also reveals that male physicians were more likely to choose not to report suspected child maltreatment. The proclivity of male physicians not to report suspected maltreatment has been noted previously.12,20 In addition, physicians in practice longer were more likely not to report suspected maltreatment. One possibility is 99 that the more years of experience a physician has, the more likely he or she will be to consider child maltreatment as a part of a differential diagnosis but later exclude it based on either their acquired knowledge or previous unconstructive interactions with child protection agencies. Studies of reporting practices have identified that some professionals do not report because they lack confidence in the child service agency’s ability to handle their large case loads; to provide a competent, prompt response; and to provide feedback about the outcome of the report.13,17–19,26,27 Another possible explanation for nonreporting is that the current system may reinforce the practice of nonreporting. A combination of poor feedback from child service agencies, an awareness of civil suits or threats of suits against mandated reporters, and a lack of legal accountability for nonreporting may have collectively created an environment that dissuades reporting. Alternatively, the high rate of nonreporting in our study may reflect providers’ efforts to be more discerning in their decision-making regarding what is a reasonable level of suspicion. In the qualitative analysis, respondents listed the hassles associated with reporting (eg, time involved, going to court, amount of paperwork) as one of the most common reasons that suspected maltreatment is not reported. Therefore, it is not surprising that those providers who had previous experience giving depositions or testifying in child maltreatment cases (Table 1) were more likely not to report. Providers who had chosen not to report any suspected child abuse or neglect were also more likely to cite lack of knowledge about reporting laws as a reason for not reporting. Many respondents stated they were reluctant to report until they were certain of the diagnosis of abuse or neglect, despite that in the study state—and in most states—one only has to have a ‘‘reasonable suspicion’’ of maltreatment in order to report.38 The reluctance to report until the physician is convinced maltreatment has occurred has been noted in other studies as well.11,24,39 Other respondents indicated lack of knowledge about the process of reporting (‘‘don’t know who to report to’’) and about their role in the reporting process (‘‘minimal concrete substantiating evidence’’ and ‘‘lack of proof or evidence’’). Such statements suggest that providers may benefit from education regarding the reporting process and the physician’s role in that process. Previous studies have demonstrated that physician education may improve the probability that physicians report suspected maltreatment.11,28,29,40 Limitations of this study include possible selection factors; 56% of eligible physicians participated in this study. Other published child abuse reporting surveys have yielded similar response rates.13,14,24 Because our study depended upon respondent recall of their practice behavior, results may be affected by recall bias, or respondents may have felt compelled to give the socially acceptable answer. We tried to address the latter concern by using a previously published technique of having respondents describe both their own behaviors as well as reasons for other’s nonreporting behavior.13 In addition, participants were as- 100 Gunn et al sured that their responses were confidential. Participants were asked if they had ever considered reporting abuse or neglect but had chosen not to do so. Although this question was purposefully worded in such a manner as to allow open-ended explanations of their reasoning, rather than potentially leading respondents to an answer or limiting their response options, it is possible that the lack of definition of ‘‘considered’’ may have confused some respondents. Two study findings appear to be contradictory. Pediatricians with more experience reporting and those with experience with depositions and testifying in child maltreatment cases were less likely to report suspected maltreatment. In contrast, lack of knowledge was a commonly offered reason as to why other physicians might not report suspected child maltreatment. One explanation is that although physicians with more experience reporting may choose not to report for reasons previously noted, their experience may help them identify many reasons why colleagues might not report. Finally, because our study was conducted in a single state to eliminate the effect of different reporting laws and protections in various states, results may not be representative of physician child maltreatment reporting practices in other states. Results of this study provide important insights into the reporting behavior of pediatricians. Throughout the qualitative portion of the survey, fear was a recurrent theme: fear of lawsuit, fear of physical harm from families, and fear of the apparent impact of reporting on children and families. Physicians’ fear of civil suit in the study state still exists despite the state’s provision for absolute immunity for medical reporters of suspected abuse or neglect.34 The fear of repercussion of reporting may be reflected in the finding that physicians threatened with lawsuit were more likely to choose not to report suspected maltreatment. Providers’ comments also revealed uncertainty about their diagnosis of child maltreatment as well as uncertainty about the level of suspicion they needed to have in order to report. These physician concerns, coupled with a lack of knowledge about reporting laws and processes, create an opportunity for at-risk children to become victims of continued abuse. Results suggest the need for new educational approaches addressing fear as mitigating factor for reporting behavior. In addition, comments regarding concerns about child safety after reporting suggest that providers might be more willing to report if they knew more about the outcome of their reports. Our study illustrates that opportunities exist for continued improvement in physician reporting practice. Physician child abuse education programs should be a required part of residency training and continuing medical education. Such programs should be directed toward specific concerns elucidated in this study: concerns that reporting will have untoward consequences for the child or provider, inadequate knowledge about reporting laws and process, the capabilities of child service agencies, and—importantly—the necessity to report suspected maltreatment to protect at-risk children from injury. Despite the clear need for ongoing training in child abuse recognition and re- AMBULATORY PEDIATRICS porting, such educational programs are not yet a requirement of the Pediatric Residency Review Committee.41 Child abuse and neglect continue to threaten the safety and health of children. Many physicians who suspect child maltreatment do not report the abuse or neglect to appropriate authorities for a variety of reasons. The implications of this study suggest that specific educational interventions may be helpful in eliminating potential barriers to reporting. Further studies are needed to assess the impact of initiatives to improve reporting of child maltreatment by medical professionals. ACKNOWLEDGMENT We thank Tony N. Brown, PhD, for his assistance in the development of the survey. REFERENCES 1. US Department of Health and Human Services, Administration on Children, Youth and Families. Child Maltreatment 2001. Washington, DC: US Government Printing Office; 2003. 2. US Department of Health and Human Services, Administration on Children, Youth and Families. Child Maltreatment 2000. Washington, DC: US Government Printing Office; 2002. 3. US Department of Health and Human Services, Administration on Children, Youth and Families. Child Maltreatment 1999. Washington, DC: US Government Printing Office; 2001. 4. Warner-Rogers JE, Hansen DJ, Spieth LE. The influence of case and professional variables on identification and reporting of physical abuse: a study with medical students. Child Abuse Negl. 1996;18:11–25. 5. National Clearinghouse on Child Abuse and Neglect Information. Child abuse and neglect: statutes at-a-glance: mandatory reporters of child abuse and neglect. Available at: http:// www.calib.com/nccanch/statutes/index.cfm. Accessed January 9, 2004. 6. DePanfilis D, Zurvain S. The effect of services on the recurrence of child maltreatment. Child Abuse Negl. 2002;26:187– 205. 7. Way I, Chung S, Johnson-Reid M, Drake B. Maltreatment perpetrators: a 54-month analysis of recidivism. Child Abuse Negl. 2001;25:1093–1108. 8. DePanfilis D, Zurvain S. Predicting child maltreatment recurrences during treatment. Child Abuse Negl. 1999;23:729–743. 9. DePanfilis D, Zurvain SJ. Epidemiology of child maltreatment recurrences. Soc Serv Rev. 1999;73:218–239. 10. Herrenkohl RC, Herrenkohl EC, Egolf B, Seech M. Final report from the project: an investigation of the effects of a multidimensional service program on recidivism/discontinuation of child abuse and neglect. Prepared for the National Center on Child Abuse and Neglect, Grant No. 9SC-428. Bethlehem, Pa: Lehigh University, Center for Social Research; 1978. 11. Flaherty EG, Sege R, Binns HJ, et al. Health care providers’ experience reporting child abuse in the primary care setting. Arch Pediatr Adolesc Med. 2000;154:489–493. 12. Van Haeringen AR, Dadds M, Armstrong KL. The child abuse lottery—will the doctor suspect and report? Physician attitudes towards and reporting of suspected child abuse and neglect. Child Abuse Negl. 1998;22:159–169. 13. Vulliamy AP, Sullivan R. Reporting child abuse: pediatricians’ experiences with the child protection system. Child Abuse Negl. 2000;24:1461–1470. 14. Zellman GL. The impact of case characteristics on child abuse reporting decisions. Child Abuse Negl. 1992;16:57–74. 15. Jenny C, Hymel KP, Ritzen A, et al. Analysis of missed cases of abusive head trauma. JAMA. 1999;282:621–626. 16. Prevent Child Abuse America. Total Estimated Cost of Child Abuse and Neglect in the United States: Statistical Evidence. Chicago, Ill: Prevent Child Abuse America; 2001. AMBULATORY PEDIATRICS Pediatricians’ Reporting of Suspected Child Maltreatment 17. Morris JL, Johnson CF, Clasen M. To report or not to report: physicians’ attitudes towards discipline and child abuse. Am J Dis Child. 1985;139:194–197. 18. Kenney MC. Child abuse reporting: teachers’ perceived deterrents. Child Abuse Negl. 2001;25:81–92. 19. Crenshaw WH, Crenshaw LM, Lichtenberg JW. When educators confront child abuse: an analysis of the decision to report. Child Abuse Negl. 1995;19:1095–1113. 20. Finlayson LM, Koocher GP. Professional judgment and child abuse reporting in sexual abuse cases. Prof Psychol Res Pract. 1991;22:464–472. 21. Zellman GL. Report decision-making patterns among mandated child abuse reporters. Child Abuse Negl. 1990;14:325–336. 22. Tite R. How teachers define and respond to child abuse: the distinction between theoretical and reportable cases. Child Abuse Negl. 1993;17:591–603. 23. O’Toole R, Webster S. Teachers’ recognition and reporting of child abuse: a factorial survey. Child Abuse Negl. 1999;23: 1083–1101. 24. Saulsbury FT, Campbell RE. Evaluation of child abuse reporting by physicians. Am J Dis Child. 1985;139:393–395. 25. Abrahams N, Casey K, Daro D. Teachers’ knowledge, attitudes and beliefs about child abuse and its prevention. Child Abuse Negl. 1992;16:229–238. 26. Compaan C, Doueck HJ, Levine M. Mandated reporter satisfaction with child protection: more good news for workers? J Interpers Violence. 1997;12:847–857. 27. Brosig CL, Kalichman SC. Child abuse reporting decisions: effects of statutory wording of reporting requirements. Prof Psychol Res Pract. 1992;23:486–492. 28. Flaherty EG, Sege R, Mattson CL, Binns HJ, for the Pediatric Research Group. Assessment of suspicion of abuse in the primary care setting. Ambul Pediatr. 2002;2:120–126. 101 29. Woolf A, Taylor L, Melnicoe L, et al. What residents know about child abuse: implications of a survey of knowledge and attitudes. Am J Dis Child. 1988;142:668–672. 30. Martinez L. Department of Children’s Services: Tennessee Child Abuse and Neglect Totals: Report to the Tennessee Legislature; 2001. 31. Tennessee Annotated code §37-1-401(a). 32. Harville v Vanderbilt University, Inc, 2003 US 18053. 33. Bryant-Bruce v Vanderbilt University, Inc, 974 F Supp 1127 (MD Tenn 1997). 34. Tennessee Annotated code §37-1-410(a). 35. Field PA, Morse JM. Nursing Research: The Application of Qualitative Approaches. Rockville, Md: Aspen Systems Corp; 1985. 36. Deisz R, Doueck HJ, George N. Reasonable cause: a qualitative study of mandated reporting. Child Abuse Negl. 1996;20:275– 287. 37. Council of American Survey Research Organizations. On the definition of response rates: a special report of the CASRO task force on completion rates. Available at: http://www.casro.org/ resprates.cfm. Accessed August 14, 2003. 38. Tennessee Commission on Children and Youth. Tennessee Compilation of Selected Laws on Children, Youth and Families. Charlottesville, Va: Lexis Law Publishing; 1999. §37-1-403. 39. US Department of Health and Human Services. Executive Summary of the Third National Incidence Study of Child Abuse and Neglect (NIS-3). Washington, DC: US Government Printing Office; 1999. 40. Lawrence LL, Brannen SJ. The impact of physician training on child maltreatment reporting: a multi-specialty study. Mil Med. 2000;165:607–611. 41. Residency Review Committee. Program Requirements for Residency Education in Pediatrics. Available at: http://www. acgme.org. Accessed April 8, 2004.