Pediatrics, Vol. 51, No. 5, May 1973, 840-848.

Reducing the Literal and Human Cost of Child Abuse: Impact of a New Hospital Management System

Eli H. Newberger, M.D., John J. Hagenbuch, A.C.S.W., Nancy B. Ebeling, A.C.S.W., Elizabeth Pivchik Colligan, A.C.S.W., Jane S. Sheehan, R.N., and Susan H. McVeigh, B.A.

From the Children’s Hospital Medical Center, the Division of Family and Children’s Services of the Massachusetts Department of Public Welfare, and Children’s Protective Services, Boston, Massachusetts

 

ABSTRACT. Social service personnel from one public and two voluntary agencies were integrated into a consultation group in an academic pediatric hospital, leading to a reduction in the actual cost of medical services and the risk of reinjury subsequent to the diagnosis of child abuse. In the 1969-1970 hospital year, 62 cases of child abuse were seen, of which 39 were hospitalized. The average hospital stay was 29 days; the average hospital cost $3,000. Total hospital costs for the 39 cases were $123,000, of which bed costs made up $95,000. There were at least three subsequent incidents of child abuse in these 39 cases, and there was one subsequent death; the reinjury rate was 10% for hospitalized cases.

In September 1970 the Trauma X Group, an interdisciplinary, interagency consultation unit based in the hospital, was formed. With formal consultation and continued surveillance after discharge by the Trauma X Group, the following data were obtained from the 1970-1971 hospital year. Of 86 cases, 60 were hospitalized. The average hospital stay was 17 days; the average hospital cost $2,500. Total hospital costs for the 60 cases was $150,000, of which bed costs made up $101,000. There was one incident of reinjury and no deaths subsequent to diagnosis in these 60 cases; the reinjury rate was 1.7%. The risk of reinjury calculated from a modified life table was reduced from 8% in the year previous to the formation of the group to 7% and 2%, respectively, in the subsequent year and six-month periods, supporting the dollar-cost impression of effectiveness. Foster placement, furthermore, was infrequent and does not explain the differential impact of the Trauma X Group in the intervals under study. Pediatrics, 51:840, 1973, CHILD ABUSE, BATTERED CHILD SYNDROME, QUALITY OF CARE, EVALUATION OF FAMILY INTERVENTION.

The challenge posed by the initial recognition or suspicion of child abuse is traditionally met by the personnel of a hospital in an urban setting in a variety of ways, ranging from frank refusal to accept the diagnosis to a quick proferring of a variety of professional services.¹ Occasionally, but not always, a long-range therapeutic plan, with particular attention to continuing supportive relationships, is developed by physician, nurse, psychiatrist, and social worker as the diagnostic formulation and communication with the relevant community agencies proceeds.

Experience at our hospital and elsewhere indicates, however, that even the best-laid plans seem frequently to be frustrated by a combination of factors associated with those severe family crises where a child’s life falls into jeopardy.² These include the personalities of his parents, for whom denial and projection often serve as principal means of ego defense³; his family’s anxious confusion in confronting an array of clinical specialty services and social agencies working disconjugately to protect their child from themselves^4,5; the exigencies of life in poverty, including mistrust of community institutions, racism, unemployment, and drugs⁶; the clinical team’s frustration generated by missed appointments, confrontations with angry parents, and time-consuming contacts with outside agencies; and conflicts among the responsible personnel stemming from the emotions brought forth by prolonged contact with disturbed families.⁷

TABLE 1 SOCIAL DIAGNOSTIC DATA FOR PRESENTATION AT TRAUMA X MEETING
		Date Division/Clinic Patient’s First Name DOB___________________
Social Worker:______________________________________		Address_________________
REFERRAL:_______________	DATE:_______________
DESCRIPTION OF INJURY:______________________________________________________
FAMILY CONSTELLATION:
Names/B.D.-Parents:
Ethnic Group:
Employment-Source of Income:
Names/B.D.-Siblings:
PREVIOUS AND CURRENT COMMUNITY CONTACTS:
Worker:
Agency:
When:
INTERVIEWS WITH PARENTS:	DATES:	WHERE:
Attitude toward injury/hospitalization:
Perceptions/Expectations of child:
Parents’ view of own upbringings:
Family Stresses:
POTENTIAL STRENGTHS:
1. Family
2. Supportive personal ties: Neighborhood Agency School Church, etc.
DIAGNOSTIC IMPRESSION:
TREATMENT PLAN:
GAPS IN MANAGEMENT:
IMPRESSIONS OF OUTCOME: Without intervention With recommended plan
QUESTIONS FOR DISCUSSION:
TRAUMA MEETING NOTES:

No one can adequately measure the human cost of child abuse. The present study is designed to demonstrate the effectiveness of a collaborative community effort to provide more nearly adequate preventive and therapeutic help to the victims of child abuse and their families. The evaluation of dollar costs and of a very crude measure of the behavioral effectiveness of intervention—the reinjury of a child—is intended both to promote awareness of the considerable literal and human expense of these cases, and to show how this toll can be reduced by an effort to coordinate human services of a kind which in the United States are generally isolated in State Departments of Welfare and fragmented among institutions whose programs attend to specific aspects of child health, mental health, and social welfare.

METHODS

In September 1970 an interdisciplinary, interagency consultation unit, the Trauma X Group, was formed at Children’s Hospital in response to a crisis in the system for the protection of children living in Massachusetts.⁸

A shortage of personnel in the state agency designated by law to receive reports of child abuse cases and to provide protective services led to an appraisal of the prevailing method of hospital case management. Until that time, physicians had simply reported cases of suspected inflicted injury to the Department of Public Welfare, whose resources to deliver protective social services were severely limited. The number of injuries subsequent to initial diagnosis were apparently high, and our staff-as well as our colleagues in the Welfare Department-agreed that a more systematic program of case-finding, evaluation, intervention, and follow-up was necessary.

Accordingly, a group of interested individuals reviewed the problem; house management guidelines were written; and a concerted effort was made to recruit the interest and support of the chiefs of the clinical departments and the administration of the hospital. The functions of the Trauma X Group were defined as follows:

1. To become informed of each new case of inflicted injury or neglect.

2. To participate in the formulation of the plan for disposition and continuing care of the child.

3. Periodically to review the entire active hospital roster of Trauma X cases for purposes of evaluation and quality control.

4. To develop, in conjunction with the Department of Public Welfare, a unified proposal to the Juvenile Court in those cases where legal action was necessary. A designated member of the Trauma X Group would represent the hospital staff in court in such situations.

5. To educate the house staff, social service staff, and nursing staff in the pathogenesis of the problem and enlightened approaches to its control.

6. To serve as a focus at the Hospital for advancing the competence of the Boston community to deal effectively with child abuse and neglect. This function included the development of better channels of communication with other agencies and the discussion of health policy alternatives which affected the index population.

7. To carry on research into the determinants and concomitants of child abuse and neglect. This role integrated the responsibilities to monitor the clinical progress of the hospital population and to advance knowledge of the field in etiology and in the methodology of intervention.

A management model for the management of all Trauma X cases was proposed to the hospital staff at the time of the organization of the Trauma X Group (Fig. 1).

The three social agencies who regularly offer consultation on the management of Children’s Hospital cases of child abuse, through participation in the Trauma X Group, are the Department of Public Welfare (Inflicted Injury Unit, Division of Family and Children’s Services), and two voluntary agencies, Children’s Protective Services and Parents’ and Children’s Services. Regular representation from the Children’s Hospital staff at Trauma X meetings comes from the administration and from the departments of medicine, psychiatry, radiology, social service, and nursing. Legal consultation is acquired through several sources: from the hospital’s own legal counsel, from a lawyer associated with the Laboratory of Community Psychiatry at the Harvard Medical School, and from a consultant retained for certain protective actions in the Juvenile Court.

A full-time Case Data Coordinator arranges case conferences, takes detailed minutes of the meetings, informs the various personnel involved in management of new developments, and reviews and updates the data on all cases in the file.

Three to five new cases are discussed at a two-hour luncheon conference each week. An effort is made to create an atmosphere which is congenial and unintimidating, both for the personnel from community clinical practice settings who come to the hospital to discuss their cases and for the benefit of our own house officers and professional staff, from whom an adequate flow of information on these cases is essential.

Systematic review of the data pertinent to the formulation of management judgments is helped by a short handout on each case, which is prepared in advance by the hospital social worker. The format of the handout is displayed in Table I. In December 1972, a problem-oriented record keeping system was introduced for Trauma X cases. It will be reported subsequently.

RESULTS

Figure 2 summarizes the Trauma X patients seen at the Children’s Hospital in the hospital years before and after the organization of the Trauma X group in September 1970.

Figure 3 summarizes the literal cost of inpatient Trauma X cases at Children’s Hospital. With a reduction of the mean hospital stay from 29 to 17 days in the year following the organization of the Trauma X Group, the disproportionate contribution of the day rate to the total cost was changed appropriately, demonstrating somewhat more efficient management by this criterion of effectiveness. (These fiscal data have been particularly useful in contacts with Welfare Department officials and state legislators in an effort to improve services for these families.)

The rates of rein jury among the hospitalized cases in the hospital years before and after the formation of the Trauma X Group are summarized in Table II.

TABLE II
REINJURY AFTER DIAGNOSIS*
	1969-1970	1970-1971
Reinjuries after Trauma X diagnosis	3	1
Deaths subsequent to diagnosis	1	0
Rate of reinjury	4/39=10% (data to 1/1/72)	1/60=1.7%
* Trauma X inpatients, Children’s Hospital Medical Center, Boston, Massachusetts

A more sensitive and accurate estimate of the risk of reinjury can be computed by constructing a life table which measures the total number of person-months contributed by the population of cases under observation in a given time interval. Table III displays the results of the life table analysis of the data from the calendar year preceding the formation of the Trauma X Group and for the subsequent year and half-year intervals. The details of the calculation of risk for the periods under evaluation (9/15/69 to 9/14/70,9/15/70 to 9/14/71, and 9/15/71 to 2/14/72) are shown on Tables IV, V, and VI. Follow-up information complete to May 1, 1972, is included in this table. The tests of significance of these data are summarized in Figure 4.

TABLE III
TRAUMA X CASES CUMULATIVE LIVE TABLE DATA
	9/15/69 to 9/14/70	9/15/70 to 9/14/71	9/15/7 to 2/14/72
Person-months observed	307	648	150
Reinjuries in the interval	4	5	1
Risk of reinjury	8%	7%	2%

 

TABLE IV
LIFE TABLE: 9/15/69 to 9/14/70
Month After Entry X	Observed At X	Last Anniversary At X	Observed X to (X+1)	First Reinjury X to (X+1)
0-	50	0	50	0
1-	50	3	47	2
2-	45	3	42	1
3-	41	1	40	1
4-	39	3	36	0
5-	36	8	28	0
6-	28	9	19	0
7-	19	5	14	0
8-	14	0	14	0
9-	14	5	9	0
10-	9	3	6	0
11-	6	4	2	0
12-	2
	Probability of Reinjury X to (X+1)	Probability of Survival X to (X+1)	Cumulative Probability of Survival to X
0-	0	1.00	1.00
1-	.04	.96	1.00
2-	.02	.98	.96
3-	.02	.98	.94
4-	0	1.00	.92
5-	0	1.00	.92
6-	0	1.00	.92
7-	0	1.00	.92
8-	0	1.00	.92
9-	0	1.00	.92
10-	0	1.00	.92
11-	0	1.00	.92
12-	0	1.00	.92
Risk of reinjury in the interval=8%

 

TABLE V
LIFE TABLE: 9/15/70 to 9/14/71
Month After Entry X	Observed At X	Last Anniversary At X	Observed X to (X+1)	First Reinjury X to (X+1)
0-	107	0	107	0
1-	107	6	101	3
2-	98	11	97	0
3-	87	9	78	0
4-	78	5	73	1
5-	72	16	56	0
6-	56	8	48	0
7-	48	7	41	0
8-	41	9	32	1
9-	31	14	17	0
10-	17	10	7	0
11-	7	6	1	0
12-	1
	Probability of Reinjury X to (X+1)	Probability of Survival X to (X+1)	Cumulative Probability of Survival to X
0-	.00	1.00	1.00
1-	.03	.97	1.00
2-	.00	1.00	.97
3-	.00	1.00	.97
4-	.01	.99	.97
5-	.00	1.00	.96
6-	.00	1.00	.96
7-	.00	1.00	.96
8-	.03	.97	.96
9-	.00	1.00	.93
10-	.00	1.00	.93
11-	.00	1.00	.93
12-
Risk of reinjury in the interval=7%

 

TABLE VI
LIFE TABLE: 9/15/71 to 2/14/72
Month After Entry X	Observed at X	Last Anniversary at X	Observed X to (X+1)	First Reinjury X to (X+1)
0-	51	0	51	1
1-	50	9	41	0
2-	41	12	29	0
3-	29	7	22	0
4-	22	15	7	0
5-	7
	Probability of Reinjury X to (X+1)	Probability of Survival X to (X+1)	Cumulative Probability of Survival to X
0-	.02	.98	1.00
1-	0	1.00	.98
2-	0	1.00	.98
3-	0	1.00	.98
4-	0	1.00	.98
5-
Risk of reinjury during this interval=2%

DISCUSSION

The impression of effectiveness in reducing the human cost of child abuse suggested by the crude rates of reinjury is supported equivocally by the life table data, which are not significant when treated by the standard method.⁹ Cautious interpretation of this analysis is indicated, furthermore, because the outcome data are affected by an ascertainment bias comprising three principal elements:

1. Follow-up information was more easily available once the program of regular surveillance began.

2. Earlier cases have had more time in which to be exposed to reinjury.

3. With increasing diagnostic sophistication, cases with more subtle clinical signs have been included in the index population.

The effects on the outcome data of the bias of ascertainment are summarized on Table VII.

TABLE VII
ASCERTAINMENT BIAS
Effect on Outcome Data(+=toward better outcome)
	1969-1970	1970-1971	1971-1972
Adequacy of follow-up data	+	–	–
Length of follow-up interval	–	+/-	+
Breadth of definition of Trauma X	–	+	+

Were children with less serious injuries preferentially selected for inclusion in the series as time went on, the likelihood of reinjury might also have progressively diminished. The nature of the presenting symptoms in the cases in the two hospital years under study indicates that the severity of their injuries did not change. These data are summarized in Table VIII. An indirect inference of the risk to the children in the series may be drawn from the rates at which the judgment was made to separate them from their families. As shown in Table IX, these rates increased slightly. Interestingly, they approximate Kempe’s estimate of the number of child abuse cases in which drastic protective intervention, placing children in foster care, is necessary.”¹⁰

TABLE VIII
PRESENTING SYMPTOMS: TRAUMA X CASES
	July 1, 1969, To June 30, 1970	July 1, 1970, To June 30, 1971	July 1, 1971, To March 30, 1972
Symptoms:
Bruises	19	17	36
Burns	5	16	8
Skull fractures	6	6	10
Other bone fractures	11	9	12
Neglect	15	14	16
Head injuries	2	6	7
Lacerations	2	4	3
Poisonings	1	6	4
Abandonments	1	2	1
Deaths
Inpatients	39	60	91
Outpatients	23	26	42
Total cases	62	86	133

 

TABLE IX
FOSTER PLACEMENT: TRAUMA X CASES
	1969-1970	1970-1971	1971-1972
Patients	62	86	93
Placed in foster care	9	16	20
Rate	9/26=15%	16/86=19%	20/93=22%

There is a discrepancy, however, between Kempe’s conception of “the battered child” and our definition of “Trauma X.” In his classic paper, Kempe defined the syndrome as follows: “The battered child syndrome is a term used by us to characterize a clinical condition in young children who have received serious physical abuse, generally from a parent or foster parent.”¹¹

The Children’s Hospital euphemism, Trauma X, is defined as a syndrome with or without inflicted injury, in which a child’s survival is threatened in his home.

This definition focuses on the risk to a child rather than on the intentions of a family which has not adequately been able to protect him. In distinction to the process of making a diagnosis of child abuse in the conventional sense, where the intent to batter is estimated from contacts with the family, this diagnostic concept tries to measure the capacity of parents to protect their children.

In making a diagnosis of child abuse, a clinician can use his technical and human skills to identify what has gone wrong in the family’s ability to nurture a child and in an unpunitive way to help them solve their problem with him. Furthermore, these data seem to show that a hospital can serve as an effective portal of entry into the child health and welfare service system for disorganized families whose children’s lives are in jeopardy.

REFERENCES

1. Holter, J. C., and Frideman, S. B.: Principles of management in child abuse cases. Amer. J. Orthopsychiat., 38: 127, 1968.

2. Rowe, D. S., Leonard, M. F., Seashore, M. R, Lewiston, N. J., and Anderson, F. P.: A hospital program for the detection and registration of abused children. New Eng. J. Med., 282:17, 1970.

3. Steel, B. F., and Pollack, C. B.: A psychiatric study of parents who abuse infants and small children. In Helfer, R. E., Kempe, C. H., eds.: The Battered Child. Chicago, Illinois: University of Chicago Press, 1968, p. 130.

4. Helfer, R E., and Kempe, C. H.: The child’s need for early recognition, immediate care and protection. In Helping the Battered Child and His Family. Philadelphia: I B. Lippincott, 1972, pp. 70-71.

5. Silver, L. B., Dublin, C. C., and Louris, R S.: Agency section and interaction in cases of child abuse. Social Casework, pp. 164-71, 1971.

6. Gil, D. G.: Violence Against Children. Cambridge, Massachusetts: Harvard University Press, 1970.

7. Galdston, R: Violence begins at home. J. Amer. Acad. Child Psych., 10:2, 1971.

8. Newberger, E., Hass, G., and Mulford, R: Child abuse in Massachusetts. Mass. Physician, 32: 31, 1973.

9. Cutler, S. J., and Ederer, F.: Maximum utilization of the life table method in analyzing survival. J. Chronic Dis., 8:699, 1958.

10. Kempe, C. H.: Pediatric implications of the battered baby syndrome. Arch Dis. Child., 46:24, 1971.

11. Kempe, C. H., Silverman, F. N., Steele, B. F., Droegemueller, W., and Silver, H. K.: The battered child syndrome. JAMA, 181:1, 1962.

Acknowledgment

The authors wish to thank Mrs. Joanne B. Bluestone, Associate Hospital Director, Miss Elizabeth Maginnis, formerly Director, Social Service Department, and Drs. C. A. Janeway and J. B. Richmond, respectively, Chiefs, Departments of Medicine and Psychiatry, Children’s Hospital Medical Center, for their enthusiastic support of this work. The able technical assistance of Ms. Ann Marshall, Ms. Anne Matthews, and Ms. Geraldine Farrell is gratefully acknowledged.

 

(Received November 1, 1972; revision accepted for publication January 22, 1973.)

The work reported in this paper was supported in part by grants from the Office of Child Development, Department of HEW (Project OCD-62-141), and from the Office of Justice Administration, Boston, Massachusetts (Project 70-046). It was presented in part at the Annual Meeting of the American Pediatric Society in May 1972.