The Death of a Child With Diabetes from Neglect A Case

Study

Gary Geffken, Ph.D.* Suzanne Bennett Johnson, Ph.D.* Janet Silverstein, M.D.** Arlan Rosenbloom, M.D.**

Summary: This case report of a child with insulin-dependent diabetes mellitus (IDDM) describes a naturally occurring ABABCA design. The A condition represents poor diabetes management provided in the home setting, and the B and C conditions represent improved diabetes management in residential (condition C). The A condition was consistently associated with episodes of diabetic ketoacidosis, high glycosylated hemoglobin percentage, and school failure. In contrast, the B/C conditions were treatment

(condition B)

or

foster

care

consistently associated with improved health status and school performance. two occasions, the child was returned to condition A by the state’s protective

On

service agency (HRS) in direct conflict with the recommendations of the child’s psychological and medical treatment staff. During her last condition A placement,

Chronically ill children who are neglected may not receive the protection they need because of lack of awareness about the psychosomatic aspects of their problem. the youngster died.

Introduction

edical neglect is

a

prob-

/ lem that is often seen as less serious than other forms of child maltreatment. One

V

**

* Department of Psychiatry Department of Pediatrics University of Florida Gainesville, Florida Address correspondence to: Gary Geffken, Ph.D., Assistant Professor, University of Florida, P.O. Box 100234, Gainesville, FL 32610-0234

study’

ical

yers,

ousness, after

found that doctors, lawsocial workers, and lay respondents rated case vignettes of medical neglect as less serious than physical abuse, even when the medical neglect may have resulted in more serious physical damage. Another study, ~ using a similar methodology, found episodes of medical neglect were perceived as less serious than physical abuse, even when the medical neglect required hospitalization and the physical abuse did not. In a recent survey,3 med-

neglect ranked fourth in seriphysical abuse, sex-

ual abuse, mistreatment. These

and

studies

suggest

that

and

sexual physical abuse are more dramatic and emotionally stirring than medical neglect, even when the consequence of medical neglect may be cases

as,

or

of

emotional

more,

physically damaging

serious. In an address to the Section on Pediatrics for the American Medical Association, C. Henry Kempe4 noted the need for or

Downloaded from cpj.sagepub.com at SIMON FRASER LIBRARY on June 12, 2015

325

further education of pediatricians about atypical presentations of child abuse. He described a variety of conditions that present as medical problems but are actually the result of child abuse. He stressed the need for training of otherwise well-informed pediatricians who serve as primary child advocates. We wonder ifwe can ask for less for our child-protection service workers. We report the case of a child with diabetes who died of neglect. In describing signs of medical neglect, the AMA Council on Scientific Affairs55 lists as the first indication, &dquo;lack of appropriate medical care in the presence of chronic illness.&dquo; The case study we have reported poignantly illustrates the danger when chronically ill children are served by state child protection agents who may not have all the information they need about the medical implications of their decisions. The ABABCA design described in this report is a variant of the ABAB design, a single-subject methodology commonly used in psychological research when sample size is limited. The method involves observing any changes in levels of dependent variables (e.g., episodes of diabetic ketoacidosis [DKA] and

glycosylated hemoglobin levels) coincident with systematic variation in the independent variable (e.g., the environmental setting of the subject). In this case, dramatic changes in episodes of DKA and glycosylated hemoglobin in home versus other controlled settings give a much stronger data base for conclusions than does the usual

study. The single-subject design replicates systematic variation of the dependent case

variables in response to different environmental settings.

Case The

326

Report patient was born and raised

in

a

small

years

in Florida. Her first spent with her mother,

town

were

father, older brother, and an older sister. When she was 2 years old, her parents separated, and one year later her father moved out of the state. After that time she saw her father once or twice a year. At the age of 7 years the patient developed insulin-dependent diabetes mellitus (IDDM). She entered school on time but failed the second and fourth grades. By then she had been hospitalized 13 times for diabetic ketoacidosis. The summer following her fourth-grade failure she was admitted to the Diabetes Project Unit (DPU) at the University of Florida Health Science Center. The move from home to the DPU represents the transition from the first phase of condition A to the first phase of condition B. The DPU is an intermediate-term (two to six months) residential treatment facility housed in the Department of Psychiatry’s Children’s Mental Health Unit. It is staffed by health-care professionals from the Departments of Psychiatry and Pediatrics. The patients are children and adolescents with major adjustment problems affecting their diabetes control, resulting in frequent hospitalizations and/or school failure. Psychological testing of our patient indicated borderline intellectual ability. Achievement testing found her several grades behind grade level in math and reading. However, when feeling well, she was capable of friendship with adults and peers, having good verbal skills and an outgoing personality. She was affectionate and had a sense of humor, traits which contributed to the attachment many adults at the Health Science Center felt toward her. However, she also displayed serious behavior problems. With adults she was fre-

Downloaded from cpj.sagepub.com at SIMON FRASER LIBRARY on June 12, 2015

quently oppositional, had tantrums, swore, and was physically aggressive. With peers, she was sometimes verbally vicious. Her ability to deal with anger was very poor. At times she acted out, and at other times she became with-

drawn, sometimes refusing to communicate for days. The patient’s first DPU admission lasted six months. Efforts to include the mother as an integral component of the rehabilitation program failed. The mother refused to participate in diabetes education and family therapy and repeatedly disappointed her daughter by not showing up when she agreed to visit the DPU or take the child for a home visit. On several occasions, it was quite apparent to the DPU staff that the mother arrived at the unit very intoxicated. In addition, when the child was on home visit, the mother did not comply with staff requests to keep behavioral records of the diabetes regimen or to supervise the diabetes regimen as prescribed by the treatment team. When these issues were explored in therapy, the mother frequently became belligerent. Consequently, the Department of Health and Rehabilitative Services (HRS), Florida’s child protection agency, was contacted and a report of medical neglect was filed. The DPU staff’s recommendation was to place the child outside the home. Extensive efforts were made to educate the protective service workers through repeated phone calls, consultations, and letters. However, the HRS caseworker assigned to the case did not concur with the medical and psychological staff’s recommendations. He reportedly advised the patient’s mother that the DPU staff was not acting in her child’s best interest and that the mother should remove the child from the DPU facil-

itv against medical advice, which the mother did. Grave

concern

this action was expressed by the DPU medical and psychological staff to the HRS. Through repeated efforts of the DPU staff, a over

guardian ad litem was appointed by the court. The function of the guardian ad litem was to act as a disinterested individual on behalf of the patient. This was necessary, since there was disagreement between the patient’s family, health-care providers, and the state agency personnel regarding the best care of this child. In this case, the guardian ad litem agreed with the DPU staff; he indicated that the child would be best served by placement outside the home. These recommendations were not followed. During the year following her discharge from the DPU, the patient lived at home and again experienced multiple episodes of diabetic ketoacidosis. The move from the DPU to home represents the return from the first phase of condition B to the baseline condition A. At age 12 she was readmitted to the DPU. At the time of this admission the mother gave up custody of her child. Although HRS agreed to place the youngster outside the home, it took approximately eight months before an appropriate placement was found. This eight-month period represents the second phase of condition B. The child was finally discharged to the foster care of her great-aunt. During the time she lived with her great-aunt the patient did well, with no episodes of ketoacidosis. The time at the greataunt’s home represents condition C. This placement was short-lived, however; the patient was returned to the mother’s care by HRS after lived at the having great-aunt’s house for only six months. The child’s final placement with the

Figure 1. HgbA1

at different

placements.

mother represents the last return to baseline condition A. This action was taken despite verbal and written opposition by both the unit director and medical director of DPU, the guardian ad litem, and the great-aunt. The great-aunt reported repeated instances of irresponsible behavior by the mother, such as extreme intoxication and lack of supervision. Seven months after the HRS decision to return this child to her natural mother, the patient was dead. The data shown in Figures 1 and 2 graphically demonstrate the signs of medical neglect evidenced when this child was in the care of her mother. The glycosylated hemoglobin (HgbA1) values during the course of the patient’s treatment are shown in Figure 1. Glycosylated hemoglobin levels provide an indication of the patient’s average blood-sugar level over the previous six to 12 weeks. A value greater than 12 % is indicative of moderate to severe hyperglycemia, while values between

8.6% and 12% were considered indicative of mild hyperglycemia. For the two years before admission to the DPU, the child’s HgbA1 values were 14.3% to 14.9% (during the first phase of condition A). At the time of discharge from the DPU, her HgbAi was 9.3% (reached during the first phase of condition B), indicating dramatic improvement in diabetes control. Her percent

HgbA1

rose

to

16.9%* only

two

months after her discharge from the DPU. Nine months later, while the patient remained in her mother’s care (second phase of condition A) , her diabetes control remained poor. During the course of her second DPU admission (second phase of condition B), she had improved metabolic control, with HgbAi values dropping from 13.9% one month after admission * Both HgbAi and HgbAiC values were obtained for this patient. For the sake of comparability, all HgbAlC values have been converted to total HgbAi values using the following formula: Total HgbAi 1.5xHgbAiC

Downloaded from cpj.sagepub.com at SIMON FRASER LIBRARY on June 12, 2015

=

327

12.4%

discharge. When the discharged to foster care in her great-aunt’s home, conto

patient

on

was

trol of her metabolism was maintained, with her HgbAi values ranging from 12.3% to 12.6% (condition C). In contrast, once the patient was returned home (first condition A) against medical advice, the patient’s diabetes control deteriorated severely. Five months after she had been returned to her mother, her HgbA1 was 31.0%, indicative of untreated

diabetes.

Figure 2 shows the patient’s record of episodes of DKA. From the time of the onset of disease until her first admission to the DPU, she had had 13 episodes (first phase of condition A), but none during the six-month period she first spent on the DPU (first phase of condition B). Upon return to her mother’s care, she had four documented episodes of DKA (second phase of condition B), but again no episodes during another eight-month period in the DPU, an effect sus-

tained while in her great-aunt’s home for six months (condition

C). Upon this patient’s return to her mother’s home (final phase of condition A), she resumed having frequent episodes of DKA, the last episode associated with mucormycosis, which eventually led to her death. The patient’s school performance paralleled her metabolic status. In the year preceding the patient’s first admission to the DPU, she missed 58 days of school. During her periods of residence on the DPU, her attendance at school was good, her grades ranged from Cs to As, and she was promoted. This stands in sharp contrast to her previous record of repeating two grades before her first admission to the DPU. When the patient was returned to her mother from her foster placement, she failed the sixth grade, her third school failure. During the month before her admission to the University of Florida Health Science Center for mucormycosis, she had been ad-

mitted to a local hospital three times for DKA. During the last of these admissions she complained of left facial pain. Her local physician diagnosed mucormycosis. She was transferred to the University of Florida Health Science Center, where exploratory surgery indicated extensive involvement of the infection in the patient’s facial tissue and bone. Because of the nature of the treatment

below, the tee was

options described hospital ethics commit-

consulted.

The only therapeutic option with any chance of cure was to treat with IV amphotericin B and surgically remove the left side of the patient’s face, including her eye and ear. Had this option been chosen, the patient would have had chronic severe pain, been markedly disfigured, and would have required full-time nursing-home care. In addition, she would probably have required several surgeries, with no guarantee of extended life. The family chose to treat with amphotericin B but to avoid surgical intervention. During the patient’s hospitalization at the Health Science Center, she and her mother were

provided

support. The

with

psychological patient regressed

emotionally and spent most of the time whining and crying. On day 16 of the hospitalization, a CT scan indicated that the fungal infection had spread to her brain. Five days later she died.

Discussion

Figure 2. Episodes of DKA at different placements.

328

Downloaded from cpj.sagepub.com at SIMON FRASER LIBRARY on June 12, 2015

It is our impression that a variety of factors contributed to the untimely death of our patient. We cite selected literature below to elucidate some we believe may have been involved. These include the child’s characteristics and their role during interaction with her mother, issues involved in parental

rights, and the needs and decisionmaking processes of child-protection workers. In an interesting conceptualization of etiological factors contributing to the maltreatment of children,’ the abuse-eliciting characteristics of children were discussed as one determinant or potential contribution to the phenomenon of child abuse. In this conceptualization, the author discusses the match between the tempermental characteristics of parent and child, that is, the characteristics of the child that may act as elicitors of maltreatment when considered in relation to the caregiver’s attributes. The point made was that while some children may play a role in their abuse, they do not cause it; in addition, the author did not rule out the distinct possibility that certain parental behaviors may foster negative child behavior, thereby eliciting parenting hostility. In our case study, it is plausible that the socially undesirable behaviors of this child may have contributed to her maltreatment. Parental rights are another issue that is discussed in the literature and that we believe has implications for this case study. One author’ has described the long tradition in the United States of custodial rights for biological parents, which has not appreciably diminished with newer legislation for the protection of abused and neglected children. In addition, itwas noted that implicit in many laws is the assumption that honoring the rights of parents will have no ill effect on their children. Another author’ maintains that the decision in the case of Santosky v Kramer further undermined the rights of abused and neglected children. In that case, the Supreme Court ruled that &dquo;clear and convincing&dquo; proof is required to terminate parental rights. This ruling overturned the

law of many states, which required the less stringent standard of &dquo;fair preponderance&dquo; of evidence to allow termination of parental rights. In our case study, we believe that the rights of the child, as judged by the DPU staff and the guardian ad litem, were not given serious enough consideration. Our data demonstrated that the patient re-

peatedly experienced recurrent DKA when placed in the home environment with her mother. While of the episodes of DKA could be connected with instances when the patient had no parental supervision or when she skipped her insulin injections, the treatment team believes insulin omission was the cause for the majority, if not all, of the episodes of DKA. In a study of 44 patients with recurrent DKA’ it was concluded that the proximate cause of recurrent DKA is omission of insulin, and its prevention requires the establishment of a support system to ensure adherence. While the DPU staff and the guardian ad litem were able

only some

to

recognize this,

the

child-protec-

tion system was not. The decision-making processes of child-protection workers are another important factor in protecting children from maltreatment. In a review of relevant literature,lo eight factors were identified to assess risk in child abuse and neglect investigations: 1) age of the child, 2) functioning of the child, 3) cooperation of the caretaker, 4) functioning of the caretaker, 5) intent of the perpetrator, 6) current access of the perpetrator to the child, 7) severity of the current incident, and 8) existence of previous incidents. Rather than the criteria being considered individually, they should be viewed as a constellation of factors the worker can use in assessing the potential risk for a child. Our patient was an emotionally disturbed adolescent. While in

the custody of the mother, the child had repeated episodes of preventable life-threatening medical crises. The needs of child-protection workers are an important consideration in their ability to protect children from maltreatment. In one study, II workers’ confidence in the quality of the service they provided was associated with their training as well as their familiarity with the professional literature. In addition, membership in multidisciplinary or professional associations was associated with worker job satisfaction and self-perceived professional skills. It is our contention that in cases involving medical neglect, childprotection workers need training and/or available consultation. When making decisions, workers must be able to separate parental rights from children’s medical needs and characteristics. We believe the ABABCA design clearly demonstrates that the parent’s right to have the child in her custody was associated with repeated life-threatening problems that were not evident in controlled settings. We believe that cases of neglect resulting in death such as the one described here are preventable through increased training for protective service workers. Our patient had repeatedly demonstrated her ability to thrive and grow in structured settings where she was supervised as well as nurtured. The inability of the child’s home environment to provide the care necessary for her health and well-being was also redemonstrated. The mother failed to participate in diabetes education and family therapy. The mother repeatedly missed appointments and disappointed her daughter by not keeping appointments to visit or take her daughter for home visits. Multiple

peatedly

Downloaded from cpj.sagepub.com at SIMON FRASER LIBRARY on June 12, 2015

329

indicated the mother was excessively intoxicated at inappropriate times and provided inadequate supervision for her daughter. Neglect can be subtle.’ There may be no overt signs, as in physical or sexual abuse. Neglect of chronically ill children can be fatal. At the time of the HRS decision to reunite this child with her mother, strong opposition was expressed by the medical and psychological staff of the DPU who knew the youngster and her mother well, as well as by the guardian ad litem. We believe the HRS workers followed guidelines based on the theory that children who are not being overtly physically or sexually abused belong with their parents. For this patient, the system designed to prosources

330

her failed. Social-service workwho devote their careers to the protection of children deserve a higher degree of training about the grave consequences of neglect, particularly for chronically ill children. tect

5.

ers

REFERENCES 1.

2.

JM , Becarra RM. Defining ChildAbuse. NewYork, NY: Free Press; 1979. Fox S, Dingwall R. An exploratory study of variations in social workers’ and health visitors’ definitions of child malGiovannoni

Br J Soc Work. 1985; 15:467-477. Misener LR. Toward a nursing definition of child maltreatment using seriousness vignettes. ANS. 1986;8:1-14. Kempe HC. Child abuse — the pediatrician’s role in child advocacy and preventive pediatrics. Am J Dis treatment.

3.

4.

Child.

1978;132:255-260.

Downloaded from cpj.sagepub.com at SIMON FRASER LIBRARY on June 12, 2015

Council on Scientific Affairs. AMA diagnostic and treatment guidelines concerning child abuse and neglect. JAMA.

1985;254.796-800. Belsky J. Child maltreatment: an ecological 1980;35:320-335. Psychol integration. .Am 7. Derdeyn AP. Child abuse and neglect: the rights of parents and the needs of their children. Am J Orthopsychiatry. 1977;47:377-387. 8. Besharov DJ. Santoskyv Kramer. an unfortunate misunderstanding. Child Abuse Negl. 1983;7:117-118. 9. Golden MP, Herrold AJ, Orr DP. An 6.

approach to prevention of recurrent diabetic ketoacidosis in the pediatric

. 1985;107:195-200. Pediatr J population. 10.

Meddin

BJ. The assessment of risk in child abuse and neglect investigations. Child Abuse Negl. 1985;9:57-62.

11.

Fryer GE, Poland JE, Bross DC, Krugman RD. The child protective service worker: a

profile

tion of

of needs, attitudes, and utilizaprofessional resources. Child

Abuse Negl. 1988;12:481-490.

The death of a child with diabetes from neglect. A case study.

This case report of a child with insulin-dependent diabetes mellitus (IDDM) describes a naturally occurring ABABCA design. The A condition represents ...
513KB Sizes 0 Downloads 0 Views