Characteristics of Medically Hospitalized Pediatric Patients With Somatoform Diagnoses

DISCUSSION

Although medically unexplained physical symptoms are common among youth⁶ and are associated with the overuse of health care services,^2,5 there is a need for additional research on pediatric somatoform disorders with regard to diagnostic criteria and psychosocial characteristics of these children and adolescents.²³ Few studies exist that describe the characteristics of youth diagnosed with somatoform disorders seen in the acute setting of inpatient medical hospitals. Instead, the majority of studies of pediatric somatization or functional somatic symptoms have used samples from outpatient settings, such as tertiary care medical clinics,¹⁸ pediatric primary care centers,² or schools.^1,19

Therefore, this chart review is the first of its kind to describe common characteristics seen in an understudied population of pediatric patients whose somatic complaints and functional disability were severe enough to warrant a medical inpatient admission, necessitate consultation with a psychiatric consultation service, and result in a somatoform disorder diagnosis. The cross-sectional design of this study precludes determination of causality; however, results highlighted a constellation of characteristics that, when combined, may result in a perfect storm and contribute to the development or maintenance of pediatric somatoform disorders (Table 3).

The selection of characteristics investigated in the chart review was guided by existing literature, and many of the findings corroborated results from previous outpatient studies on pediatric somatization and functional somatic symptoms. Thirteen percent of youth seen by the PCS during the 2-year period were diagnosed with a somatoform disorder, which is consistent with rates reported in outpatient community samples.⁶

Also supported by previous literature, the patient most commonly diagnosed with a somatoform disorder was an adolescent girl²⁴ presenting with multiple physical complaints,⁷ primarily pain and neurologic symptoms,⁶ and who was described as a “good” child¹⁵ with an internalizing coping style and a comorbid anxiety or mood disorder (Table 3).^6,12,24,25 Kozlowska¹⁵ describes the process by which good, compliant, perfectionistic children obtain relief from pressures to succeed by assuming the sick role. By maintaining the sick role, such children are released from their exemplary functioning without the expression of open rebellion and can also avoid experiencing anxiety about facing failure.

Younger children in our sample tended to complain of abdominal symptoms more frequently than adolescents, again supporting findings from previous studies.^7,26 It is important to note that the presence of a physical illness does not exclude the possibility of somatization playing an important role in the child’s presentation.²⁷ The current study highlights that medical conditions and psychosomatic symptoms or diagnoses are not mutually exclusive: approximately two-thirds of patients were identified as having a medical condition, and the majority of these youth reported unexplained symptoms that overlapped with their medical diagnosis (Table 3).

Findings from the current study also add to the existing literature. Previous research points to somatization processes as most common among low-income, disadvantaged adults and youth.^28,29 In contrast the average median household income of our sample was $80 000, which is ∼$30 000 greater than the national average.²² In fact, 24% of the patients lived in communities where the median household income was >$100 000. This finding is notable in the context of recent research highlighting youth from high socioeconomic status communities as an at-risk group.³⁰ Furthermore, although previous research has identified trauma as strongly correlated with or a cause of somatization processes,³¹ less than one-third of the youth in the current study reported traumatic experiences, and only 1 in 10 received a posttraumatic stress disorder diagnosis. A recent study using the same sample demonstrated that, although rates of trauma in a sample of medically admitted pediatric and young adult patients with somatoform diagnoses matched national norms,³² patients with trauma histories had unique psychiatric and psychosocial profiles compared with those without trauma histories.³³ Our sample is also unique with regard to patients’ psychiatric profile predating the medical admission for somatic symptoms, with high numbers of preexisting psychiatric diagnoses, history of psychiatric treatment, and ongoing psychotropic medication treatment.

This study highlights the pitfalls of looking for a single psychological factor to explain the emergence and maintenance of medically unexplained symptoms. Instead, our findings demonstrate that multiple contributing aspects are common, and medical providers should consider a combination of several salient psychosocial and medical characteristics.⁹ For example, family and academic contexts were highlighted in our investigation as associated with pediatric somatoform disorders.^23,34 Approximately 50% of patients had parents or siblings with mental or medical illnesses and reported family losses and conflict. A large proportion of the sample also endorsed self- and parent-imposed pressures or expectations for success and high academic demands. Significant impairment in school functioning was demonstrated by school absences in half of the patients and dropping grades for a quarter of the sample. These characteristics, coupled with the fact that the great majority of the sample was described as having an anxious temperament³⁵ and tendency to internalize feelings and use avoidant and passive coping styles, may point to an underlying mechanism by which pediatric somatoform disorders emerge and are maintained.

Although the majority of the sample was referred to outpatient therapy, 26% of the sample needed a higher level of psychiatric care, including partial or day treatment programming or inpatient hospitalizations on specialized medical–psychiatric units. Although it was beyond the scope of our study to assess for follow-up outcomes, the current study demonstrated that despite the psychiatric treatment recommendations given at discharge, in 14% of the cases PCS was reconsulted during subsequent medical admissions within the 2-year study period. Future studies should assess postdischarge patient outcomes including how many patients adhere to mental health treatment recommendations and how many patients benefit from psychiatric treatment, given the rate of readmissions seen in our sample for persistent somatic complaints.

Our findings suggest that pediatric patients with somatoform disorders are diagnostically and psychosocially complex, posing significant challenges for medical and behavioral health care providers to manage in isolation. A multidisciplinary approach in which both the medical and psychological needs of these youth are conceptualized and treated is most appropriate.^9,13 Thus, alerting medical providers in pediatric hospitals and primary care to the characteristics seen among these youth (Table 3) may help prevent or reduce medical service utilization and emotional and functional costs of pediatric somatoform disorders. Additional studies are needed to assess potential paths for these patients from initial presentation in primary care offices to specialty clinics, medical hospitalizations, and, potentially, medical–psychiatric treatment units. The movement toward integrated care in which behavioral health services are colocated in primary care offices, for example, may create the context for a comprehensive collaborative multidisciplinary approach³⁶ and for evidence-based models for managing pediatric somatoform disorders.

This study has several limitations. Given the nature of retrospective chart review, data were not collected prospectively or in a standardized fashion with validated measures. Second, this study lacks a comparison sample and includes only youth referred to a tertiary urban medical hospital in New England, thus potentially limiting the generalizability of findings. An additional limitation is the small sample size; although it allowed a more comprehensive and thorough investigation of the patients included in the study, it might have been a limitation in detecting potential associations between the variables of interest. Although the use of DSM-IV-TR diagnostic criteria for somatoform disorders in this study allowed comparison with the current literature, the recent publication of the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition³⁷ has prompted notable changes in diagnostic terminology and categorization of the diagnoses within the Somatic Symptom and Related Disorders,³⁸ rendering a potential limitation of this study. Furthermore, the sample represents only patients referred to the PCS for psychiatric assessment; it is possible that there were other medically hospitalized pediatric patients who may have met criteria for somatoform disorder but were not referred. Finally, important information about outcomes for patients after discharge was not assessed; this includes the possibility that patients sought care at different institutions and received alternative diagnoses or treatment recommendations.