Clinical Characteristics, Outcomes, Disposition, and Acute Care of Children and Adolescents Treated for Acetaminophen Toxicity

This study received institutional review board approval from Mayo Clinic prior to any research activity. Data were collected by using the Rochester Epidemiology Project (REP), an ongoing collaboration of medical facilities in Olmsted County, Minnesota, that collects data from community members who have agreed to share medical records for research. (Additional details on the REP and Olmsted County population are available in an online supplement to this article.)

We identified all patients in the REP database ages 0–18 who were evaluated for excessive acetaminophen exposure between January 1, 2004, and December 31, 2010 (the period for which data were available). We extracted information from the electronic medical record; procedures have been described elsewhere (26). Data on overdose intentionality and psychiatric diagnoses were based on provider documentation from the ED evaluation. For patients who received psychiatric consultation, the description contained in the psychiatrist’s note was used to determine intentionality; for those not receiving a consultation, the ED physician’s note was the data source. When intentionality was documented as ambiguous or ambivalent, the overdose was classified as intentional.

The sample included 110 cases representing 110 unique patients (i.e., none had multiple overdose presentations) under age 19 (range 0–18). Eighty-one percent were female, and 19% were male. All presented to the county’s tertiary referral center. Among the 110 overdoses, 97 (88%) were intentional, and 13 (12%) were unintentional. Information on demographic characteristics, overdose characteristics, medical sequelae and management, comorbid psychiatric diagnoses, initial disposition from the ED, and psychiatric consultation and hospitalization is presented in Table 1 for the entire sample and for intentional- and unintentional-overdose groups. Table 2 contains detailed characteristics of the two patients listed for liver transplant.

Our study examined characteristics and acute care of pediatric patients who overdosed on acetaminophen, both accidentally and intentionally. To our knowledge, this is the first comprehensive report of clinical characteristics, short-term medical management and outcomes, psychiatric and substance-related comorbidities, initial disposition, and utilization of psychiatric hospital care among youths evaluated for acetaminophen toxicity. In our sample, intentional overdoses, especially among adolescents, accounted for most exposures presenting to clinical attention.

A 2019 REP study of a cohort of 207 adults ages 18 and older treated between 2004 and 2010 employed the same methodology, and the results demonstrated many similarities to, and some differences with, our findings (26). In that study, the most common psychiatric diagnosis among adults who overdosed on acetaminophen was depression, as it was in our study. Substance use was more prevalent among the adults, and alcohol use disorder was the second most common psychiatric condition. The proportion of adults with pain disorder was almost four times that of our pediatric sample. Adults who overdosed inadvertently while attempting to treat pain may have contributed to the lower proportion of intentional overdoses in the adult sample. Although NAC was administered extensively in our pediatric sample, most of whom had overdosed intentionally, a slightly higher proportion of adults received NAC, and almost four times as many adults as youths were transferred to the liver transplant unit. Although adults had low overall rates of liver transplant and death, no patients in our pediatric sample required transplant due to recovery, and no deaths were reported.

Greater incidence and severity of medical complications among adults may be explained by factors such as extended histories of alcohol use, chronic pain with misuse of acetaminophen-containing analgesics, and other medical comorbidities (26, 27). Accidental acetaminophen overdose among adults is associated with greater morbidity and mortality, compared with overdose with suicidal intent (27). By contrast, our study’s unintentional-overdose group, which was younger than the intentional-overdose group, had less severe medical sequelae, compared with patients who intentionally overdosed, consistent with the extant pediatric literature (9, 11, 13, 15, 25). Psychiatric consultation was requested for similar proportions of adults and youths, although children and adolescents had higher rates of psychiatric hospitalization than did adults.

Despite low rates of acute adverse medical outcomes in our pediatric sample, primary prevention strategies should be considered. In the United States, acetaminophen is readily available in large quantities from unrestricted retail sources. In much of Europe, acetaminophen is available only in blister packs; this simple measure significantly limits overdoses (28, 29) and is particularly relevant for impulsive ingestions often seen among adolescents (19). Limiting access via more restrictive packaging, checking identification, and limiting quantities (as has been implemented for some over-the-counter medications, such as pseudoephedrine), or requiring provider prescriptions, could substantially reduce cases of pediatric acetaminophen toxicity.

Unlike our unintentional-overdose group, a notable characteristic of those who intentionally overdosed was the frequent presence of psychiatric comorbidity. Alcohol and cannabis use also were common. More than half of intentional-overdose patients were depressed, and adjustment disorder was the second most common diagnosis. Almost a quarter of patients who overdosed intentionally had attempted suicide previously. The high rate of psychiatric comorbidity in our sample is consistent with studies worldwide; among adolescents who presented to EDs after overdosing, depression was present among 79% of South Korean youths (16) and 57% of Australian youths (19).

Some characteristics of our sample and aspects of disposition and care received may be specific to the geographic and institutional environment. Although our data source should capture all cases in the county, all patients were evaluated and treated at one tertiary medical center, where the ED primarily serves local patients from communities across eight counties in which there are few other facilities for emergency care. The proportion of intentional versus unintentional overdoses, characteristics of patients who overdosed, and medical and psychiatric management patterns may differ in other community practices and hospitals. Whereas inpatient psychiatric care is often preferred for adolescents after intentional overdoses, actual treatment practices and disposition patterns differ widely. In a large U.S. sample of adolescents medically hospitalized after suicide attempts, most were discharged home, with only 21% admitted to a psychiatric or rehabilitation facility (30). More recent studies of intentional and unintentional pediatric poisonings evaluated in EDs in Lebanon (25) and Singapore (14) also found that most patients were discharged home. Similarly, among youths who presented to Australian EDs after overdosing intentionally, 60% were discharged home, 19% were admitted for medical hospitalization, and 13% underwent psychiatric hospitalization (19). Among French adolescents under age 16 who attempted suicide (most by overdose), nearly 93% were hospitalized medically, but only 7.5% received inpatient psychiatric care (24). A mere 2% of patients under age 15 who were evaluated for deliberate self-harm (mostly overdoses) in a United Kingdom study were hospitalized, although nearly half were offered psychiatric aftercare (20). Only 25% of a South Korean adolescent sample received outpatient psychiatric care following an intentional overdose (16).

Determinants of disposition of suicidal adolescents include patient-related characteristics, sociodemographic and geographic factors, and hospital setting. Diagnosis (bipolar disorder and substance abuse and dependence), clinical acuity (prior attempts or hospitalizations, psychotropic medications, treatment nonadherence, and lower global functioning), and lack of services are associated with likelihood of inpatient psychiatric care following pediatric ED visits for suicidal ideation and behavior (31). Quantity of medication ingested and presence of psychiatric diagnosis on specialist evaluation also are associated with hospitalization (32). Psychiatric hospital admission is more likely for adopted youths (31) and less likely for female adolescents (33). Youths residing in rural areas (33) and outside the northeastern United States (30) also are less likely to undergo inpatient psychiatric care after suicide attempts. Children’s medical units are more likely than adult hospitals to transfer adolescent patients to psychiatric, rehabilitation, or chronic care facilities after attempts (30).

Furthermore, having a psychiatric inpatient unit within the same facility as the ED increases the likelihood of hospitalization of youths evaluated for mental health concerns (34). The presence of a dedicated pediatric psychiatry inpatient unit at one of the REP facilities, combined with a dearth of readily accessible outpatient programs across the large, semirural geographical area served by our ED, may have contributed to the high rate of inpatient care utilization in our sample, in contrast with findings for other rural populations (33). However, availability of inpatient psychiatric care for adolescents is severely limited in many locations (35). Even with a pediatric psychiatry inpatient unit on site, patients under age 18 who require psychiatric admission from the particular ED in our study are more likely to be transferred and are transferred greater distances, compared with adults in need of inpatient care (36). Transfer to another facility prolongs ED length of stay (36) and creates substantial burden for patients and families. Outpatient alternatives for youths who overdose may need to be used when access to inpatient beds is difficult.

Uniquely, nearly 90% of our sample received psychiatric consultation. Specialist child psychiatry clinicians can play important roles in determining overdose circumstances (particularly in intentional cases or cases with ambiguous intent), identifying psychiatric comorbidities, and recommending disposition after medical stabilization. Near-universal consultation in our sample was another possible reason for the high rate of psychiatric hospital admission (almost three-quarters of patients), in contrast with the low rates of hospitalization after intentional self-harm found in other studies (19, 20, 30). French adolescents who attempted suicide had high rates of psychiatric evaluation (93%) but low rates of hospitalization (7.5%) (24), suggesting that other factors may have influenced our high hospitalization rate. Of interest, in our sample, approximately one-third of unintentional-overdose patients received psychiatric consultation, and one was admitted for psychiatric hospitalization. This finding implies that medical providers suspected or identified psychiatric symptoms among youths presenting with ostensibly accidental overdoses—or perhaps requested consultation in many pediatric cases out of an abundance of caution.

Although in-person psychiatric consultation is not universally available, increasing use of telepsychiatry services allows pediatric mental health specialists to participate in assessment and treatment planning. Tele-consultations for psychiatric emergencies not only can improve access to specialized care, but they are also efficient from cost, clinical, and operational standpoints; additionally, providers and patients’ caregivers report high satisfaction (37, 38). Multidisciplinary teams with nurses and social workers trained in suicide-specific interventions may also improve care in settings where access to psychiatrists and psychologists or to tele-consultation is limited (39, 40).

Providers working with adolescents for whom hospitalization is not feasible should liaise with families and local outpatient resources to develop alternative treatment plans. This may involve clinic-based intensive outpatient programs, more frequent psychotherapy appointments, and timely follow-up care with established providers. Newer treatment models specifically designed for suicidal patients may help reduce risk and bridge gaps in care delivery. Family-based crisis intervention (FBCI), an emergency intervention for suicidal adolescents, is an alternative to typical ED care and inpatient hospitalization (41). FBCI utilizes modules to teach concrete tools and safety-planning skills, aiming to stabilize patients during the ED visit and permit safe discharge home. A randomized controlled trial of FBCI versus treatment as usual showed reduced hospitalization rates and higher family empowerment and satisfaction in the FBCI group (41). Other innovative interventions involve technology tools to deliver care to suicidal patients, such as a novel avatar system that aims to standardize and improve ED care and to reduce hospitalization (42), and computerized adaptive testing methods for risk assessment (43).

Our study had several methodological limitations. We classified overdose intentionality retrospectively as a dichotomous variable based on providers’ descriptions from the initial medical-psychiatric evaluation. Prospective work should employ more descriptive methods, including patient self-report measures, to assess nuances of patients’ motives and examine relationships to other characteristics and outcomes. Future studies also should use multiple coders and assess agreement when data are extracted from subjective descriptions in clinical records. Additionally, our data were limited to cases through 2010. Shifts in the prevalence of suicide attempts and completions among children and adolescents and in health care utilization for suicidality (44, 45) demand analysis of newer data, with comparison to our findings, to understand recent changes in medical and psychiatric care delivered to youths who overdose. Comparison of different time periods is particularly necessary, considering reductions in inpatient psychiatric beds and lengths of stay worldwide, as well as increased boarding of psychiatric patients in EDs and medical floors (46).

In addition, patients presenting with polypharmacy overdoses may have inadvertently been excluded from our study, although in our study’s ED, it was institutional standard of care to obtain serum acetaminophen levels whenever there was concern for medication ingestion. Furthermore, data on timing of presentation and NAC administration relative to ingestion were incomplete and unable to be analyzed. Clinical records from mental health–related pediatric ED visits are often inconsistent and incomplete (47). More complete and accurate data are needed to understand characteristics of patients with delays between overdose and evaluation and treatment to improve care for these youths at higher risk of hepatic injury and other adverse outcomes.

We did not conduct prospective, longitudinal follow-up assessments after hospital-based care. Many patients in our study received posthospitalization care in geographically distant facilities not participating in the REP, and we did not have access to records of longer-term services and outcomes. Recent expansion of the REP (48) to include patients in a larger geographical area and additional outpatient clinics will enable future work examining longitudinal care delivery after overdose. Examining long-term outcomes and medical and psychiatric service utilization will be necessary to understand the physical and psychiatric sequelae of pediatric acetaminophen toxicity. Considering the potentially serious consequences and frequent need for high-level medical intervention, youths who overdose on acetaminophen may have different long-term outcomes, compared with those who overdose on other, less toxic medications. One study in a predominantly adult sample found distinctions in demographic factors and clinical course between patients overdosing on acetaminophen versus other substances, including a lower likelihood of recurrent attempts among those who ingested acetaminophen (49); however, it is unknown whether youths demonstrate similar patterns. No patients in our pediatric sample had multiple presentations for acetaminophen overdose during the study period; it is conceivable that the experience of receiving medical care for acetaminophen toxicity served as a deterrent for future overdoses for some youths.

Conversely, the acuity of medical assessment and intervention could prove traumatic and worsen anxiety among some patients, potentially increasing the risk of subsequent attempts and other psychiatric sequelae. Prolonged or multiple hospitalizations affect academic performance and friendships, resulting in stress and school avoidance upon discharge (50, 51). Although few patients in our study required admission to a pediatric intensive care unit (PICU), there is a high risk of psychological sequelae, including posttraumatic stress symptoms, for both children and families following PICU admission (52). Patients with serious or persistent medical complications, PICU treatment, or prolonged stays may require special attention to mitigate psychological consequences, including frequent, ongoing assessment of trauma-related symptoms; prompt management of pain and delirium; treatment of comorbid psychiatric conditions; and referral to trauma-focused psychotherapy (53, 54). The high incidence of psychiatric comorbidity and substance use among young patients who overdose also is concerning for long-term prognosis, because these conditions are rarely addressed completely in acute care settings focused on crisis stabilization. This highlights the critical need for a more comprehensive continuum of community mental health services (55), particularly considering suboptimal treatment adherence or nonadherence and high rates of repeat suicide attempts in this population (24). Medical and psychiatric teams caring for youths who overdose should ensure that discharge planning includes arranging longer-term mental health and substance-related treatment.

Serious acute medical outcomes among youths who overdose on acetaminophen were rare in this sample, although the study did not examine long-term medical sequelae and other psychosocial consequences. Psychiatric comorbidities and substance use were common. Psychiatric consultation and inpatient medical and psychiatric hospitalization were typical after overdose. Youths with intentional and unintentional acetaminophen overdoses may have distinct demographic and clinical characteristics affecting treatment and disposition. Rising rates of suicides and attempts (44) and increased health care utilization for suicidality (34, 45) require greater understanding of psychosocial determinants of outcomes and service utilization among pediatric patients who overdose. Innovative prevention and treatment approaches are imperative for the many youths who overdose yet struggle to access mental health treatment.

The authors acknowledge the contributions of the late Nuria J. Thusius, M.D., a skilled clinician and cherished colleague, who was instrumental in developing the study design and collecting data for this project and whose work they are honored to advance.