Medically Unnecessary Venipuncture in Autism Spectrum Disorder Behavior Assessments

Abstract

Objective:

Assess the clinical value of routine laboratory testing in the emergency department (ED) during behavioral disturbance evaluations in children with autism spectrum disorder (ASD).

Methods:

A retrospective chart review of patients ages 3 to 21 years with ASD presenting to 3 pediatric EDs with behavioral disturbance from January 2019 to January 2020. Local laboratory standards were used to determine abnormal ranges in ED screening labs. Patients with abnormal findings were reviewed for medical significance, defined as the need for a medical intervention, inpatient observation, or the inclusion of a nonbehavioral diagnostic code due to an abnormal laboratory test result.

Results:

A total of 209 eligible ED encounters were reviewed. Mean age was 14.5 ± 3.1 years, and 84% were male. Of those, 84% (176/209) received venipuncture for screening labs per protocol, of which 97% (170/176) featured abnormal test results. Only 2 abnormal labs (1%) revealed clinically significant findings. Compared with whites, more patients of non-White race received venipuncture (90% vs. 73%, P = 0.001) but less non-whites receiving venipuncture were admitted to psychiatric facilities (44% vs. 62%, P = 0.01).

Conclusion:

This study demonstrates that routine screening labs in asymptomatic children with ASD presenting to the ED with behavioral disturbances are often outside the range of normal, but without clinical relevance. This practice may lead to unnecessary and painful venipuncture. Children with ASD are a uniquely vulnerable population for whom we should choose wisely.

In recent prevalence studies, an estimated 1 in 31 children is diagnosed with autism spectrum disorder (ASD).¹ Those with ASD often behave, communicate, interact, and learn in ways that are different from most other children their age.² Often when these differences are encountered in the medical setting, there is a need for resources that are tailored to meet the psychosocial and sensory needs of this population.³ In the acute care setting of the emergency department (ED), it is often difficult to exclude patients with ASD from departmental behavioral mental health (BMH) protocols that ensure department standards and flow. As a result, department protocols and care plans are rarely tailored to the individual needs of children on the autism spectrum.^4,5

A child with ASD is more likely to visit an ED for a psychiatric complaint than children without ASD⁶ and are also more likely to have longer ED stays due to placement challenges and delays compared with children without autism.^7–9 Children presenting to the ED in behavioral crises often undergo a set of routine screening laboratory tests (labs) upon their arrival,^10,11 often through nurse-initiated protocols. Obtaining these routine screening labs presents a potentially adverse experience for patients with ASD who may not be able to communicate or express themselves throughout the process of often painful specimen collection. Patients with ASD can be physically and chemically restrained during their ED stay, which often occurs concurrently with specimen collection, to prevent any potential harm to the patient or medical staff.^12,13

Numerous studies confirm that most children with acute mental health issues do not have underlying medical etiologies for these symptoms.^10,11 A growing body of evidence in both adults and children has shown that routine laboratory testing without clinical indication is unnecessary and adds to health care costs.¹⁴ The American Academy of Pediatrics (AAP) supports the Choosing Wisely Campaign, an initiative of the American Board of Internal Medicine (ABIM), that aims to help clinicians avoid unnecessary diagnostic tests for pediatric patients in EDs. The list, “Five Things Physicians and Patients Should Question” was developed by a task force comprised of members of the AAP Section on Emergency Medicine’s Committee on Quality Transformation (COQT) subcommittee and colleagues in Canada.¹⁵ Relevant to this study is one of the recommendations: “Do not obtain screening laboratory tests in the medical clearance process of pediatric patients who require inpatient psychiatric admission unless clinically indicated”. On the basis of the Choosing Wisely campaign, any diagnostic testing should be based on a thorough history and physical examination, and universal requirements for routine testing should be abandoned.¹⁵ Despite these sound recommendations, many institutions and psychiatric inpatient units have maintained policies mandating universal lab screening for medical clearance before transfer from EDs.¹⁶

The relevance of laboratory specimens obtained as a result of intake protocols for BMH patients and their use in medical decision-making has not been previously examined specifically among children with ASD.¹⁰ We hypothesized that < 5% of screening lab tests would reveal a medically significant finding that requires a change in medical management, based upon previous publications for children with BMH complaints.^10,11 Our work aims to explore the medical relevance of these laboratory tests obtained by departmental or facility protocols, while considering the ethical implications of mandating blood draws that impose often adverse experiences on patients presenting with BMH complaints.

METHODS

A retrospective chart review was conducted examining patient encounters at 3 urban pediatric EDs associated with a dedicated comprehensive autism center. The electronic health record (EHR) from all patients between the ages of 3 and 21 years with a known autism diagnosis presenting to the ED from January 2019 to January 2020 were secondarily screened for a BMH complaint (ie, “behavioral complaint” in the EHR), and those eligible were identified for manual chart review. The age cutoff of 3 years was utilized to ensure the reliability of the child’s ASD diagnosis, consistent with other ASD-related epidemiological studies.¹⁷ The EHR was reviewed for documented ASD-related care and the identification of ASD in the patient’s medical history. Those patients without a documented ASD diagnosis in the EHR or those whose chief complaint did not include a behavioral concern were excluded.

A review of the results from screening labs obtained at the beginning of the patient’s ED encounter, including but not limited to a complete blood count (CBC), a comprehensive metabolic panel (CMP) or basic metabolic panel (BMP), and a thyroid panel (optional) was performed. Per the behavioral health screening protocol at our institution, a CBC and metabolic panel (CMP or BMP) were ordered for all patients presenting with a behavioral complaint. Utilizing the local laboratory standards at each facility, abnormal test results were identified. These abnormal test results were then reviewed alongside the EHR by a single reviewer to determine whether a subsequent care decision was made because of the laboratory finding. Pertinent care decisions could be defined as medical management by medication, procedure, the consultation of a subspecialty, the admission of the patient from the ED to inpatient care, and the inclusion of a nonbehavioral diagnostic code due to an abnormal laboratory test result. Any potentially clinically relevant laboratory abnormalities resulting in an intervention as defined above was reviewed by a secondary reviewer. The EHRs from all patients meeting potential exclusion criteria were also reviewed by the same secondary reviewer.

Patient demographics were extracted from the electronic health record. Mean ± SD, a χ² test of independence to compare categorical variables, and %, were analyzed when appropriate among variables such as age, race, sex, and disposition. The prevalence of abnormal laboratory test results and medically significant test results among the sampled population was also determined. A P-value ≤ 0.05 was considered statistically significant.

RESULTS

A total of 213 unique patients with ASD and ED encounters were identified during the time period reviewed, of whom 4 were excluded for no BMH complaint (Fig. 1). Of the eligible 209 subjects, mean age was 14.3 ± 5.8 years, and 84% were male (Table 1). Specific BMH chief complaints and ICD-10 codes are summarized in Table 2. Overall, 176 children (84%) received venipuncture for behavioral complaint-related specimen collection per the ED’s BMH intake protocol. Of whom, 97% (170 children) contained at least one abnormal laboratory result as defined by the facility’s local laboratory standards. Nearly all (99%) abnormal lab results were of no clinical relevance (Table 3). Results out of the normal range were found in 83% of CBCs and 85% of CMP/BMPs performed, and half of the thyroid panels (Table 3). Two of these encounters (1%) featured an abnormal laboratory result associated with a medically significant finding or escalation of care as outlined in our criteria. These patients will be detailed below as Patient A and Patient B.

FIGURE 1.

Review of ED Encounters. Summary of ED visits evaluating the medical necessity of laboratory blood draws. The figure summarizes all patient visits during the study period (n = 213), those excluded (n = 4). It then further stratifies those who received blood draw (n = 176), those with an abnormal result (n = 170), and finally those with a clinically significant finding (n = 2).

TABLE 1.

Participant Characteristics

		Disposition			Received Blood Draw Per Protocol
Variable	Eligible Analyzed Cohort (n = 209)	Home (n = 105)	Facility (n = 104)	P	Yes (n = 176)	No (n = 33)	P
Age (y)
M (SD)	14.3 (5.8)	14.6 (3.3)	14.3 (2.8)	0.44	14.5 (3.0)	14.4 (3.1)	0.64
Sex*				0.44			0.85
Male	175 (83.7)	90 (85.7)	85 (81.7)		147 (84)	28 (16)
Female	34 (16.3)	15 (14.3)	19 (18.3)		29 (85.3)	5 (14.7)
Race†				0.01			0.001
White/Caucasian	69 (33)	43 (62.3)	26 (37.7)		50 (72.5)	19 (27.5)
Non-White‡	140 (67)	62 (44.3)	78 (55.7)		126 (90)	14 (10)
Disposition							< 0.001
Home	105 (50.2)				78 (74.3)	27 (25.7)
Facilityδ	104 (49.8)				98 (94.2)	6 (5.8)

Results are n (%) unless otherwise specified.

^*Assigned at birth (male/female) per the medical record.

^†Race as self-reported in the medical record.

^‡American Indian/Alaskan Native, Asian, Black or African American, Native Hawaiian/Pacific Islander, or Multiracial

^δIn-state or out-of-state facility, no admission for psychiatric inpatient care in the health system.

TABLE 2.

Behavioral Mental Health ICD-10 Codes and Diagnosis Summary for Children with Autism Spectrum Disorder

ICD-10 Code and Summary	No. Patients, N (%)
R 46.89	98 (47.0)
Other symptoms and signs involving appearance and behavior
R 45.851	33 (15.8)
Suicidal ideation
R 45.1	7 (3.3)
Restlessness and agitation
R 45.6	4 (1.9)
Violent behavior
R 45.851	3 (1.4)
Homicidal ideation
Others (Frequency ≤ 2)	64 (30.6)

TABLE 3.

Abnormal Laboratory Result Summary

Laboratory Test	No. Tests - N	No. Abnormal Tests - N (%)	Medically Significant Finding – N (%)
CBC	176	146 (83)	1 (< 1)
CMP/BMP/Chem14/Chem8	176	150 (85.2)	1 (< 1)
Thyroid Panel (TSH, T4 Free)	4	2 (50)	0

Patient A presented with asymptomatic anemia with a hemoglobin of 7.1 g/dL on a CBC. The patient was not clinically symptomatic during the ED visit and had a previous history of microcytic anemia up to 3 years before the reviewed encounter. Oral iron supplementation was prescribed to the patient upon discharge.

Patient B presented with the incidental finding of hypokalemia on a CMP. The patient had a low potassium of 2.9 mmol/L from an unknown etiology and was not clinically symptomatic during the ED visit. One dose of oral potassium was given in the ED to address hypokalemia. A subsequent redraw of the potassium level revealed a normal potassium level of 3.9 mmol/L and the patient was deemed medically clear for transfer to inpatient psychiatric care. Due to the rapid improvement in the potassium level after oral supplementation (40 mEq), it is unclear if the hypokalemia was accurate or due to a spurious lab error.

Racial Disparity

Half of this ASD cohort were ultimately transferred to an inpatient psychiatric facility for further evaluation and treatment. There were no age- or sex-related differences in patients dispositioned home versus those admitted and no differences were noted in patients who underwent venipuncture compared with those without labs ordered. The majority of children transferred to a psychiatric facility underwent venipuncture (94% vs. 74%, P < 0.001, admission vs. home disposition, respectively). More white children were dispositioned home than non-white children (P = 0.01), and a greater percentage of non-white children had protocoled BMH screening labs obtained compared with white children (P = 0.001, Table 1). When disposition is considered, a difference in venipuncture by race was lost in the group admitted to a facility, but a race-based disparity remains in patients discharged home, with more non-white children undergoing venipuncture compared with white/Caucasian children, respectively (84% vs. 61%, P = 0.007, Table 4).

TABLE 4.

Disposition Cohort Analysis by Race

	Disposition Home (n = 105)			Disposition Facility (n = 104)
	Received Blood Draw Per Protocol			Received Blood Draw Per Protocol
Variable	Yes 78 (74.3)	No 27 (25.7)	P	Yes 98 (94.2)	No 6 (5.8)	P
Racea			0.007			0.63
White/Caucasian	26 (60.5)	17 (39.5)		24 (92.3)	2 (7.7)
Non-Whiteb	52 (83.9)	10 (16.1)		74 (94.9)	4 (5.1)

Results are n (%) unless otherwise specified.

^aRace as self-reported in the medical record.

^bAmerican Indian/Alaskan Native, Asian, Black or African American, Native Hawaiian/Pacific Islander, or Multiracial.

DISCUSSION

The AAP joined the Choosing Wisely campaign in 2012 to help identify “Things Providers and Patients should Question” to reduce low-value care. Specifically, the AAP Section on Emergency Medicine (SOEM) Committee on Quality Transformation (COQT) assembled a task force to oversee the creation of a Pediatric Emergency Medicine Choosing Wisely list. On the basis of over a decade of data to support the medically unnecessary practice of routine laboratory testing of pediatric patients with BMH needs, their second of 5 recent recommendations is: “Do not obtain screening laboratory tests in the medical clearance process of pediatric patients who require inpatient psychiatric admission unless clinically indicated.”¹⁵ There is a need for further exploration of the impact of the Choosing Wisely campaign on this practice since its most recent recommendation.

Our data reveals that these outdated institutional policies persist locally, often mandated by inpatient psychiatric facilities before a child is deemed “medically clear” for acceptance, independent of physician assessment and medical decision-making. While this study was conducted before the publication of the Choosing Wisely recommendations,¹⁵ concerns for this medically unnecessary practice is well document in the literature^6,10,11 over 5 years before our study and ultimately prompted the subsequent recommendations that target medical clearance screening laboratory testing. Impact of the new recommendations on this practice remains to be seen; however, mandatory laboratory testing for medical clearance continues for children requiring inpatient BMH services at our center. Ultimately, outside facilities are dictating clinical care and prohibiting ED physicians from choosing wisely when it comes to ordering medically unnecessary blood tests.

Even more concerning is the application of these medically unnecessary policies to a particularly vulnerable group of children with ASD, where 84% underwent venipuncture for routine lab assessment, including 74% of children who were ultimately discharged home. Furthermore, nearly all of the laboratory test results reviewed (97%) were out of the range of the local laboratory standard. However, in 99% of cases, these abnormal findings were not deemed clinically relevant by our criteria. Abnormal or spurious lab results often require retesting,¹⁸ adding to patient pain, health care costs, and time to medical clearance. In addition, the incidental findings outlined in both Patient A and Patient B were not related to the behavioral disturbances that prompted the ED visit. Also, given the rapid resolution of the case of hypokalemia, a spurious lab error is likely that led to a medically unnecessary intervention to address the lab result, and additional painful venipuncture for the child, while delaying medical clearance.

Routine laboratory screening is a medically unnecessary practice that is not supported by the evidence, nor is it free of harm,¹⁴ particularly for children with ASD. While this ethos applies to all pediatric patients with psychiatric complaints, the ED presents an environment in which the unique needs of vulnerable patient populations may be overlooked or unmet and thus require specific attention in this regard.¹⁹ Patients with ASD necessitate care delivery that takes into consideration the multifaceted nature of the patient’s autism diagnosis through the development of coping plans and the integration of crisis prevention programs.^20,21 In our center, the current utilization of department-based nurse-initiated protocols in the ED setting is implemented before a patient is even seen by a physician or advanced practice provider. It is imperative that treatment teams exercise caution when including vulnerable patient populations in general patient categories in the emergency care setting (ie, behavioral complaints with and without an ASD diagnosis). While there is more variability in utilization of the nurse-initiated BMH protocol after the publication of Choosing Wisely, many receiving BHM facilities in our area still require medical screening laboratory testing results before accepting patients for inpatient admission. Moreover, recent literature shows that almost half of all ED encounter for BMH concerns involved medical clearance testing and that these practices varied across centers.²²

In our experience, it was evident that protocol-initiated laboratory blood draws being performed in children with autism presenting with behavioral complaints were medically unnecessary, and potentially unethical, which was the onus for this study. While we hypothesized that < 5% of these tests would be abnormal, the high number of abnormal test results outside of range was unexpected. However, only 1% of these were potentially clinically relevant, which is an even smaller number than the 5% to 6% previously reported in children with BMH complaints, where autism was not a specified diagnosis.^10,11 We recommend that treatment teams prioritize evaluating patients presenting with a behavioral complaint and a history of ASD with a thorough history and physical exam before any invasive procedures for specimen collection. Obtaining specimens when not clinically indicated has the potential to present adverse experiences for patients with autism and their families in the acute care setting. While the AAP has acknowledged that routine specimen collection and testing of BMH patients lack utility, it remains a part of nurse-initiated protocols throughout pediatric EDs despite recommendations to abandon this practice for over a decade.^23,24 Avoiding unnecessary specimen collections may reduce the need for chemical or physical restraint and overall adverse experiences for patients with ASD during the ED encounter.

Our data also revealed potential racial disparities in care for children with ASD. Reasons behind a greater discharge home of white/Caucasian versus non-white children are unknown but could reflect access to resources, outpatient follow-up, or differences in insurance not evaluated in this study. A discrepancy in venipuncture by race in children with ASD discharged home was unexpected since these BMH protocols apply to all children. Further investigation is warranted to explore reasons for such racial disparities and to ensure optimal treatment of all ASD patients.

Our study has several limitations. As a single pediatric health care system with a dedicated autism center, our observations may not translate to other hospitals and institutions. In addition, policies in Georgia inpatient facilities may differ across state lines. However, mandatory medical clearance screening policies are not state mandates, but rather policies established among the limited inpatient psychiatric facilities in the region, impacting all acute care settings serving children with BMH needs.

Furthermore, there are limitations of a retrospective chart review design, and the data analyzed was isolated to only one year of indexed visits to the system. It is also possible that the electronic health record review did not include all patients with ASD seen in the ED with a behavioral complaint. In addition, it cannot be determined from chart review as to why a treating physician or the nursing care team chose not to obtain behavioral screening laboratory draws per protocol. Often, however, these patients presented with additional chief complaints that were addressed during the visit that would not require an inpatient psychiatric admission.

CONCLUSION

Laboratory values obtained through routine screenings were often out of the normal ranges, but were rarely clinically significant, leading to further investigation by the medical team to secure medical placement. Pediatric emergency centers need to re-evaluate routine screening protocols for asymptomatic patients with ASD, as this practice is not supported by evidence, is not free from harm, and is not medically necessary. Ongoing institutional policies mandating clinically unnecessary lab screening for medical clearance in children with ASD, alongside all children with a BMH need, require national attention in order to effect change in policies and practice.