Family navigation to increase evaluation for autism spectrum disorder in toddlers: Screening and Linkage to Services for Autism pragmatic randomized trial

Abstract

The Screening and Linkage to Services for Autism pragmatic trial examined the effect of family navigation on referral and evaluation for autism spectrum disorder in an urban healthcare system serving low-income families. Children aged 16–30 months with initial scores ⩾3 on the Modified Checklist for Autism in Toddlers–Revised with Follow-up (M-CHAT-R/F) were randomized to usual care (n = 133) or navigation if indicated (n = 142). Bilingual, bicultural autism family navigators offered navigation to intervention-group families if the completed M-CHAT-R/F indicated the need for autism spectrum disorder diagnostic evaluation. Autism family navigators also assisted families to arrange for completion of the M-CHAT-R/F interview if not done, and then offered navigation if indicated. Of 49 families for whom navigation was indicated, 45% received navigation, including care coordination, education, and practical support. In intention-to-treat analyses, the intervention did not increase referrals for evaluation but more than doubled completed autism spectrum disorder diagnostic evaluations. Intervention children were also more likely to complete the M-CHAT-R/F interview with fidelity. Research is needed to identify effective methods to improve uptake of family navigation. Family navigation may be an important tool for reducing systemic inequalities in the early identification and treatment of young children with autism spectrum disorder (ClinicalTrials.gov, https://clinicaltrials.gov/; NCT02374541).

Lay abstract

Young children with autism spectrum disorder can benefit from early detection and intervention to moderate autism spectrum disorder’s effect on their development. Routine screening with standardized instruments facilitates early identification. However, many screen-positive children do not receive referrals for evaluation, complete diagnostic evaluations, or engage in services. This is especially true for children of color and those from less educated or low-income families. The Screening and Linkage to Services for Autism trial evaluated the effect of family navigation on referral and evaluation for autism spectrum disorder in screen-positive toddlers. The trial was conducted in an urban, low-income population. It was designed to test “real-world” implementation of family navigation. We randomly assigned 275 children aged 16–30 months with an initial positive autism spectrum disorder screen to receive navigation if indicated or usual care. Navigation increased the number of at-risk toddlers who were evaluated for autism spectrum disorder. It also improved the quality of autism spectrum disorder screening in primary care clinics. However, many children failed to receive indicated referrals for evaluation. Navigation did not improve referral rates. Furthermore, nearly half of families who were potentially eligible for navigation could not be contacted. Other ways of structuring the navigation program may be necessary to improve families’ access to and acceptance of navigation, especially in the setting of financial, cultural, and linguistic barriers. Our trial nevertheless showed that family navigation may be an important tool for reducing systemic inequalities in the early identification and treatment of young children with autism spectrum disorder.

Keywords

autism spectrum disorders diagnosis early intervention education services patient navigation toddlers underserved populations

Introduction

The importance of identifying autism spectrum disorder (ASD) in young children is increasingly recognized because of its high prevalence (Baio et al., 2018) and evidence that young children with ASD can benefit from early detection and early intervention (EI) to moderate ASD’s effects on development (French & Kennedy, 2018; Warren et al., 2011; Zwaigenbaum, Bauman, Choueiri, et al., 2015). Routine screening with standardized instruments facilitates early identification and is recommended in national guidelines (American Academy of Pediatrics, 2014; Johnson et al., 2007; Zwaigenbaum, Bauman, Fein, et al., 2015). However, many screen-positive children fail to receive referrals for evaluation, complete diagnostic evaluations, or engage in services (Chlebowski et al., 2013; Pierce et al., 2011; Windham et al., 2014).

Children of color and those from less educated or lower-income families are referred for evaluation, diagnosed with ASD, and provided services later and less consistently (Fountain et al., 2011; Irvin et al., 2012; Liptak et al., 2008; Mandell et al., 2005; Thomas et al., 2007; Zuckerman, Mattox, et al., 2014). Their families report having less information about and more problems obtaining specialty care, fewer support services to connect to care, and less satisfaction with EI services; additional barriers include community stigma and differing beliefs regarding mental health and disability, provider dismissal of parental concerns, and parental distrust of providers (Liptak et al., 2008; Thomas et al., 2007; Zeleke et al., 2019; Zuckerman et al., 2015, 2017; Zuckerman, Mattox, et al., 2014; Zuckerman, Sinche, Cobian, et al., 2014; Zuckerman, Sinche, Mejia, et al., 2014; Zwaigenbaum, Bauman, Stone, et al., 2015).

One potential strategy to address these barriers is patient navigation (Freeman & Rodriguez, 2011), in which a trained navigator guides patients through and around barriers to ensure timely diagnosis and treatment (Freeman, 2004). Family navigation (FN), an adaptation of patient navigation that incorporates a family systems approach, improves preventive care delivery in low-income children (Hambidge et al., 2009; Szilagyi et al., 2011) and may improve management of chronic conditions in children (Raphael et al., 2013).

Several studies have examined FN for young children with or at risk of developmental disorders. Of 53 poor, urban children <34 months old referred to EI services for failing a developmental or ASD screening test, not meeting developmental milestones, or parental concerns who received navigation, 79% completed an evaluation for EI eligibility compared with 51% of historical controls (Guevara et al., 2016). In a randomized trial of paraprofessional home visiting services (providing assistance consistent with the FN model) for children with or at risk of any developmental disability or inadequate parental care, children receiving these services were more likely to receive an individualized family service plan (IFSP) (Rosenberg et al., 2002). Crossman et al. (2020) examined 260 families of children with ASD in the Autism Treatment Network and found modest improvements in parent activation and internalizing caregiver strain after FN implementation. In a trial of 39 children diagnosed with ASD, half of whom were Medicaid-eligible, families randomized to receive navigation 1 week post-ASD diagnosis were significantly more successful in scheduling or completing appointments for recommended services compared with families randomized to navigation 3 months post-diagnosis (Roth et al., 2016). Lopez et al. (2019) randomized Latinx parents of children with ASD or at risk of ASD to receive an intervention delivered by promotoras that was consistent with some aspects of FN and found improved family outcomes (e.g. comfort asking for services and access to health care and transportation) compared with the usual care group. We identified only one randomized trial, a small pilot study, that has examined FN specifically to increase receipt of diagnostic evaluation among low-income children who screened positive for ASD. This study demonstrated that children ages 15–72 months referred for ASD evaluation who were randomized to receive navigation were more than 3 times as likely as controls to complete a diagnostic assessment within 1 year (Feinberg et al., 2016). There were also beneficial effects on perceived stress and social support. As all enrolled children had already been referred for diagnostic assessment, however, this trial could not assess the effect of FN on referral. Furthermore, 69% of potentially eligible children did not enroll in this efficacy trial; hence, findings may not be generalizable to the broader population.

Screening and Linkage to Services for Autism (SaLSA) aimed to test, under “real-world” conditions, the effect of autism FN on each of the steps required to obtain referral for and complete an ASD diagnostic evaluation and receive referral for EI services, among children aged 16–30 months who received healthcare in an urban, safety net healthcare system.

Method

Study design, setting, and sample

The SaLSA pragmatic (Loudon et al., 2015) parallel-group randomized controlled trial was planned and implemented in collaboration with El Grupo Vida, a community-based network of Hispanic/Latino people with autism and other disabilities and their families. Each week throughout the study period, an independent data analyst based at the study site applied a computer algorithm to electronic health records (EHR) that automatically identified children meeting eligibility criteria (described below) and randomly assigned them in equal numbers to the intervention or control arm using a computerized random number generator. It should be noted that although using the EHR to identify and invite eligible participants may be easily achievable in usual care depending on local resources, the additional effort for recruitment may have reduced the trial’s pragmatism (Loudon et al., 2015). Due to the nature of the intervention, neither participants nor service providers were blinded to study arm after allocation. The Colorado Multiple Institutional Review Board approved this study.

The trial was conducted at Denver Health (DH), a community health system with 27 primary care and school-based clinics serving mainly low-income patients. DH began routine screening for ASD in February 2015. Children are screened at 18- and 24-month well visits using the Modified Checklist for Autism in Toddlers–Revised with Follow-up (M-CHAT-R/F) (Robins, 2008; Robins et al., 2009, 2014), a two-step parent-report screener. In Step 1, parents complete the M-CHAT-R; scoring 3–7 indicates moderate risk and ⩾8 indicates high risk. For moderate-risk children, clinical staff clarify at-risk responses using the follow-up interview (Step 2) to reduce false positives. Follow-up scores ⩾2 are considered positive. An initial score ⩾8, or 3–7 with follow-up score ⩾2, indicates the need for referral for evaluation to determine the eligibility for EI and presence of ASD diagnosis (Robins et al., 2009, 2014).

Denver residents are referred to Rocky Mountain Human Services (RMHS), which is responsible for ensuring that, within 45 days after the family accepts referral, the child receives an EI eligibility evaluation under “Child Find” (a legal requirement for schools to find children who have disabilities and need services, under Part C of the Individuals with Disability Education Act (IDEA)), for children 0–2 years old) and, if eligible for services due to disability, an IFSP. If the referral to RMHS specifies a positive ASD screen or concern about possible ASD, or if requested by a parent or service provider of an EI-eligible child, RMHS provides a multidisciplinary team evaluation for ASD. The evaluation is performed by trained clinicians who use behavioral observation, parent report, and standardized testing, including the Autism Diagnostic Observation Schedule (ADOS) (Charman & Baird, 2002; Falkmer et al., 2013).

All Denver residents aged 16–30 months without a prior ASD diagnosis who were seen for well visits at a DH primary care clinic from February 2015 to February 2018, and had an initial M-CHAT-R score ⩾3 documented in the EHR by their provider, were eligible for the study. Based on periodic audits throughout the study period, the M-CHAT-R was administered to an estimated 76% of children eligible for the trial during the study period. Siblings of randomized patients were excluded. No other exclusions were applied.

Randomization was based on an initial score ⩾3 because research has shown that for many children whose score indicates moderate risk, the recommended follow-up interview is not administered (e.g. Guthrie et al., 2019). Therefore, one aspect of the planned intervention was for the autism family navigators (AFNs) to assist families of moderate-risk children to obtain the indicated follow-up interview, as required for test fidelity and accuracy (Robins et al., 2009, 2014). While the follow-up interview is primarily intended to reduce false positive tests, confirming an above-normal score might also motivate clinicians to refer the child for evaluation. Because randomization was based only on the initial score, it was expected that a substantial proportion of moderate-risk children randomized to the intervention group would not require referral for ASD evaluation based on a negative follow-up interview and would therefore not be offered navigation. Such children were considered to have received the allocated intervention.

The control group children received usual care. It was similarly expected that a proportion of moderate-risk children randomized to the control group would not require referral for ASD evaluation based on a negative follow-up interview.

Autism FN intervention

The AFNs aimed to assist families whose screen-positive children needed ASD evaluation to overcome individual and system barriers to obtaining referral, evaluation, and services (Wells et al., 2008). The two AFNs were bilingual, Latina research staff already employed by DH; although we did not require additional qualifications, one had personal experience with ASD and the other had general (not ASD-specific) clinical training. The AFNs completed standard navigator training, including interpersonal communication, problem-solving skills, motivational interviewing, and navigation (e.g. locating resources, strategies for working with healthcare teams, and community agencies), with modeling, practice, and feedback (https://patientnavigatortraining.org/courses/level1/). Additional training about ASD symptoms, screening, diagnosis, and treatment; EI processes and services; research procedures and human subjects protection was provided.

For each child randomized to the intervention group, an AFN reviewed the EHR within 1 week of randomization to confirm the M-CHAT-R score (and follow-up score if completed), noting missing follow-up score and any scoring or recording errors (Figure 1). If a referral to RMHS was indicated or made, the AFN attempted up to seven contacts with the family by telephone, mail, and/or home visit to confirm eligibility and offer navigation. Where the child scored 3–7 but had no follow-up score, she similarly attempted to contact the family to confirm eligibility, and then arranged for administration of the follow-up interview. If clinical staff subsequently completed and documented the follow-up interview and the score was ⩾2, the AFN then offered navigation. If the follow-up interview was never administered despite AFN efforts, but the child was nevertheless referred for EI evaluation (e.g. for language delay or other developmental concern) and was otherwise eligible for the trial, the family was also offered navigation. The AFN obtained verbal informed consent for navigation.

Figure 1.

Procedure for implementation of autism family navigation.

An AFN administered a standardized intake interview to consenting families, usually by telephone, asking open-ended questions with probes to identify financial, logistical, sociocultural, educational, language, or medical barriers to evaluation. Intake data were collected and managed using Research Electronic Data Capture (REDCap), a secure, web-based application hosted at the University of Colorado (Harris et al., 2009).

The AFNs used a flexible approach to navigation to address barriers identified by the families. Navigation activities potentially included providing information and education on ASD, referral, evaluation, and EI, care coordination, scheduling assistance, coaching, advocacy, psychosocial support (e.g. attending evaluations with the family), interpretation and translation, and resources for insurance, transportation, child care, or physical needs. Most AFN assistance was provided by telephone call (or sometimes text message) with the family or service providers. As requested by families, AFNs also occasionally met with families at home, attended evaluations or IFSP meetings, or mailed resource information. The AFN maintained contact and provided navigation as needed for up to 3 months after the IFSP meeting (longer if requested). Families were free to accept as much or as little assistance as desired. Typically, each family worked with one AFN throughout the process.

At a system level, AFNs worked directly with clinical and EI providers and staff to overcome barriers to high-fidelity screening, referral and linkage to services, through training, education, and facilitation of communication between the family and the providers and staff.

Outcome measures

The primary trial outcomes were completed evaluations to (1) determine Part C eligibility for EI services and (2) diagnose ASD. Secondary outcomes included referral to RMHS for evaluation, development of an IFSP (with service recommendations as needed), and time to and age of referral, eligibility determination, and IFSP development. We also assessed completion of the M-CHAT-R/F with fidelity, which was defined as follows: M-CHAT-R score ⩾8 or M-CHAT-R = 3–7 and a follow-up score recorded in the EHR. Diagnosis of ASD is also reported. Outcomes were assessed as of the date the child reached 36 months of age or the trial ended, whichever came first.

Data collection and management

All trial outcomes were assessed using an anonymized data set linking DH and RMHS electronic records. Variables obtained from clinical and administrative patient records collected by DH and maintained in the DH data warehouse included family language and child sex, race, ethnicity, insurance status, International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) or International Classification of Diseases, Tenth Revision (ICD-10) codes, age at initial M-CHAT-R screen, M-CHAT-R/F initial and follow-up scores, referral to RMHS, and age at referral. Insurance status was categorized as Medicaid (a public health insurance program for people with low income), Title XXI State Children’s Health Insurance Program/Denver Health Financial Assistance Program (CHP/DFAP) (public and local programs, respectively, for low-income children/families), and other (commercial insurance or self-pay). ICD-9-CM codes were used to identify complex chronic conditions, as defined by Feudtner et al. (2000). ICD-10 codes were converted to ICD-9-CM equivalents using the Centers for Disease Control and Prevention’s General Equivalency Mappings table (National Center for Health Statistics, 2019).

At RMHS, data on referred children are maintained in an electronic record system. As required in statute, RMHS transmits back to DH information about their referred patients, including assignment to care coordinator, eligibility determination and reason for eligibility, completion of IFSP, and recommendations for EI services. For research purposes, RMHS created a single data set for all children referred from DH for Part C evaluation that included these data plus receipt of diagnostic evaluation and ASD diagnosis (if any), which was transmitted to the DH data warehouse. RMHS did not have data on engagement in EI services for many participants so these outcomes could not be evaluated as originally planned.

An independent DH analyst linked the RMHS data to the DH electronic data on randomized study participants to create an analyzable data set. DH staff also manually reviewed completed M-CHAT-R/F forms (when available in the EHR) for all randomized patients and, when errors were identified, documented correct scores in the data set. Dates were converted to ages (in days) and other personal identifiers were removed before transmission to research staff.

Data analysis

Intention-to-treat analyses were performed using data from all randomized children, regardless of receipt of navigation, subsequent determination of ineligibility, refusal, and so on, to ensure comparability between intervention and control groups. Descriptive statistics were computed for participant characteristics, comparing groups with chi-square or t tests. To test between-group differences, we used log-binomial or linear regression or Cox proportional hazards models, as appropriate. For survival analyses, observations were censored at 12 months after initial screen or at study end date, whichever came first. RMHS records included dates of eligibility determination and IFSP completion only for the first referral the agency received from DH for each child, which may have occurred prior to the positive M-CHAT-R screen (e.g. due to a previously recognized developmental condition), sometimes resulting in negative time to event. We conducted additional survival analyses excluding such children. Hypothesis tests were two-sided with the Type 1 error rate set at 0.05. A sample size of 54 referred children in the control group and 84 referred children in the intervention group was estimated to provide greater than 80% power at alpha = 0.05 to detect a clinically important increase in receipt of ASD diagnostic evaluation. Statistical analyses were performed using R (R Foundation for Statistical Computing, Vienna, Austria).

Results

We randomized 275 children to intervention (n = 142, 52%) or control (n = 133, 48%) between February 2015 and February 2018 (the planned trial end date). Most randomized children were male, Hispanic ethnicity, White race, and insured by Medicaid (indicating low income); had no complex chronic conditions; lived in English-speaking households; and had moderate ASD risk. Study groups did not differ significantly on measured characteristics (Table 1). All 275 children were analyzed (Figure 2).

Table 1.

Characteristics of 275 children at risk of ASD randomized to family navigation if indicated or to usual care.

Characteristic	Usual care (n = 133)		Family navigation (n = 142)^a
Characteristic	n	%	n	%
Male sex	85	64	86	61
Hispanic ethnicity	79	59	88	62
Race
White	76	57	79	56
Black	23	17	29	20
Other	34	26	34	24
Insurance
Medicaid	124	93	131	92
CHP/DFAP	5	4	5	4
Commercial or self-pay	4	3	6	4
Complex chronic conditions
0	118	89	130	92
⩾1	15	11	12	8
Language
English	92	69	99	70
Spanish	35	26	32	23
Other/unknown	6	5	11	8
M-CHAT-R initial screen
High risk = 8+	12	9	12	8
Moderate risk = 3–7	111	83	116	82
Low risk = 0–2^b	10	8	14	10

ASD: autism spectrum disorder; CHP: Title XXI State Children’s Health Insurance Program; DFAP: Denver Health Financial Assistance Program; M-CHAT-R: Modified Checklist for Autism in Toddlers–Revised.

There were no statistically significant differences between groups (all p values > 0.45). ^bChildren with initial positive screen who were randomized, but after review of M-CHAT-R form and correction of scoring errors, determined to be low risk.

Figure 2.

CONSORT flow diagram for SaLSA trial.

Of 142 children randomized to the intervention group, 68 (47.9%) were not offered navigation because the child did not require referral based on their M-CHAT-R follow-up score, completed either at their initial visit or with AFN assistance. Three children (2.1%) were determined to be ineligible post-randomization. The need for referral or navigation could not be determined for 22 children (15.5%) because the provider never administered the indicated M-CHAT-R follow-up interview or made a referral. The remaining 49 children were eligible for family navigation. Of these, 22 (44.9%) consented to navigation, 6 (12.2%) refused (four because the child was already receiving EI services), and 21 (42.9%) could not be contacted (including four whose foster care caseworker refused guardian contact).

Families more often accepted navigation if they spoke Spanish (28.1%, compared with 11.8% of families speaking English or another language) and were Hispanic ethnicity (19.3% vs 9.3% of non-Hispanic families) or White race (22.8% vs 6.3% of non-White families). No other measured characteristics differed between intervention group families who did and did not accept navigation.

Including the initial intake call, the AFNs completed on average three phone calls with each family and interacted in person with 23% of families (e.g. attending evaluations). The AFNs most often provided care coordination (95%) and information about ASD screening, evaluation, and EI services (68%).

Compared with control children, intervention children were 25% more likely to complete the M-CHAT-R/F as recommended (Table 2). In the subgroup of 226 children for whom the follow-up interview was indicated based on M-CHAT-R scores of 3–7, 119 (53%) completed it; intervention children were 33% more likely to complete an indicated follow-up (60% vs 45%, RR = 1.33, 95% confidence interval (CI) = [1.03, 1.72]). Among moderate-risk children with a completed follow-up interview, similar proportions in both groups scored ⩾2 (38% vs 40%, RR = 0.94, 95% CI = [0.60, 1.49]). Intervention and control children did not differ significantly in their receipt of referral to RMHS (45% in each group), Part C eligibility determination (25% vs 23%), IFSP (23% vs 22%), or referral for EI services (20% in each group) (Table 2).

Table 2.

Effect of family navigation on M-CHAT-R/F screening fidelity, referral, and evaluation for ASD and referral for EI services.

Outcome	Controln = 133n (%)	Interventionn = 142n (%)	Risk ratio (95% confidence interval)^a	p value
Completed two-step M-CHAT-R/F	72 (54%)	96 (68%)	1.25 [1.03, 1.52]	0.02
Referred for evaluation to determine eligibility for EI services	60 (45%)	64 (45%)	1.00 [0.77, 1.30]	0.99
Received evaluation to determine eligibility for EI services	31 (23%)	36 (25%)	1.09 [0.72, 1.65]	0.69
Received individualized family service plan	29 (22%)	32 (23%)	1.03 [0.66, 1.61]	0.88
Referred for EI services	26 (20%)	29 (20%)	1.04 [0.65, 1.68]	0.86
Received ASD diagnostic evaluation	5 (4%)	15 (11%)	2.81 [1.05, 7.52]	0.04
Received diagnosis of ASD	5 (4%)	12 (8%)	2.25 [0.81, 6.21]	0.12

Bold values indicate results that are statistically significant at p < 0.05.

M-CHAT-R/F: Modified Checklist for Autism in Toddlers–Revised with Follow-up; ASD: autism spectrum disorder; EI: early intervention.

All analyses based on intention to treat (i.e. as randomized).

Intervention group children were nearly 3 times as likely to undergo diagnostic evaluation for ASD (Table 2). When limited to the 124 intervention and control group children who were referred to RMHS, the estimated effect was the same (23% vs 8%; RR = 2.81 (1.09, 7.26), p = 0.03). Twelve intervention children (8%) and five control children (4%) were diagnosed with ASD.

Of the 22 children whose families consented to navigation, 100% were referred for EI service eligibility evaluation, 16 (73%) completed the eligibility evaluation, 15 (68%) were either evaluated for ASD (n = 10) or scheduled for ASD evaluation at trial close (n = 5), and 8 children (36%) were diagnosed with ASD (80% of the 10 evaluated), all of whom engaged in EI services. Excluding these 22 children, 4% of intervention group children underwent a diagnostic evaluation and 3% received a diagnosis of ASD, similar to the proportions in the control group (4% and 4%, respectively), indicating that benefits accrued only to intervention children engaged in navigation.

Time from initial M-CHAT-R screen to referral to RMHS, Part C eligibility determination, IFSP (Table 3), and age at which each occurred (Table 4) were similar between groups in intention-to-treat analysis. Exclusion of children with negative time-to-event (i.e. children who had been referred to RMHS, completed an eligibility determination, or received an IFSP before the initial M-CHAT-R, for example, due to chromosomal anomaly or other condition identified earlier in life) did not influence the magnitude or direction of the results.

Table 3.

Effect of family navigation on time from initial positive ASD screen to receipt of referral, evaluation for EI eligibility, and individualized family service plan.

Event	Hazard ratio (95% CI)^a	p value
Referral for evaluation to determine eligibility for EI services	1.00 [0.70, 1.43]	0.99
Completion of evaluation of eligibility for EI services	1.10 [0.68, 1.78]	0.69
Receipt of individualized family service plan	1.08 [0.65, 1.78]	0.77

ASD: autism spectrum disorder; EI: early intervention; CI: confidence interval.

All analyses based on intention to treat.

Table 4.

Effect of family navigation on mean age at receipt of referral, evaluation for EI eligibility and individualized family service plan, in children at risk of ASD.

Outcome	Control	Intervention	Mean difference in months (95% CI)^a	p value
Outcome	M age (months)	M age (months)	Mean difference in months (95% CI)^a	p value
Referred for evaluation to determine eligibility for EI services (n = 124)	20.1	21.0	0.9 [–1.3, 3.1]	0.43
Completed evaluation for EI service eligibility (n = 67)	20.5	21.6	1.0 [–2.5, 5.4]	0.61
Received individualized family service plan (n = 61)	26.1	23.5	−2.6 [–6.5, 1.2]	0.18

EI: early intervention; ASD: autism spectrum disorder; CI: confidence interval.

All analyses based on intention to treat, among those with the completed outcome.

Discussion

The most important effect of FN for young children who screened positive for ASD was an increased likelihood that they would receive a comprehensive diagnostic evaluation for ASD. Although only about half of families eligible for navigation engaged with the navigator in this pragmatic trial implemented in an urban, safety net system, children in the intervention group was nearly 3 times as likely as those in the control group to receive a diagnostic evaluation. Eighty percent of intervention group children evaluated for ASD were diagnosed with ASD, indicating that navigation did not increase unnecessary evaluations.

By liaising between the family and the provider and advocating for follow-up interview administration, the navigator significantly increased the likelihood of its administration. A theoretical advantage of completing the interview is avoiding unnecessary referral for ASD evaluation when the follow-up result is negative, potentially saving time and resources for families and the healthcare and EI systems. Nearly half of children for whom an M-CHAT follow-up interview was indicated did not receive it. Previous studies have similarly reported low rates of implementation of M-CHAT follow-up interviews for screen-positive children in pediatric primary care practices (Carbone et al., 2020; Guthrie et al., 2019), confirming that completion of the M-CHAT with fidelity is challenging in community settings. In such settings, relying on clinical staff to complete the follow-up interview as intended is unlikely to be effective without additional supports or resources. Use of digital technology has been shown to significantly improve the fidelity of M-CHAT-R/F implementation (Campbell et al., 2017). Based on our results, FN is another important resource for improving M-CHAT-R/F fidelity. Additional research on methods to improve the quality and completeness of ASD screening in busy primary care settings is needed.

FN had no effect on the likelihood of referral to EI, which is perhaps not surprising since the AFNs’ efforts were focused largely on the families, whereas clinicians were responsible for making referrals. Even with a positive ASD screen confirmed by follow-up interview, some providers decided against referral. Failure of clinicians to refer despite a positive ASD screen has been reported by others (Monteiro et al., 2019; Robins et al., 2019). Research is needed to better understand the reasons providers choose not to refer screen-positive patients and to develop and test interventions to increase provider referrals after positive ASD screens.

FN also did not significantly increase the likelihood of completing subsequent steps in the EI enrollment process, despite the provision of care coordination, resources, and other assistance by AFNs. Both study groups showed a similar fall-off after referral: only 56% of intervention children and 52% of control children referred to RMHS were evaluated for eligibility for EI services, and 50% and 48%, respectively, received an IFSP. While we anticipated that AFNs might influence whether families accepted the EI referral, the small number of families that received navigation may have reduced our ability to detect differences in these outcomes in intention-to-treat analyses. Among families that did receive navigation, the proportion that received an EI eligibility evaluation (73%) was substantially higher than in the overall sample. It is also possible that the disadvantaged families referred from DH need a more flexible timeframe to respond positively to EI case managers. The EI agency terminates the case if the family does not respond to two attempted telephone calls, followed by a letter, typically within 30 days of referral if no response is received. Once a case is terminated, a new referral is required to reopen it. Perhaps this fixed timeframe allowed the AFNs too little latitude to influence these outcomes. Finally, the processes at Part C programs designed to achieve these outcomes in a timely manner are well established and, in addition, the local EI and health systems had previously worked to improve communication and address barriers such as those related to transportation and interpretation. AFNs may have been unable to offer enough incremental benefit to existing practices to influence these outcomes. The AFNs focused most of their efforts on ensuring ASD evaluations were completed; such evaluations are not governed by Part C requirements and may therefore have been more susceptible to improvement with navigation. This may explain why the trial found a significant increase in ASD diagnostic evaluations with navigation, despite little effect on earlier steps in the EI process.

Feinberg et al. (2016) also conducted a randomized trial that examined FN targeting children who screened positive for ASD and had been referred for ASD evaluation. Our finding of a nearly threefold increase in the likelihood of receiving ASD evaluation with navigation is similar to the previous trial’s finding. Their trial also reported that twice as many children in the navigation group were diagnosed with ASD as in the control group, as did we, although differences were not statistically significant in either trial due to small numbers. Unlike Feinberg et al., our pragmatic trial examined the effectiveness of autism family navigation, delivered in a real-world setting using a flexible approach to navigation, starting from the initial positive screen and continuing through referral for ASD evaluation and diagnosis, yet found similar beneficial effects. Another trial, currently in progress, may offer additional insights on effects of autism FN (Broder-Fingert et al., 2018).

While most families who were successfully contacted consented to navigation, AFNs were unable to reach about 40% of families for whom navigation was indicated, despite repeated efforts using varied modes of contact. Low-income populations typically have greater residential mobility and up-to-date contact information was not always available. Furthermore, Latinx or Black families may have hesitated to respond to offers of assistance due to issues of language, culture, or trust. The existing sociopolitical climate and the fact that AFNs contacted families by telephone rather than in clinic may have led some families to disregard AFN calls, messages, and letters. If an AFN program were proposed for this population, program developers would need to consider and address these issues. Feasibility and acceptability of a navigation program within an urban community health system likely depends on enhanced visibility and understanding of the AFN role among clinicians, development and maintenance of relationships with primary care and EI staff and ensuring their understanding of the navigation program’s objectives and protocols, and community outreach and partnerships with community organizations to enhance trust. In our study, Hispanic White Spanish speakers were more likely to receive navigation than other families, suggesting that acceptance of navigation was supported by AFNs’ ethnic and linguistic similarity to the families. Navigated families were otherwise similar to families allocated to the intervention group that did not receive navigation, indicating that our results are broadly applicable to families seen in community health centers. Furthermore, our use of intention-to-treat analysis for primary outcomes avoided selection bias related to which families were successfully contacted and accepted navigation.

This study had several limitations. The M-CHAT R/F was first implemented in three DH clinics at the start of the study period and then implemented in other DH clinics. Despite clinic-level audit and feedback, multiple training sessions, as well as individual staff observation and feedback by the AFNs, failure to administer the screening test or record results, and errors in its administration, scoring and documentation, persisted. We had no data on otherwise eligible children who were not randomized due to such omissions or errors (accounting for about one fourth of eligible children). DH personnel conducting periodic audits reported anecdotally that these failures related primarily to provider characteristics (e.g. inexperience with the test) rather than child characteristics, suggesting against important selection bias. Nonetheless, incomplete screening did reduce study power to detect between-group differences. Study power was also reduced because fewer children presented for well visits at 18 or 24 months of age (when ASD screening was normally provided) than expected based on DH’s patient population. Children from lower-income or less educated households or who have no insurance or public insurance are more likely to miss well-child visits (Selden, 2006; Wolf et al., 2018), contributing to disparities in ASD screening and diagnosis. Families of children who fail to attend well visits may encounter even greater barriers to obtaining EI evaluation and services than children who do attend. Our study may therefore underestimate barriers faced by such families. Finally, unlike most US EI agencies, RMHS itself conducts diagnostic evaluations for children at risk of ASD. Results may differ in locales where families must obtain this evaluation through yet another system, potentially introducing additional barriers such as negotiating the new system or waiting longer for appointment times, or where ASD diagnostic evaluations are offered within the safety net system, potentially reducing barriers the families experience and therefore the impact of family navigation.

In conclusion, autism FN implemented in an urban, safety net system is feasible and effective for increasing the number of at-risk toddlers who are evaluated for ASD and appears to increase the number of children diagnosed with ASD. Having family navigators specifically dedicated to ASD enables development of expertise in ASD and EI as well as ongoing relationships with staff at EI agencies, which may improve care coordination. While AFNs were effective in assisting families whom they were able to reach and consent, in this pragmatic trial, many eligible families could not be contacted or declined to take part. Research is needed to test different ways of structuring the intervention (e.g. embedding AFNs within clinics) to improve access to and acceptance of navigation, especially in environments where language and immigration status may be important barriers. While recognizing the need to better understand how to engage low-income families and families of color in FN programs, our trial nevertheless demonstrates that FN has the potential to serve as an important tool for reducing significant, systemic racial, ethnic, and socioeconomic inequalities in the early identification and treatment of young children with ASD.

Supplemental Material

sj-pdf-1-aut-10.1177_1362361320974175 – Supplemental material for Family navigation to increase evaluation for autism spectrum disorder in toddlers: Screening and Linkage to Services for Autism pragmatic randomized trial

Supplemental material, sj-pdf-1-aut-10.1177_1362361320974175 for Family navigation to increase evaluation for autism spectrum disorder in toddlers: Screening and Linkage to Services for Autism pragmatic randomized trial by Carolyn DiGuiseppi, Steven A Rosenberg, Margaret A Tomcho, Kathryn Colborn, Kristina Hightshoe, Silvia Gutiérrez-Raghunath, Jeanette M Cordova, Jodi K Dooling-Litfin and Cordelia Robinson Rosenberg in Autism

Footnotes

Acknowledgements

We gratefully acknowledge contributions by Kristin Breslin, MPH; Simon Hambidge, MD, PhD; Sarah Leslie, MPH; Rosse Rodriguez-Perez, MD; Anita Roberts, RN; and Sarah Sabolot, BA (Denver Health and Hospital Authority) and Lindsay Krings, BS, and Beth Scully, BA (Rocky Mountain Human Services).

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This project was supported by HRSA (Health Resources & Services Administration)/MCHB (Maternal and Child Health Bureau) (Grant No. R40MC27702) and NIH (National Institutes of Health)/NCRR (National Center for Research Resources) Colorado CCTSI (Colorado Clinical and Translational Science Institute) (Grant No. UL1 RR025780). Its contents are the authors’ sole responsibility and do not necessarily represent official HRSA or NIH views. The funding agencies had no involvement in study design; collection, analysis, and interpretation of data; writing of the report; and the decision to submit the paper for publication.

ORCID iD

Carolyn DiGuiseppi

Presentation information

This study was presented at the International Society for Autism Research (INSAR) 2019 Annual Meeting, Montreal, Canada, May 1–4, 2019.

References

American Academy of Pediatrics. (2014). AAP publications reaffirmed or retired. Pediatrics, 134, e1520.

Baio

Wiggins

Christensen

D. L.

Maenner

M. J.

Daniels

Warren

Kurzius-Spencer

Zahorodny

Robinson Rosenberg

White

Durkin

M. S.

Imm

Nikolaou

Yeargin-Allsopp

Lee

L.-C.

Harrington

Lopez

Fitzgerald

R. T.

Hewitt

, et al. (2018). Prevalence of autism spectrum disorder among children aged 8 years—Autism and Developmental Disabilities Monitoring Network, 11 sites, United States, 2014. Morbidity and Mortality Weekly Report. Surveillance Summaries, 67(6), 1–23.

Broder-Fingert

Walls

Augustyn

Beidas

Mandell

Wiltsey-Stirman

Silverstein

Feinberg

(2018). A hybrid type I randomized effectiveness-implementation trial of patient navigation to improve access to services for children with autism spectrum disorder. BMC Psychiatry, 18(1), Article 79. https://doi.org/10.1186/s12888-018-1661-7

Campbell

Carpenter

K. L. H.

Espinosa

Hashemi

Qiu

Tepper

Calderbank

Sapiro

Egger

H. L.

Baker

J. P.

Dawson

(2017). Use of a digital Modified Checklist for Autism in Toddlers—Revised with follow-up to improve quality of screening for autism. The Journal of Pediatrics, 183, 133.e1–139.e1.

Carbone

P. S.

Campbell

Wilkes

Stoddard

G. J.

Huynh

Young

P. C.

Gabrielsen

T. P.

(2020). Primary care autism screening and later autism diagnosis. Pediatrics, 145(2), e20192314.

Charman

Baird

(2002). Practitioner review: Diagnosis of autism spectrum disorder in 2- and 3-year-old children. Journal of Child Psychology and Psychiatry, 43(3), 289–305.

Chlebowski

Robins

D. L.

Barton

M. L.

Fein

(2013). Large-scale use of the modified checklist for autism in low-risk toddlers. Pediatrics, 131(4), e1121–e1127.

Crossman

M. K.

Lindly

O. J.

Chan

Eaves

Kuhlthau

K. A.

Parker

R. A.

Coury

D. L.

Zand

D. H.

Nowinski

L. A.

Smith

Tomkinson

Murray

D. S.

(2020). Families’ experiences with family navigation services in the Autism Treatment Network. Pediatrics, 145(Suppl. 1), S60–S71.

Falkmer

Anderson

Falkmer

Horlin

(2013). Diagnostic procedures in autism spectrum disorders: A systematic literature review. European Child & Adolescent Psychiatry, 22(6), 329–340.

10.

Feinberg

Abufhele

Sandler

Augustyn

Cabral

Chen

Diaz Linhart

Cesar Levesque

Aebi

Silverstein

(2016). Reducing disparities in timely autism diagnosis through family navigation: Results from a randomized pilot trial. Psychiatric Services, 67, 912–915.

11.

Feudtner

Christakis

D. A.

Connell

F. A.

(2000). Pediatric deaths attributable to complex chronic conditions: A population-based study of Washington State, 1980–1997. Pediatrics, 106(1 Pt 2), 205–209.

12.

Fountain

King

M. D.

Bearman

P. S.

(2011). Age of diagnosis for autism: Individual and community factors across 10 birth cohorts. Journal of Epidemiology and Community Health, 65(6), 503–510.

13.

Freeman

H. P.

(2004). A model patient navigation program. Oncology Issues, 19, 44–46.

14.

Freeman

H. P.

Rodriguez

R. L.

(2011). History and principles of patient navigation. Cancer, 117(S15), 3537–3540.

15.

French

Kennedy

E. M. M.

(2018). Annual research review: Early intervention for infants and young children with, or at-risk of, autism spectrum disorder: A systematic review. Journal of Child Psychology and Psychiatry, 59(4), 444–456.

16.

Guevara

J. P.

Rothman

Brooks

Gerdes

McMillon-Jones

Yun

(2016). Patient navigation to facilitate early intervention referral completion among poor urban children. Families, Systems, & Health, 34(3), 281–286.

17.

Guthrie

Wallis

Bennett

Brooks

Dudley

Gerdes

Pandey

Levy

S. E.

Schultz

R. T.

Miller

J. S.

(2019). Accuracy of autism screening in a large pediatric network. Pediatrics, 144(4), e20183963.

18.

Hambidge

Phibbs

Chandramouli

Fairclough

Steiner

J. F.

(2009). A stepped intervention increases well-child care and immunization rates in a disadvantaged population. Pediatrics, 124(2), 455–464.

19.

Harris

P. A.

Taylor

Thielke

Payne

Gonzalez

Conde

J. G.

(2009). Research Electronic Data Capture (REDCap)—A metadata-driven methodology and workflow process for providing translational research informatics support. Journal of Biomedical Informatics, 42(2), 377–381.

20.

Irvin

D. W.

McBee

Boyd

B. A.

Hume

Odom

S. L.

(2012). Child and family factors associated with the use of services for preschoolers with autism spectrum disorder. Research in Autism Spectrum Disorders, 6(1), 565–572.

21.

Johnson

C. P.

Myers

S. M.

, & American Academy of Pediatrics Council on Children with Disabilities. (2007). Identification and evaluation of children with autism spectrum disorders. Pediatrics, 120(5), 1183–1215.

22.

Liptak

G. S.

Benzoni

L. B.

Mruzek

D. W.

Nolan

K. W.

Thingvoll

M. A.

Wade

C. M.

Fryer

G. E.

(2008). Disparities in diagnosis and access to health services for children with autism: Data from the National Survey of Children’s Health. Journal of Developmental & Behavioral Pediatrics, 29(3), 152–160.

23.

Lopez

Magaña

Morales

Iland

(2019). Parents taking action: Reducing disparities through a culturally informed intervention for Latinx parents of children with autism. Journal of Ethnic & Cultural Diversity in Social Work, 28(1), 31–49.

24.

Loudon

Treweek

Sullivan

Cheesbrough

G. F.

Donnan

Thorpe

K. E.

Zwarenstein

(2015). The PRECIS-2 tool: Designing trials that are fit for purpose. British Medical Journal, 350, Article h2147.

25.

Mandell

D. S.

Novak

M. M.

Zubritsky

C. D.

(2005). Factors associated with age of diagnosis among children with autism spectrum disorders. Pediatrics, 116(6), 1480–1486.

26.

Monteiro

S. A.

Dempsey

Berry

L. N.

Voigt

R. G.

Goin-Kochel

R. P.

(2019). Screening and referral practices for autism spectrum disorder in primary pediatric care. Pediatrics, 144(4), e20183326.

27.

National Center for Health Statistics. (2019). International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM). https://www.cdc.gov/nchs/icd/icd10cm.htm

28.

Pierce

Carter

Weinfeld

Desmond

Hazin

Bjork

Gallagher

(2011). Detecting, studying, and treating autism early: The one-year well-baby check-up approach. The Journal of Pediatrics, 159(3), 458.e1–465.e6.

29.

Raphael

J. L.

Rueda

Lion

K. C.

Giordano

T. P.

(2013). The role of lay health workers in pediatric chronic disease: A systematic review. Academic Pediatrics, 13(5), 408–420.

30.

Robins

D. L.

(2008). Screening for autism spectrum disorders in primary care settings. Autism, 12(5), 537–556.

31.

Robins

D. L.

Casagrande

Barton

Chen

C.-M. A.

Dumont-Mathieu

Fein

(2014). Validation of the modified checklist for autism in toddlers, revised with follow-up (M-CHAT-R/F). Pediatrics, 133(1), 37–45.

32.

Robins

D. L.

Fein

Barton

(2009). Modified Checklist for Autism in Toddlers, Revised with Follow-Up (M-CHAT-R/F)^TM. https://cms.m-chat.org/LineagenMChat/media/Lineagen-M-Chat-Media/mchatDOTorg.pdf

33.

Robins

D. L.

Fernandes

Rocha

M. L.

Stahmer

A. C.

Fein

D. A.

McClure

Vivanti

Dumont-Mathieu

Lee

N. L.

Yudell

(2019). Implementation of screening in pediatric primary care [abstract]. In INSAR 2019 annual meeting abstract book (pp. 252–253). https://insar.confex.com/insar/2019/webprogram/Paper29927.html

34.

Rosenberg

S. A.

Robinson

Fryer

G. E.

(2002). Evaluation of paraprofessional home visiting services for children with special needs and their families. Topics in Early Childhood Special Education, 22(3), 158–168.

35.

Roth

B. M.

Kralovic

Roizen

N. J.

Spannagel

S. C.

Minich

Knapp

(2016). Impact of autism navigator on access to services. Journal of Autism and Developmental Disorders, 37, 188–195.

36.

Selden

T. M.

(2006). Compliance with well-child visit recommendations: Evidence from the Medical Expenditure Panel Survey, 2000-2002. Pediatrics, 118, e1766–e1778.

37.

Szilagyi

P. G.

Humiston

S. G.

Gallivan

Albertin

Sandler

Blumkin

(2011). Effectiveness of a citywide patient immunization navigator program on improving adolescent immunizations and preventive care visit rates. Archives of Pediatrics & Adolescent Medicine, 165(6), 547–553.

38.

Thomas

Ellis

McLaurin

Daniels

Morrissey

(2007). Access to care for autism-related services. Journal of Autism and Developmental Disorders, 37(10), 1902–1912.

39.

Warren

McPheeters

M. L.

Sathe

Foss-Feig

J. H.

Glasser

Veenstra-VanderWeele

(2011). A systematic review of early intensive intervention for autism spectrum disorders. Pediatrics, 127(5), e1303–e1311.

40.

Wells

K. J.

Battaglia

T. A.

Dudley

D. J.

Garcia

Greene

Calhoun

Mandelblatt

J. S.

Paskett

E. D.

Raich

P. C.

, & Patient Navigation Research Program. (2008). Patient navigation: State of the art or is it science? Cancer, 113(8), 1999–2010.

41.

Windham

G. C.

Smith

K. S.

Rosen

Anderson

M. C.

Grether

J. K.

Coolman

R. B.

Harris

(2014). Autism and developmental screening in a public, primary care setting primarily serving Hispanics: Challenges and results. Journal of Autism and Developmental Disorders, 44, 1621–1632.

42.

Wolf

E. R.

Hochheimer

C. J.

Sabo

R. T.

DeVoe

Wasserman

Geissal

Opel

D. J.

Warren

Puro

O’Neil

Pecsok

Krist

A. H.

(2018). Gaps in well-child care attendance among primary care clinics serving low-income families. Pediatrics, 142(5), e20174019.

43.

Zeleke

W. A.

Hughes

T. L.

Drozda

(2019). Disparities in diagnosis and service access for minority children with ASD in the United States. Journal of Autism and Developmental Disorders, 49, 4320–4331.

44.

Zuckerman

K. E.

Lindly

O. J.

Reyes

N. M.

Chavez

A. E.

Macias

Smith

K. N.

Reynolds

(2017). Disparities in diagnosis and treatment of autism in Latino and non-Latino white families. Pediatrics, 139(5), e20163010.

45.

Zuckerman

K. E.

Lindly

O. J.

Sinche

B. K.

Nicolaidis

(2015). Parent health beliefs, social determinants of health, and child health services utilization among U.S. school-age children with autism. Journal of Developmental and Behavioral Pediatrics, 36, 146–157.

46.

Zuckerman

K. E.

Mattox

K. M.

Sinche

B. K.

Blaschke

G. S.

Bethell

(2014). Racial, ethnic, and language disparities in early childhood developmental/behavioral evaluations: A narrative review. Clinical Pediatrics, 53(7), 619–631.

47.

Zuckerman

K. E.

Sinche

Cobian

Cervantes

Mejia

Becker

Nicolaidis

(2014). Conceptualization of autism in the Latino community and its relationship with early diagnosis. Journal of Developmental and Behavioral Pediatrics, 35(8), 522–532.

48.

Zuckerman

K. E.

Sinche

Mejia

Cobian

Becker

Nicolaidis

(2014). Latino parents’ perspectives of barriers to autism diagnosis. Academic Pediatrics, 14(3), 301–308.

49.

Zwaigenbaum

Bauman

M. L.

Choueiri

Kasari

Carter

Granpeesheh

Mailloux

Smith Roley

Wagner

Fein

Pierce

Buie

Davis

P. A.

Newschaffer

Robins

Wetherby

Stone

W. L.

Yirmiya

Estes

, et al. (2015). Early intervention for children with autism spectrum disorder under 3 years of age: Recommendations for practice and research. Pediatrics, 136(Suppl. 1), S60–S81.

50.

Zwaigenbaum

Bauman

M. L.

Fein

Pierce

Buie

Davis

P. A.

Newschaffer

Robins

D. L.

Wetherby

Choueiri

Kasari

Stone

W. L.

Yirmiya

Estes

Hansen

R. L.

McPartland

J. C.

Natowicz

M. R.

Carter

Granpeesheh

, et al. (2015). Early screening of autism spectrum disorder: Recommendations for practice and research. Pediatrics, 136(Suppl. 1), S41–S59.

51.

Zwaigenbaum

Bauman

M. L.

Stone

W. L.

Yirmiya

Estes

Hansen

R. L.

McPartland

J. C.

Natowicz

M. R.

Choueiri

Fein

Kasari

Pierce

Buie

Carter

Davis

P. A.

Granpeesheh

Mailloux

Newschaffer

Robins

D. L.

, et al. (2015). Early identification of autism spectrum disorder: Recommendations for practice and research. Pediatrics, 136(Suppl. 1), S10–S40.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.17 MB