Integrating a New Online Autism Screening Tool in Primary Care to Lower the Age of Referral

Introduction

The earliest signs of autism spectrum disorder (ASD) may appear at 12 to 14 months of age, and a diagnosis of ASD can be made in the second year of life by experienced professionals.^1-3 Research indicates that early intervention is effective in improving children’s developmental trajectories and reducing autism symptoms.^4,5 However, the average age of diagnosis for ASD is approximately 4½ years, ⁶ and the age of diagnosis is further delayed for children from underserved areas, low-income, and minority families. ⁷

The American Academy of Pediatrics (AAP) recommends screening all children for ASD at 18 to 24 months. ⁸ However, a recent US Preventive Services Task Force report found inconclusive evidence to support universal screening in primary care prior to parents’ expression of concerns. ⁹ One barrier to consistent early screening is lack of well-validated ASD-specific screeners for younger toddlers. ¹⁰ Studies of the Modified Checklist for Autism in Toddlers (M-CHAT), ¹¹ a widely used screening tool for ASD, suggest that it may miss more children with ASD than it catches when implemented at 18 months ¹² and may perform better for older (eg, 24-30 months) than younger (eg, 16-23 months) toddlers.^13,14 In addition, surveys of health care professionals suggest variability in the rates at which they follow the screening guidelines set forth by the AAP.^15,16

Bridging research to practice gaps is a national priority, as the reported prevalence of ASD ⁶ and societal costs increase. ¹⁷ Building capacity of providers to recognize early signs of ASD is critical to early detection and increased access to early intervention services. Approximately 95% of children complete routine health supervision visits from birth to 3 years of age. ¹⁸ Therefore, primary care physicians represent a crucial target for support on screening, diagnosis, and evidence-based intervention.^19,20 Online professional development and the utilization of digital screening tools are 2 potential avenues to facilitate large-scale practice change. Online coursework is cost-efficient, time-flexible, delivered with high procedural fidelity, and has been found to be as effective as other training methods. ²¹ Evidence suggests that using digital platforms for screening in primary care is feasible, may identify more children, may increase follow-up screens, and that families from diverse socioeconomic backgrounds report positive experiences with online screening tools.^22,23

Purpose of the Study

The purpose of this study was to report on a quality improvement (QI) project designed to systematically analyze, review, and improve screening practices for ASD in a busy pediatric office in central Florida. The project addressed 3 aims: (1) to examine office-based screening practices, (2) to integrate an online screening tool during eligible well-child visits, and (3) to compare the average age of referral for ASD evaluation and eligibility for intervention before and after integration of online screening.

Methods

The QI team included medical providers, students, and office staff. Providers and medical students were trained using a new online professional development course, Autism Navigator for Primary Care (ANPC). ²⁴ ANPC is an 8-hour course designed to increase the capacity for universal screening in primary care to improve early detection and referral for young children with or at risk for ASD. The course content includes a video library of more than 2 dozen toddlers with ASD at 18 to 24 months of age and features side-by-side videos to contrast toddlers with typical development with those showing early signs of ASD. Video clips illustrate core diagnostic features, key social communication milestones and early signs of ASD, gathering and sharing information with families, and evidence-based early intervention strategies used by families of toddlers with autism in everyday activities.

Autism Navigator for Primary Care also provides training for administration and interpretation of the Smart Early Screening for Autism and Communication Disorders (Smart ESAC). ²⁵ The Smart ESAC is a digital tool designed as a universal screener beginning with a brief, 10-question broadband screen for communication delay, which, if positive, is followed by 20 autism-specific screening questions. Field testing with follow-up of more than 850 children indicated good sensitivity (range = .81-.84 across age intervals) and specificity (range = .70-.89). ²⁶ The Smart ESAC can be administered via computer, tablet, or smartphone with automated scoring to simplify result sharing with parents and upload to an electronic health record (EHR). Once screened, parents are invited to an online portal through which subsequent screenings can be completed prior to future office visits. Parents can also access several resources and tools based on the outcome of their child’s screen.

Evaluation Methods and Data Collection

The QI project was based on Model for Improvement. ²⁷ The Smart ESAC was integrated into standard practice for 1 year, spanning from August 2017 to September 2018, and the number of eligible 12- to 30-month well-child visits and screenings administered was recorded through 5 periodic reviews of medical records. Chart review was then conducted on all patients in the practice with an International Classification of Diseases, 10th Revision (ICD-10) code for ASD (F84) between 2001 and 2017 to calculate the average age of referral for early intervention eligibility. Prior to the integration of the Smart ESAC, the M-CHAT was administered at 18- and 24-month visits per AAP recommendations. Pre- and post-Smart ESAC implementation data were compared to assess impact on referral and intervention timing.

Results

A total of 391 well-child visits was reviewed over the 1-year QI project in five 2-month intervals (n = 86, n = 61, n = 69, n = 99, and n = 76, respectively; see Figure 1). The following trends were observed: (1) screening rates of 100% were observed for 12- and 18-month visits by the final interval in the study, indicating full integration of the Smart ESAC; (2) there was marked improvement in screening rates for the 15-month visit; and (3) the 24- and 30-month screenings showed a strong increase in integration until the final time interval.

Figure 1.

Integration of Smart Early Screening for Autism and Communication Disorders in well-child visits over time.

Chart review revealed that, in the 16 years prior to Smart ESAC integration, 33 patients had an ICD-10 code for ASD with an average age of referral for a diagnostic evaluation at 37.2 months (range = 15-66 months; SD =13.73). After implementation of the Smart ESAC, 20 patients were referred for a diagnostic evaluation at an average age of 19.7 months (range = 12-30 months; SD = 4.52). Results of an independent samples t test indicated a statistically significant reduction in average age of referral, with a large effect size: t(51) = 5.51, P < .001; Cohen’s d = 1.71.

Discussion

The purpose of this study was to implement a new online professional development course and digital screening tool combining broadband and autism-specific questions into a busy pediatric practice to promote earlier detection of young children with signs of ASD. Prior to the study, the average age of referral using the M-CHAT was around 3 years of age. Over the course of 1 year, results indicated an increase in the number of children screened earlier at well-child visits, beginning at 12 months of age. At the end of data collection, over 70% of visits at 15 months included a Smart ESAC screening, with a screening rate of 100% for the 12- and 18-month visits. In addition, the average age of referral for early intervention eligibility following a positive screen for ASD dropped to 20 months. Results of this study support previous research demonstrating that an online screening tool was feasible for both parents and clinicians.^21,22 Within 1 year, integration of the Smart ESAC was identified by this practice as a primary change associated with improvements in the detection of early signs of autism.

Strengths and Limitations

The study has limitations. There were barriers to initiating this QI project, including connecting touchscreen tablets to the secure office network, which was resolved through communication and consultation between ANPC staff and IT (information technology) personnel at the practice. Next, parents were permitted to refuse the screening, although this was rare during the project. After the first interval, the Smart ESAC was integrated as standard practice and no patients refused in the final 4 intervals. Third, we combined results from 24- and 30-month age intervals and coded them as “last screen” because we observed that our patients were either screened at 24 or 30 months. This occurred for 2 reasons: (1) most patients with positive screens for ASD had already been referred for evaluation by 30 months, and (2) patients who had several consecutive negative screenings were less likely to be screened again at 30 months. Finally, it will be important to monitor the patients’ outcomes to 36 months and beyond.

Conclusions

Building on results of this QI project, we are working to further integrate screening with the Smart ESAC for all eligible children at well-child visits. Future PDSA (Plan-Do-Study-Act) cycles will include encouraging parent portal utilization to increase at-home screening prior to office visits, which could streamline efficiency and office flow. Efforts are underway to incorporate the Smart ESAC into the standard of care at 21 pediatric offices with over 50 clinicians. Other next steps are to create a direct link between the Smart ESAC parent portal and the EHR platform to automatically transfer the ESAC report into the EHR. In conclusion, we found that using a digital broadband and autism-specific screening tool has great potential to improve earlier detection and lower the age of referral for evaluation, ultimately allowing families to access early intervention and promote better outcomes for our patients with ASD.