Secondary Logo

Journal Logo


Primary care of children with autism spectrum disorder

Pilot program results

Kilmer, Michele DNP, APRN, CPNP-PC

Author Information
doi: 10.1097/01.NPR.0000669128.73279.70
  • Free


This is the final article in a three-part series on pediatric autism spectrum disorder (ASD) management in primary care clinics. The first article provided an overview of the neurodevelopmental disorder and emphasized the importance of the primary care provider (PCP) as the leader of the healthcare team for pediatric patients with ASD. Realizing the importance of supporting PCPs to assume this leadership position, the second article addressed evidence-based interventions that should occur when a child is identified as at-risk for ASD, giving practical solutions for incorporating these interventions into primary care practice. This final article converges the concepts of the first two by detailing a pilot program, which aimed to equip PCPs to overcome barriers associated with pediatric ASD management in primary care clinics.

The CDC recognizes delays in evidence-based care in primary care clinics as a crucial concern for pediatric patients with ASD that must be addressed.1 Despite an extensive amount of research asserting evidence-based practice innovations in the care of pediatric patients with ASD, current primary care practices with these patients remain related to their poor health outcomes.1 Barriers to implementing evidence-based changes into practice warrant further investigation and must be resolved before vital improvements in care can occur. Common hindrances include time constraints, an overwhelming amount of available healthcare literature, and organizational structure demands.2,3 Another well-documented modern barrier is the research-to-practice gap phenomenon. PCPs have trouble keeping up with the numerous articles and recommendations from researchers and ASD advocacy organizations. Some resources are too lengthy and can seem overwhelming to follow considering the time limitations most PCPs must adhere to during each healthcare visit. Delays in evidence-based care will continue until research results are translated into a proven functional method for use in primary care clinics. Unfortunately, many efforts to reduce the research-to-practice gap are short-lived and do not result in sustainable or reproducible change.4 These obstacles also must be addressed to provide meaningful practice recommendations for primary care clinics, regardless of location or patient demographics.

The foundation of improved health outcomes for this patient population is built on PCP practice. Evidence-based care starts in the primary care clinic when at-risk children are identified and appropriate plans of care are created and implemented. Translational researchers must consider all known barriers when developing ASD management guidelines, programs, or educational aids for PCPs. PCPs need guidance provided in a clear and easily understood format. Clinical practice guidelines with algorithms may be the answer to the research-to-practice gap in pediatric patients with ASD. Clinical practice guidelines are another well-documented solution.5 These statements include recommendations that serve as a framework for clinical decisions that may optimize patient care.6 The creation and dissemination of clinical practice guidelines with algorithms for pediatric patients with ASD that are easy for PCPs to understand and follow are essential in the effort to meet recommendations for the evaluation and treatment of this patient population.

This 2-month translational research pilot program tested the usability of the author-developed clinical practice guideline with a corresponding algorithm for the initial management of pediatric patients at risk for ASD. The clinical practice guideline detailed evidence-based interventions for the initial management of pediatric patients, ages 9 to 60 months, who are identified as at risk for ASD through developmental screening in well-child visits. The pilot program evaluated the effect of the author-developed guideline on PCPs' understanding of pediatric ASD management, ability to correctly implement the guideline's interventions, and satisfaction with the practice change.


The participants in this pilot program included nine family nurse practitioners (FNPs) employed at one practice in two separate family medicine clinics. Their experience level ranged from 1 to 26 years as an FNP. Each FNP evaluated an average of 10 to 13 patients per day, which totaled approximately 1,500 patients per month. During the 8-week implementation period, approximately 2,000 children/adolescents were evaluated.

The formation of the author-developed guideline began with a detailed literature review using the Cumulative Index to Nursing and Allied Health Literature (CINAHL), PubMed, Cochrane Collection Plus, Ovid, and the National Guideline Clearinghouse. Recommendations from the CDC, National Institute of Mental Health (NIMH), the American Academy of Pediatrics (AAP), the American Academy of Neurology (AAN), and Healthy People 2020 were reviewed. Barriers to implementing evidence-based practice in primary care clinics were considered during guideline development. The second article in this three-part series expounded on the evidence supporting each part of the author-developed guideline.

The author-developed guideline and corresponding algorithm were created specifically for this pilot program, along with three surveys: the Provider Understanding of Pediatric Autism Management survey, the Health Provider Competency survey, and Clinical Practice Guidelines Satisfaction survey. The Provider Understanding of Pediatric Autism Management survey consists of 58 true/false questions written to evaluate the providers' knowledge of interventions in pediatric ASD management as well as prevalence and incidence data. To improve internal validity, each item in the provider knowledge survey corresponded with an evidence-based article. The Health Provider Competency survey includes 20 items formatted with a Likert scale to assess providers' confidence levels in managing the care of pediatric patients with ASD.

The authors scheduled an online meeting with the participating FNPs to discuss the details of the pilot program. Electronic versions of the provider knowledge survey and competency survey were sent to participating providers 1 week prior to the scheduled meeting to assess their understanding and confidence in managing pediatric ASD. Both surveys' results were analyzed and presented to the providers during the online meeting. Items incorrectly answered on the provider knowledge survey were reviewed along with the purpose of the pilot program and the proper use of the author-developed guideline and algorithm. The developmental screening tool used in the office prior to program implementation was the Modified Checklist for Autism in Toddlers, Revised with Follow-Up (M-CHAT-R/F). During the educational meeting, the recommendation to only use the M-CHAT-R/F for patients ages 16 to 30 months was discussed.7 The providers decided to use the Infant/Child Development Inventory (IDI/CDI) as a developmental screening tool for patients who fell outside of that age range.8 After the online educational session, the providers began an 8-week implementation phase using the author-developed guideline to direct care for all pediatric patients ages 9 to 60 months who presented for a well-child exam.

Study measurements

Outcome measures. Outcome measures assessed if evidence-based care was provided and were supported by the recommendations of the author-developed guideline. The first outcome measure calculated the percentage of patients, ages 9 to 60 months, who received a developmentally appropriate screening assessment during well-child exams. The second outcome measure investigated the percentage of patients who were identified with concerns for developmental delay during the well-child visit. The third outcome measure analyzed which of the identified patients were referred to specialists, including audiology, psychiatry, psychology, and occupational, physical, speech, and developmental therapies, per guideline recommendations. The fourth outcome measure examined the percentage of identified patients who received genetic or other lab testing.

Process measures. Process measures demonstrated how well the author-developed guideline was implemented within the daily practice of the primary care clinic. The first process measure evaluated the effect of the author-developed guideline implementation on provider understanding of pediatric ASD diagnosis, evaluation, and management. It was expected that the FNPs would experience increased understanding of evidence-based practice for the care of pediatric patients with ASD and that this enhanced understanding would be demonstrated in their practice. This process measure would be evaluated with the provider knowledge survey. Another process measure examined the FNPs' willingness to follow the author-developed guideline recommendations, as noncompliance could have affected outcome measures with no reflection of the accuracy of the guideline. Providers were to note all well-child evaluations on the data collection sheet and submit the sheets to the secretary on a weekly basis. If noncompliance was identified as a concern, more information would be gathered as to why noncompliance was occurring so that changes could be made to the guideline.

Balancing measures. Balancing measures assessed the FNPs' satisfaction with the author-developed guideline. This measurement was defined as the individual provider's opinion of guideline usability and effectiveness. A survey addressing provider satisfaction was distributed at the end of the implementation period to analyze this measure. This pilot program was approved by the University of Alabama Institutional Review Board. Anonymity was maintained in the following way: providers documented their practice in an Excel spreadsheet and sent these data to the pilot program author; the author did not have access to patient charts.


Two large pediatric practices and four family medicine practices were invited to participate in the pilot program. Five groups of providers declined participation. Considerable resistance to participate in a project where practice changes are expected was noted in two clinics. One clinic cited cultural differences between staff and the patient's families, financial restraints to purchase developmental screening tools, inadequate staffing resources, and time constraints that prohibited them from participating. Another clinic also stated time constraints and staffing concerns as the reasons for opting out of the program. Two clinics did not give a reason for not participating. The last facility, which was an NP-led primary care practice agreed to participate in the pilot study. The sampling size, set at 10 at the beginning of the recruitment period, was reduced because only nine PCPs were employed by the participating family medicine practice.

Provider knowledge survey results

The average score for the preimplementation provider knowledge survey was 72%, with a range of 62% to 86% and median score of 73%. The questions missed by most of the providers concerned the Healthy People 2020 objective MICH-29.3 that calls for an increase in enrollment in special services by age 48 months for patients with a diagnosis of ASD, the prevalence of pediatric patients with ASD that are insured by Medicaid, and the core symptoms of ASD. During the implementation period, one of the FNPs left the clinic, so that provider did not participate in the postimplementation surveys. The remaining eight providers completed the same provider knowledge survey they had taken at the start of the implementation period prior to the educational meeting. There was one striking outlier, as one provider's results were drastically different than the others. That provider completed the surveys in 2 minutes and 30 seconds compared with seven other practitioners whose survey time averaged 7 minutes and 27 seconds. That provider's answers were removed from the analysis of the project results. The average score of the remaining seven providers increased to 92% with a standard deviation of 5%, a range of 86% to 100%, and median score of 92%. The questions most missed were consistent with the preimplementation phase: The Healthy People 2020 objective MICH-29.3 and the core symptoms of ASD. A third question, involving ASD prevalence, was missed more in the postimplementation phase than in the preimplementation phase.

Provider competency survey results

The preimplementation competency survey indicated that many providers, despite having a basic understanding of ASD, did not feel comfortable managing the care of pediatric patients with ASD, especially patients with ASD who also have metabolic or genetic concerns. The majority felt comfortable referring patients to speech therapy and other specialty providers for further workup but did not feel confident in ordering genetic or other lab workup. All nine providers were adept at using the M-CHAT-R/F developmental screening tool, but none were using any tool to screen children whose ages fell outside of the M-CHAT-R/F testing range.

The postimplementation competency survey confirmed the effectiveness of the author-developed guideline in building the providers' confidence level in caring for pediatric patients with ASD. All providers stated that they felt confident in evaluating and treating the general needs of pediatric patients with ASD after the implementation period. The competency scores related to ordering genetic or other lab workup and referring patients to physical therapy, occupational therapy, and audiology improved after the implementation period. Provider insight also grew in understanding pediatric ASD symptoms and diagnostic criteria. One hundred percent of providers not only felt comfortable with the developmental tools used during the implementation period, but also felt confident in providing care for pediatric patients with ASD as they aged. Sixty-two percent of providers stated that they planned to seek yearly educational opportunities to enhance their understanding of pediatric ASD after the implementation phase, and the percentage of providers who expressed familiarity with ASD advocacy and family support groups in their area grew from 11% to 50%. (See Pre- and postimplementation Health Provider Competency survey scores.)

Satisfaction survey

The satisfaction survey was administered to the eight FNPs who completed the implementation period. The survey consisted of nine questions that used a Likert scale with Strongly Disagree, Somewhat Disagree, Neither Disagree nor Agree, Somewhat Agree, and Strongly Agree to analyze the degree to which participation in the pilot program improved their ability to deliver evidence-based care to pediatric patients with ASD. Outlier data for this survey were included in the results. The results of this survey indicate the program was very beneficial for the providers and achieved its goal of increasing access to evidence-based care for pediatric patients with ASD. Seventy-three percent of the providers stated that the pilot program improved their ability to provide evidence-based care to pediatric patients for whom there is a concern for ASD. Eighty-eight percent stated they learned new evidence-based recommendations by participating in the program. One hundred percent stated they were able to expand their professional network by participating in the project. Eighty-eight percent stated the online introductory meeting and author-developed guideline with algorithm improved their understanding of ASD incidence, screening, evaluations, and evidence-based interventions. One hundred percent stated it was easy to use in their everyday practice, the format was easy to understand, and was applicable to their practice's patient population. The results of the satisfaction survey indicate that the author-developed guideline and algorithm could easily be adapted into the daily routine of a primary care practice.

Pre- and postimplementation Health Provider Competency survey scores

Secondary data

Secondary data were collected to analyze the effect of the author-developed guideline on patient care. A total of nine pediatric patients were evaluated during the implementation period; six were males and three were females. The patients were all White, and English was the primary language spoken in the home for all. These nine patients were the only ones who presented to the clinics for well-child evaluations during the implementation period. The first outcome measure was met as all of the nine patients received developmental screening using an appropriate screening tool for each patient's age. The providers used the Child Development Inventory screening tool on seven of the patients. Two of the patients were evaluated with the M-CHAT-R/F. The patients ranged between 12 and 49 months, with a median age of 24 months, and an average age of 22.2 months. The second outcome measure was to evaluate the number of patients identified as at-risk for developmental delay. Three out of the nine patients were identified as having developmental delay concerns per the screening tools used. The providers then followed the author-developed guideline and algorithm to direct the care of these patients based on their individual needs.

The third outcome measure analyzed which specialty referrals were made once patients were identified as at-risk for developmental delay. One patient (age 20 months) was referred to speech therapy and further developmental evaluation. Another patient (age 37 months) was referred for speech therapy and to the local Early Childhood Intervention program. A third patient (age 18 months) was referred to occupational therapy, further developmental evaluation, and the Early Childhood Intervention program. This patient also presented with sleep dysfunction. None were referred for an audiology evaluation, physical therapy, psychiatry or psychology, gastroenterology, or genetic specialist, and none underwent serum lab testing.

Dissemination to the clinical site

The providers' scores on the surveys and the secondary findings were reviewed with the providers and office manager within 1 month after the implementation phase ended. ASD prevalence, screening guidelines, core ASD symptoms, and the importance of referring to audiology for evaluation of developmental or speech delay were also discussed. The clinic providers felt the following adaptions to the program would be beneficial: The use of a clinic liaison to assist with questions and the referral process, the consistent use of one developmental screening tool instead of two or three, a contact list of allied health professionals who accept evaluation and treatment prescriptions for pediatric patients with ASD, and concise algorithms detailing the different action courses for management of comorbid conditions.


Limitations of the pilot program include the small number of patients who were evaluated during a short 8-week implementation period and unintentional bias in the surveys created for the pilot program. The provider knowledge survey was created through a lengthy process examining current evidence-based literature. Each item on the survey corresponded with different evidence-based articles. While this method may improve face validity of the tool, methods to improve concurrent and population validity should be undertaken for future use. The providers completed the same knowledge survey pre- and postimplementation, which may account for the improvement in scores; however, it is noted that the tests were administered over 8 weeks apart to decrease this possibility. Also, only FNP providers elected to participate in the pilot program even with concentrated effort to obtain a varied group of PCPs from different regions and educational backgrounds. Despite this, the preliminary results warrant a second study that addresses these concerns that result in PCP resistance to managing the care of these vulnerable patients. The recommendations suggested by the FNPs were reasonable and should be incorporated into the next study.

Also, even though 83% of the providers stated in the postimplementation competency survey that they routinely refer pediatric patients with a concern for ASD to an audiologist, none of the providers referred any of the identified patients for an audiology exam. None of the patients were referred to physical therapy, psychiatry or psychology, gastroenterology, or a genetics specialist, and none of the patients underwent serum lab testing. These practices should be explored further as it is unknown if these omissions were the result of examining the patients' past medical history, family history, social history, and developmental testing results or if the providers chose the referrals they are most comfortable with. This practice may improve by incorporating a clinic liaison who can oversee the process, provide educational opportunities, and answer questions as they arise.

Barriers not specifically addressed within the guideline include financial restraints, cultural differences, and inadequate staffing. The two clinics that participated in the pilot project received free access to developmental screening tools to use during the implementation phase, whereas primary care clinics must look at their resources and purchase these tools for their sites. Cultural differences in recognizing the need for developmental assessment and referral to appropriate specialty clinics are not addressed in the guideline. Parental misinterpretation of core ASD symptoms, varying ASD symptomology, perceived stigma of an ASD diagnosis, mistrust of the American healthcare system, and seeking healthcare when in crisis mode as opposed to participating in preventive healthcare services are known factors influencing the delay in diagnosis among minority populations.9 Staffing resources to oversee the developmental screening process and referrals may be alleviated with the addition of a clinic liaison.

Last, the exclusion of the results from the provider who left the clinic and the outlier results could have interfered with the comparison of the preimplementation and postimplementation provider knowledge surveys.


The preliminary results of the pilot program suggest that the author-developed guideline can easily be integrated into daily practice routine and enhance the management of pediatric patients at risk for ASD. The results are clinically significant and warrant further study. Secondary data are beneficial as they demonstrated that the author-developed guideline was successful in identifying and managing the care of the three patients who exhibited developmental delay concerns. Considering that approximately 15% of children ages 3 to 17 years have some type of developmental disorder, this is an important issue for all pediatric practices.10

Additionally, the satisfaction survey scores support the use of this program in primary care clinics. The participating providers agreed that the pilot program improved their understanding of pediatric ASD and positively affected practice change. Their recommendations of additional support measures should be incorporated into subsequent studies. It is noted that survey questions incorrectly answered in the preimplementation and postimplementation provider knowledge surveys were similar. The use of a clinic liaison could address provider understanding of ASD management and may result in decreased hesitancy in overseeing the care of pediatric patients with ASD. Future studies could incorporate a clinic liaison to supply educational materials and conferences to improve the providers' understanding of pediatric ASD through scheduled educational interventions.

The pilot program enrollment process uncovered several factors that must be investigated further to institute change in the management of pediatric ASD. Resistance to practice changes, time constraints, inadequate staffing, cultural barriers, and financial restraints must be investigated and solutions to these concerns should be presented for any changes to be successful in expanding access to evidence-based care for patients with ASD.


This translational research pilot program aimed to overcome known barriers found in primary care clinics and increase access to evidence-based care for pediatric patients with ASD through the dissemination of a clinical practice guideline and algorithm developed by the authors. The guideline and algorithm were specifically written to address time constraints and other barriers to improve providers, knowledge of ASD management. Data collected after the 8-week implementation period confirmed that the author-developed guideline was successful in identifying pediatric patients who are at risk for ASD. Patients identified with developmental delay were referred to appropriate services the day of the well-child evaluation. The results from the satisfaction survey verify that the program can be reasonably and easily integrated into practice in primary care clinics. The pre- and postimplementation provider knowledge surveys show that provider understanding improved after receiving education on recommended practices. The recruitment process uncovered hindrances to participation in a program that would change current practices of management of pediatric patients with ASD for primary care clinics. Time constraints, inadequate staffing, cultural barriers, and financial restraints were identified as specific barriers to participation in a program to increase access to care for pediatric patients with ASD. Future studies must address these concerns in order to improve health outcomes related to pediatric ASD. (See Key concepts.)

Key concepts

  • Current practice in primary care clinics does not promote optimal health outcomes for pediatric patients with ASD.
  • PCPs need educational support that is easy to understand and implement into practice.
  • Efforts to improve early identification and intervention in pediatric ASD must account for known barriers present in primary care clinics.


1. Maenner MJ, Shaw KA, Baio J, et al. Prevalence of autism spectrum disorder among children aged 8 years—autism and developmental disabilities monitoring network, 11 sites, United States, 2016. MMWR Surveill Summ. 2020;69(No. SS-4):1–12.
2. Majid S, Foo S, Luyt B, et al. Adopting evidence-based practice in clinical decision making: nurses' perceptions, knowledge, and barriers. J Med Libr Assoc. 2011;99(3):229–236.
3. Wallis L. Barriers to implementing evidence-based practice remain high for U.S. nurses: getting past “we've always done it this way” is crucial. Am J Nurs. 2012;112(12):15.
4. Kristensen N, Nymann C, Konradsen H. Implementing research results in clinical practice—the experiences of healthcare professionals. BMC Health Serv Res. 2016;16:48.
5. Graham R, Mancher M, Miller Wolman D, et al. Clinical Practice Guidelines We Can Trust. Washington, DC: National Academies Press; 2011:6.
6. AAFP. Clinical practice guideline manual. 2017.
7. Robins DL. M-CHAT-R/F. 2020.
8. Ireton H. Child Development Inventory. Minneapolis, MN: Behavior Science Systems, Inc.; 1992.
9. Zeleke WA, Hughes TL, Drozda N. Disparities in diagnosis and service access for minority children with ASD in the United States. J Autism Dev Disord. 2019;49(10):4320–4331.
10. Zablotsky B, Black LI, Maenner MJ, et al. Prevalence and trends of developmental disabilities among children in the United States: 2009-2017. Pediatrics. 2019;144(4):e20190811.

clinical practice guideline; evidence-based care; pediatric autism spectrum disorder; primary care provider; research-to-practice gap

Wolters Kluwer Health, Inc. All rights reserved.