DEVELOPMENTAL DISABILITY affects approximately 15% of U.S. children, amounting to nearly 10 million children. This represents an increase of 17% between 1997 and 2008 (Boyle et al., 2011), with higher rates reported among children who live in poverty (Stevens, 2006). The rate of children with an autism spectrum disorder (ASD) has recently been reported to be approximately one in 88 children (Baio, 2012), an increase from prior reported rates of 1% of children in the United States (Kogan et al., 2009; Rice, 2009), with some recent studies suggesting even higher worldwide prevalence rates (Kim et al., 2011).
Although early detection and intervention have been shown to improve short- and long-term outcomes (Anderson et al., 2003; Bailey et al., 2005; Hill, Brooks-Gunn, & Waldfogel, 2003), only 10% of children with delays are reported to have received services by 24 months (Rosenberg, Zhang, & Robinson, 2008) and children are still being diagnosed with ASD at 5–6 years of age on average, despite the ability to make a reliable diagnosis at 2 years of age (Shattuck et al., 2009; Wiggins, Baio, & Rice, 2006).
Primary care pediatricians play an important role in identifying children with developmental delay and ASD because they have frequent contact with children younger than 3 years. Although pediatricians endorse the importance of routine screening (Schonwald, Huntington, Chan, Risko, & Bridgemohan, 2009), frequent barriers often cited regarding the use of standardized screening tools include time, reimbursement, and lack of staff to implement a screening program (Sand et al., 2005; Sices, Feudtner, McLaughlin, Drotar, & Williams, 2003). In 2006, the American Academy of Pediatrics (AAP) issued a policy statement recommending systematic developmental screening in primary care with a standardized screening tool for children 9, 18, and 24 or 30 months of age (American Academy of Pediatrics Council on Children With Disabilities, Section on Developmental Behavioral Pediatrics, Bright Futures Steering Committee, & Medical Home Initiative for Children With Special Need, 2006). In addition, it was recommended that if a concern was raised by a parent or a clinician during a surveillance visit, developmental screening should be performed at that visit or shortly thereafter (American Academy of Pediatrics Council on Children With Disabilities, Section on Developmental Behavioral Pediatrics, Bright Futures Steering Committee, & Medical Home Initiative for Children With Special Need, 2006). In 2007, the AAP further recommended that ASD-specific screening should be performed on all 18- and 24-month-old children during well-child visits (Johnson, Myers, & American Academy of Pediatrics Council on Children With Disabilities, 2007). Prior to the publication of these guidelines, 23%–50% of pediatric providers surveyed reported always or almost always using standardized instruments for developmental screening (Sand et al., 2005; Sices et al., 2003).
Since the publication of the policy statement, a number of initiatives have been implemented across the country to improve developmental screening, including the Assuring Better Child Development screening academy (Earls, 2009) and the AAP's Developmental Surveillance and Screening Policy Implementation Project (King et al., 2010). Only a handful of studies have examined changes in screening practice since the guidelines were published. A 2011 publication based on findings from the 2009 AAP periodic survey found that the frequent use of standardized screening had increased to 47.7% (Radecki, Sand, O'Connor, Sharp, & Olson, 2011). Other studies with smaller samples have suggested that as many as 80% of pediatricians surveyed conduct routine developmental screening (Gills, 2009). A study from the 2007 National Survey of Children's Health revealed that 19.5% of parents recalled completing developmental screening about their child within the previous 12 months, with a range of 10.7%–47% across different states. A total of 29 states had a parent-completed developmental screening prevalence rate that was lower than the national rate (Bethell, Reuland, Schor, Abrams, & Halfon, 2011). Similar increases in ASD screening practice are not reported to be as high, with rates ranging from 8% to 28% of pediatricians routinely screening (Dosreis, Weiner, Johnson, & Newschaffer, 2006; Gills, 2009).
Studies to date that have examined current screening practices have been limited in scope, asking only whether screening occurred or not and what instruments were used. Furthermore, there are very few studies that have surveyed clinicians nationally, with most focusing on small, local populations. Therefore, the purpose of this study was threefold. First, to determine compliance with three AAP screening recommendations: (1) developmental screening at 9, 18, and 24 or 30 months; (2) developmental screening when concerns are raised by the pediatrician or parent at a surveillance visit; and (3) autism screening at 18 and 24 months. Second, to describe how children are currently screened in practices in terms of location, availability and roles of office personnel, and timing during the visit. Third, to examine pediatrician and practice characteristics associated with compliance with AAP screening recommendations.
Study subjects were pediatricians who were fellows of the AAP and provided primary care to children younger than 3 years. The Connecticut AAP executive director contacted all AAP state chapter executive directors by e-mail to invite them to participate in a national study of current developmental and autism screening practices. A total of eight executive directors from seven states (some larger states have more than one executive director) expressed interest, with six states ultimately participating (Connecticut, Hawaii, Massachusetts, Michigan, New Jersey, and New York). Individual state responses will not be reported in this paper because this was an agreed-upon condition of participation. The Institutional Review Board at Yale University School of Medicine approved the study.
A web-based physician-completed survey was adapted from a questionnaire by Sices et al. (2003). State AAP executive directors sent the survey to all state chapter members with available e-mail addresses. Up to three reminders were sent at the discretion of each state's executive director between April and May 2011. As an incentive to participate, subjects were given the option of participating in a drawing for a prize of $50 gift card to Amazon.com. Pediatricians who did not provide primary care to children younger than 3 years were ineligible for participation.
The survey contained a total of 38 questions that used either a 5-point Likert scale or a multiple-choice format. It contained questions related to compliance with three of the 2006 and 2007 AAP policy statement recommendations: (1) completing developmental screening at 9, 18, and 24 or 30 month visits, using a standardized instrument; (2) completing developmental or autism screening when a concern is raised by a parent or provider at a surveillance visit (not a 9-, 18-, or 24- or 30-month visit), using a standardized instrument at that visit or at a scheduled return visit; and (3) completing autism screening at 18- and 24- or 30-month visits, using a standardized instrument. For this study, compliance was defined as a pediatrician rating him- or herself as always screening on a 5-point scale ranging from “never” to “always.”
Questions were asked about the screening implementation within a practice including (1) screening instruments used, (2) when during the visit screening takes place, (3) who administers and scores the screens, and (4) where screening is completed. Pediatricians were asked about changes in their practice over the last 5 years related to developmental and autism screening and the reasons for any changes. The demographic portion of the survey included questions about the pediatrician (age, gender, race, and ethnicity) and the practice location and patient composition (percentage of non-White patients in the practice and percentage of patients who are insured by Medicaid). All questionnaire responses were anonymous.
Chi-square analyses were used to compare the characteristics both of pediatricians and of their practices, comparing those who complied with AAP screening guidelines with those who did not. Multivariate models consisting of logistic regressions were also used to compare differences between those clinicians who were compliant with guidelines and those who were not while controlling for variables found to be significant on bivariate analyses. Analyses were completed using SPSS, Version 18 (SPSS, Chicago, IL).
Within the six states that participated in the survey, state AAP executive directors reported sending the survey to approximately 4,100 members in total. A total of 408 surveys were returned for an average response rate of 10%. Incomplete surveys and surveys completed by residents were not included, resulting in a final sample of 281. The demographic and practice characteristics of the sample, compared with the 2009 AAP periodic survey of fellows, are shown in Table 1. The 2009 AAP periodic survey contains information about from where the most recently published national study of developmental screening obtained its data. The current sample was significantly different from that sample in terms of race/ethnicity and community type of practice. When examining the sample in this study, respondents were composed of more females and the majority of the sample identified their race/ethnicity as White, non-Hispanic. Most respondents were in a suburban, private practice. The 2009 sample had greater racial and ethnic diversity and more respondents from urban and rural practices.
Compliance with AAP guidelines
The findings of pediatricians who were compliant with each of the three current screening recommendations are shown in Table 2. A total of 17.8% of respondents were compliant with all three screening recommendations. Physicians reported using standardized screens for developmental delay for less than half of all visits or never 46.3% of the time for the 9-month visit, 30.2% for the 18-month visit, and 34.9% for the 24- or 30-month visit. For autism screening, physicians reported using standardized instruments less than half of all visits or never 28.5% of the time for the 18-month visit and 37.8% for the 24- or 30-month visit. Compared with 5 years ago, 44.8% of respondents reported using standardized developmental screening tools more often and 72.2% reported using standardized autism screening tools more often. The most common reasons cited for this increase were a desire to maintain a standard of care (65.5%), the collective decision of a practice (58%), the publication of the AAP policy statements (50.5%), and improved reimbursement for developmental screening (26%). Other less common reasons endorsed included attending a continuing medical education or training (16.7%), state regulation requirements (7.1%), and parent request (3.2%).
As is shown in Table 3, when standardized developmental and autism screens are administered, most screens are completed in the examination room, with the waiting room being the second most common site of screening completion. In total, 91.6% of all standardized screens are completed just before or during the well-child visit. Although one third of parents complete the screens on their own, doctors and nurse practitioners are the practice personnel who most frequently assist parents with completing screens and scoring them. The Ages and Stages Questionnaire is the most common instrument reported that is being used to screen for developmental delays. Those who checked “other” to report their use of standardized instruments for developmental screening most often endorsed the Modified Checklist for Autism in Toddlers (M-CHAT) as the instrument used. About autism screening, the M-CHAT was overwhelmingly the most common measure being used.
Characteristics associated with screening compliance
To identify differences between pediatricians who were compliant with each of the three AAP recommendations and those who were not, a series of bivariate comparisons was completed as shown in Table 4. No specific pediatrician or practice characteristic was identified that was significantly associated with compliance for all of the three recommendations. Medicaid insurance status, however, was significantly associated with compliance in autism screening and screening when concerns are raised and approached significance for developmental screening.
Multivariate analysis of compliance
A series of multivariate models examined pediatrician and practice characteristics associated with compliance with each of the AAP recommendations. Independent variables were characteristics previously identified that distinguished compliant versus noncompliant pediatricians for at least one screening recommendation on bivariate analysis. Pediatrician gender and race were also included in all models. Separate multivariate models were constructed for each of the three practice characteristics: (1) practice location; (2) non-White race/ethnicity percentage of the practice; and (3) Medicaid-insured percentage of the practice) because these variables were found to be highly overlapping. Therefore, a total of nine multivariate models were constructed (three for each AAP recommendation), and these findings are shown in Table 5.
There were no pediatrician characteristics that were associated with compliance with developmental screening. Suburban practices were significantly less likely to screen for development than urban ones, with rural practices trending toward screening less than urban practices. Pediatricians with 10%–50% of patients of non-White race/ethnicity in their practice were significantly less likely to screen for developmental delay than were those with more than 50% of patients, as were pediatricians with 10%–30% of Medicaid-insured patients compared with those with more than 30% of patients.
Very few pediatrician or practice characteristics distinguished compliant versus noncompliant pediatricians. As can be seen in Table 5, pediatricians aged 51–60 years were significantly more likely to screen children for autism than younger pediatricians. In addition, practices with 10%–30% Medicaid-insured patients were significantly more likely to be screened for autism than were those with more than 30% of patients. Although pediatricians with less than 10% Medicaid-insured patients in their practice were not significantly different from pediatricians with more than 30% of patients, the trend was in the direction of increased screening.
Screening when concerns raised
As seen in Table 5, similar to autism screening, pediatricians aged 51–60 years were significantly more likely to screen children when concerns were raised during a visit, whereas those in suburban practices were significantly less likely to screen children when concerns were raised during a visit. Pediatricians with less than 30% Medicaid-insured patients were significantly less likely to perform screening when concerns were raised, as were the pediatricians in suburban practices. Pediatricians with 21%–50% patients of non-White race/ethnicity were also less likely to screen when concerns were raised than those with more than 50% of patients.
This study suggests that pediatricians who responded to our survey were screening children more often for developmental delay and autism than had been reported in previous studies of screening practice. We found between 41.6% and 59.8% of pediatricians reported compliance with at least one of the three screening recommendations that have been advanced by the AAP. Although these rates are a significant increase compared with screening rates in earlier studies (Sand et al., 2005; Sices et al., 2003), only 17.8% of pediatricians who responded to this survey were compliant with all three AAP recommendations regarding developmental and autism screening. An additional 30%–45% of pediatricians reported never, or infrequently, using standardized screens at suggested intervals or when significant concerns were raised during a visit. These data suggest that there is still significant room for growth in the number of pediatricians who consistently implement recommended practices for the timing of developmental screenings with infants and young children.
The rates reported by pediatricians in this study of implementing developmental screens were similar to those reported on the basis of the AAP 2009 periodic survey, in which 47.7% of pediatricians reported always or almost always using a developmental screening tool (Radecki et al., 2011). Similar to the study of Radecki et al., we found that the Ages and Stages Questionnaire was the most commonly used developmental screening tool but that the Denver Developmental Screening Test II was still used by survey respondents despite its poor psychometric properties (Glascoe et al., 1992). The M-CHAT was the most commonly used screen for autism. Screening for autism has skyrocketed in comparison with rates of 8% reported in earlier studies (Dosreis et al., 2006), and nearly 75% of respondents reported using a standardized autism screen when compared with 5 years ago. In our sample, pediatricians most commonly reported increasing their use of standardized screens due to a desire to maintain a standard of care and as a result of the publication of the AAP policy statement.
In this study, pediatricians stated that they and nurse practitioners scored screens and approximately half of all screens were completed during the office visit. It was further reported that staff assisted parents approximately 50% of the time to complete screening tools (e.g., through parent report). This suggests that screening practices in this sample rely on clinician's time to complete even when a tool designed to be completed by parents is used.
This study also revealed significant differences in the implementation of screening based on practice demographics. Suburban children in practices with lower percentages of children insured by Medicaid were less likely to be screened for developmental delays at standardized intervals or when concerns were raised by parents during a visit. Pediatricians in practices with fewer non-White children were less likely to screen for developmental delays as well. In contrast, practices with 10%–30% of Medicaid-insured children were more likely to screen children for autism than pediatricians with practices with more Medicaid-insured patients.
Similar to our findings, earlier surveys of physicians and parents (Halfon et al., 2004; Sices et al., 2003) found that pediatricians practicing in urban settings with a high percentage of non-White children and children insured by Medicaid in their practice were significantly more likely to screen children for developmental delay. It is encouraging to know that children who are potentially at increased risk for developmental delays due to economic disadvantage are being screened, even though reasons for this are unclear. One reason may be the practice location of our urban respondents. For example, 68% of hospital-based and medical school clinics (30 of 44) are located in urban settings. It could be that the respondents in this study were located within academic centers with residency training programs where underserved children are cared for.
Of concern, however, is that suburban and rural nonminority children are being screened significantly less often for developmental delay at routine intervals. There may be at least three possible reasons for this. First, pediatricians may not be adequately reimbursed for screening. In this study, only 26% of pediatricians reported that changes in their screening practice over the past 5 years were due to improved reimbursement. This may suggest that reimbursement is still a barrier in more than half of the respondents in our sample who reportedly have not increased their use of standardized developmental screening measures. Second, in our sample, it was reported that clinicians implemented the significant proportion of the screening. This could create a burden of time and work within a practice that could be perceived as a disincentive to systematic screening implementation. Third, pediatricians could be more likely to normalize young children's delays when they are seeing patients from a similar demographic as their own.
In contrast to developmental screening, practices that care for children with 10%–30% of Medicaid-insured patients were significantly more likely to screen patients for autism than practices with more than 30% of patients. One possible explanation for this could be that parents with private insurance may have greater knowledge about and anxiety regarding autism because of their own educational level. As a result, these parents may ask more questions from their pediatricians about autism and request autism screening for their children. It is surprising that the younger physicians were not as compliant as pediatricians aged 51–60 years, who were significantly more likely to screen children for autism than younger pediatricians, as well as to screen children when concerns were raised. It is possible that older pediatricians were more likely to attend continuing medical education conferences or trainings. This finding does suggest a need to ensure that developmental and autism screening is being widely implemented in clinics where residents see patients.
This study has several limitations that may limit its generalizability. The most significant of these was the lower-than-anticipated response rate. Although the survey was brief, pediatricians get bombarded with surveys and electronic mails each day. We hoped that sending the survey through the executive directors of their state AAP chapter would increase completion rates due to name recognition of the sender. The final sample was also geographically limited and not necessarily representative of other locations in the United State. In addition, these finding may not represent accurately the behaviors of all practice types because most respondents were in private practice. The demographic characteristics of our sample are also not fully reflective of the demographics of U.S. pediatricians, and minority physicians are underrepresented in this sample. A final limitation is that this survey is based on pediatrician self-report, which may introduce bias. Nearly all studies to date regarding screening practice, however, are based on self-report. Bias would most likely trend toward overreporting of practice, and these data suggest that even if that is the case, there are still many pediatricians who are not compliant with AAP recommendations. Despite these limitations, this survey showed that even in a sample of pediatricians presumably self-selected for clinicians with an interest in developmental issues, screening practices still have ample room for improvement and there are disparities in screening practice.
Pediatricians have made significant gains in their rates of developmental and autism screening since the publication of relevant practice parameters. With alarmingly high rates reported for children with developmental disability and especially autism, it is critical that developmental and autism-specific screenings be implemented by primary care providers who continue to be the most consistent entry point for all children during their first years of life. This is even more important because indicators for autism are now found in earlier ages among children who may present with this diagnosis.
This study found that suburban, nonminority children not living in poverty are less likely to be routinely screened for developmental problems but more likely to be screened for autism. Additional research is needed to understand these practice variations and to identify barriers and facilitators to screening. Pediatricians and practices need to explore the most efficient and effective ways to screen children, including the implementation of more novel strategies that include online and electronic methods for parents who have access to computers or even smartphones. This study also suggested that pediatricians may assume the primary role in the implementation of screening, and additional efforts need to be devoted toward helping practices involve all personnel in various aspects of screening. Finally, reimbursement for screening is an area that should be examined in that improved or consistent funding could allow more pediatricians to allocate effort toward this crucial endeavor.
American Academy of Pediatrics Council on Children With Disabilities, Section on Developmental Behavioral Pediatrics, Bright Futures Steering Committee, & Medical Home Initiative for Children With Special Need. (2006). Identifying infants and young children with developmental disorders in the medical home: An algorithm for developmental surveillance and screening. Pediatrics, 118(1), 405–420. doi:10.1542/peds.2006-1231
Anderson L., Shinn C., Fullilove M., Scrimshaw S. C., Fielding J. E., Normand J. Task Force on Community Preventive Services (2003). The effectiveness of early childhood development programs: A systematic review. American Journal of Preventive Medicine, 24(3), 32–46. doi:10.1016/S0749-3797(02)00655-4
Bailey D. B. Jr., Hebbeler K., Spiker D., Scarborough A., Mallik S., Nelson L. (2005). Thirty-six-month outcomes for families of children who have disabilities and participated in early intervention. Pediatrics, 116(6), 1346–1352. doi:10.1542/peds.2004-1239
Baio J. (2012). Prevalence of autism spectrum disorders—Autism and Developmental Disabilities Monitoring Network, 14 sites, United States, 2008. MMWR Surveillance Summaries, 61(SS03), 1–19.
Bethell C., Reuland C., Schor E., Abrams M. K., Halfon N. (2011). Rates of parent-centered developmental screening: Disparities and links to services access. Pediatrics, 128(1), 146–155. doi:10.1542/peds.2010-0424
Boyle C., Boulet S., Schieve L., Cohen R. A., Blumberg S. J., Yeargin-Allsopp M., Kogan M. D. (2011). Trends in the prevalence of developmental disabilities in US children, 1997–2008. Pediatrics, 127(6), 1034–1042. doi:10.1542/peds.2010-2989
Dosreis S., Weiner C. L., Johnson L., Newschaffer C. J. (2006). Autism spectrum disorder screening and management practices among general pediatric providers. Journal of Developmental & Behavioral Pediatrics, 27(2, Suppl.), S88–S94. doi:10.1097/00004703-200604002-00006
Earls M. (2009). Expanding innovation through networks: The Assuring Better Child Health and Development (ABCD) Project. North Carolina Medical Journal, 70(3), 253–255.
Gills J. (2009). Screening practices of family physicians and pediatricians in 2 southern states. Infants & Young Children, 22(4), 321–331. doi:10.1097/IYC.0b013e3181bc4e21
Glascoe F. P., Byrne K. E., Ashford L. G., Johnson K. L., Chang B., Strickland B. (1992). Accuracy of the Denver-II in developmental screening. Pediatrics, 89(6, Pt 2), 1221–1225.
Halfon N., Regalado M., Sareen H., Inkelas M., Reuland C. H., Glascoe F. P. (2004). Assessing development in the pediatric office. Pediatrics, 113(6, Suppl.), 1926–1933.
Hill J. L., Brooks-Gunn J., Waldfogel J. (2003). Sustained effects of high participation in an early intervention for low-birth-weight premature infants. Developmental Psychology, 39(4), 730–744. doi:10.1037/0012-16126.96.36.1990
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. doi:10.1542/peds.2007-2361
Kim Y. S., Leventhal B. L., Koh Y. J., Fombonne E., Laska E., Lim Grinker R. R. (2011). Prevalence of autism spectrum disorders in a total population sample. American Journal of Psychiatry, 168(9), 904–912. doi:10.1176/appi.ajp.2011.10101532
King T. M., Tandon S. D., Macias M. M., Healy J. A., Duncan P. M., Swigonski N. L., Lipkin P. H. (2010). Implementing developmental screening and referrals: Lessons learned from a national project. Pediatrics, 125(2), 350–360. doi:10.1542/peds.2009-0388
Kogan M. D., Blumberg S. J., Schieve L. A., Boyle C. A., Perrin J. M., Ghandour R. M., van Dyck P. C. (2009). Prevalence of parent-reported diagnosis of autism spectrum disorder among children in the US, 2007. Pediatrics, 124(5), 1395–1403. doi:10.1542/peds.2009-1522
Radecki N., Sand N., O'Connor K. G., Sharp S., Olson L. (2011). Trends in the use of standardized tools for developmental screening in early childhood: 2002–2009. Pediatrics, 128, 14–19. doi:10.1542/peds.2010-2180
Rice C. (2009). Prevalence of autism spectrum disorders—Autism and Developmental Disabilities Monitoring Network Surveillance, United States, 2006. MMWR Surveillance Summaries, 58(10), 1–20.
Rosenberg S. A., Zhang D., Robinson C. C. (2008). Prevalence of developmental delays and participation in early intervention services for young children. Pediatrics, 121(6), e1503–e1509. doi:10.1542/peds.2007-1680
Sand N., Silverstein M., Glascoe F. P., Gupta V. B., Tonniges T. P., O'Connor K. G. (2005). Pediatricians' reported practices regarding developmental screening: Do guidelines work? Do they help? Pediatrics, 116(1), 174–179. doi:10.1542/peds.2004-1809
Schonwald A., Huntington N., Chan E., Risko W., Bridgemohan C. (2009). Routine developmental screening implemented in urban primary care settings: More evidence of feasibility and effectiveness. Pediatrics, 123(2), 660–668. doi:10.1542/peds.2007-2798
Shattuck P. T., Durkin M., Maenner M., Newschaffer C., Mandell D. S., Wiggins L., Cuniff C. (2009). Timing of identification among children with an autism spectrum disorder: Findings from a population-based surveillance study. Journal of the American Academy of Child & Adolescent Psychiatry, 48(5), 474–483. doi:10.1097/CHI.0b013e31819b3848
Sices L., Feudtner C., McLaughlin J., Drotar D., Williams M. (2003). How do primary care physicians identify young children with developmental delays? A national survey. Journal of Developmental & Behavioral Pediatrics, 24(6), 409–417. doi:10.1097/00004703-200312000-00002
Stevens G. D. (2006). Gradients in the health status and developmental risks of young children: The combined influences of multiple social risk factors. Maternal & Child Health Journal, 10(2), 187–199. doi:10.1007/s10995-005-0062-y
Wiggins L. D., Baio J., Rice C. (2006). Examination of the time between first evaluation and first autism spectrum diagnosis in a population-based sample. Journal of Developmental & Behavioral Pediatrics, 27(2, Suppl.), S79–S87. doi:10.1097/00004703-200604002-00005