Outcomes of autism spectrum disorder screening and follow-up program in a sample of Turkey


Borte Gurbuz Ozgura, Hande Aytacb        


aMugla Sitki Kocman University Training and Research Hospital, Child and Adolescent Psychiatry Clinic, Mugla; 

bMugla Provincial Directorate of Health, Mugla, Turkey




Background: The mean age of diagnosis of autism spectrum disorder (ASD) was not at the desired level yet. Early diagnosis is one of the most decisive factor on prognosis. Therefore, various screening programs are implemented in different countries to provide early diagnosis. ASD screening and follow-up programs are implemented in all provinces in Turkey as a part of the National Action Plan for Individuals with ASD. The purpose of the study is to assess the outcomes of patients referred in one year by primary health care professionals who received an autism training program.

Methods: Of the 18,678 children aged 18-36 months who were screened by family physicians with a 5-question screening scale for one year, 320 children who were found to be at risk for ASD were evaluated. In addition to psychiatric examination, developmental assessment and Childhood Autism Rating Scale results were reviewed retrospectively.

Results: Two hundred sixty-four out of 320 risky children were evaluated by the child and adolescent psychiatrist. Ninety-four were diagnosed with ASD. The most frequent diagnoses that follow were language disorder and intellectual disability. While 6.8% of the referred patients (n=18) were subjects who had received ASD diagnosis previously, 77 subjects received ASD diagnosis for the first time. It can be seen that as the markings in the screening form as risky increased, the rate of being diagnosed with ASD also increases. The positive predictive value of the 5-question screening form was 35.6%.

Conclusions: This program enables the diagnosis of ASD at an early age during community-based screening. We believe that the most important advantage of the program is that it can be implemented quickly and easily by family physicians.

Keywords: Autism spectrum disorders, early diagnosis, screening, family physician




Autism spectrum disorder (ASD) is a neuro-developmental disorder seen in the early childhood period, which is characterized by impairment in social communication and interaction and restrictive-repetitive behavior patterns [1]. Its diagnosis at early ages positively affects prognosis [2, 3]. Various screening scales have been developed on diagnosis in early ages [4-6]. The purpose of screening scales is to question ASD related symptoms and to identify the risk groups in different occupation groups in contact with early age groups. Therefore, a great responsibility falls on the shoulders of health workers on screening children who are under risk for ASD. Family physicians regularly examine children under risk for ASD in their developmental follow-up and vaccination periods. The American Academy of Pediatrics suggests screening children for ASD on the 18th and 24th months [7]. It is extremely important that children aged 18-36 months in contact with family physicians are screened for ASD in early periods. ASD knowledge level and awareness have, in general, been found low in professionals such as doctors, nurses, therapists, and educators who establish the initial contact with children diagnosed with ASD [8-13]. The low level of awareness, a small number of diagnoses, insufficient access to treatment, and the increasing prevalence of ASD have made this disorder a public health problem [14, 15]. Therefore, many countries today use screening programs intended to diagnose ASD in early age periods [4, 16, 17]. In our country, the national action plan on individuals diagnosed with ASD was published in the official gazette on 13.04.2016 [18]. In the related action plan, it is stated that the health services provided to individuals with ASD need to be carried out by the Ministry of Health, health workers need to be trained to create awareness and job definitions were made to implement the ASD screening program. The action plan involves the job definitions of different ministries for children diagnosed with ASD from diagnosis to treatment, education, and social rights and the targets planned to be reached after the completion of the action plan. Within the framework of this action plan, the Ministry of Health has developed the ASD screening and follow-up program. This program is being carried out by family physicians throughout the country. With the implementation of the action plan in 2016, family physicians and family health workers were given training on ASD screening and follow-up programs by the child and adolescent psychiatrists in all cities in Turkey. After these training, the ASD screening questions designed for children aged 18-36 months consulting family physicians started to be applied in a routine manner. It was targeted to refer children who were found at risk to the child and adolescent psychiatrists (CAMH) for definitive diagnosis. 

The autism screening and follow-up program training were given to family physicians and family health workers in the city of Mugla in 2018. The aim of this study is to assess the outcomes of patients referred to in a period of one year by primary health care professionals who received an autism training program.


The ethical approvals of the Mugla Provincial Directorate of Health and Mugla Sitki Kocman University, Human Studies Ethics Committee were obtained to carry out the study. The autism spectrum disorder screening and follow-up program were initiated in the city of Mugla in 2018. As the first step of the program, training related to ASD was organized for all family physicians and family health workers (by first and second authors). The content of the training included normal developmental period features, ASD definition, and clinical findings, and the functioning of the screening and follow-up program. All subjects guided within the scope of this study were health workers who received training on how the screening form should be filled. Family physicians carried out screenings for ASD at least once for children aged 18-36 months among the patients in their area. The family health workers (midwives, nurses, etc.) made sure that all children they found at risk to be examined by the family physicians and the referral forms were provided by the family physicians. They guided all children found to be at risk who were younger than 18 months or older than 36 months within the scope of the screening program as well. In addition, subjects who had been diagnosed with ASD and referred for check-up examinations were included in the study since the program also involves ASD follow-up. The arrangement of examinations for referred children by the CAMH specialist was carried out single-handedly by one person (second author). The population sample was selected as all children who screened by family physicians during the ongoing screening program for a period of one year in Mugla. Between the dates 01.05.2018-01.05.2019, the charts of all subjects who were found at risk and referred within the scope of the ASD screening and follow-up program by the family physicians in Mugla, Turkey were reviewed retrospectively. All subjects referred within a duration of one year were examined by the same CAMH specialist (first author). The children were evaluated in the accompaniment of their parents. Besides the sociodemographic information, the children’s developmental characteristics were also questioned. In the first examination, the CAMH specialist applied the Childhood Autism Rating Scale (CARS). Denver developmental screening test- II was applied for a developmental evaluation. The cognitive development levels were determined in line with the clinical examination. The criteria involving identification for risk included at least one risky answer for the 5-question screening form. The psychiatric diagnoses were put in accordance with DSM-5 [1].

The children who did not come for examination were identified as patients for whom at least two appointments were set, informed by the family physicians, but did not show up for the appointments. As for the period of follow-up, they were identified as subjects who showed up for check-up within a period of one year and subjects who showed up for check-up for a period of three months following the selection of patients for the study was completed. Patients with ASD, suspected ASD, other psychiatric diagnoses, or psychiatric symptoms were identified as follow-up criteria.

Evaluation tools

Autism spectrum disorder screening and follow-up program referral form

This is the form that contains the name, surname, birth date, and contact information of the referred patient and the 5-item ASD risk questions. The questions used in the screening form were prepared by the Ministry of Health on the basis of the Modified Checklist for Autism in Toddlers (M-CHAT) [19]. The questions in the screening form consisted of whether the child has a speech delay; has repetitive movements; responds to his/her name when called; establishes eye contact, and looks at something which is pointed at with a finger. The answers to these questions are marked by the primary health care workers through observation and asking questions to the families of the children. It was not regarded as sufficient to fill out the form only with the information received from the parents. If one of the five questions was identified as risky in the first step, the form became risky, and the child was referred to the CAMH specialist.

Childhood Autism Rating Scale (CARS)

This is a rating scale consisting of 15 items developed by Schopler et al. with the purpose of distinguishing children with mental deficiencies who are not autistic from children with autistic symptoms [20]. The 15 sub-headings of the scale consist of, Relationship to People; Imitation; Emotional Response; Body Use; Object Use; Adaptation to Change; Visual Response; Listening Response; Taste, Smell and Touch Responses and Use; Fear and Nervousness; Verbal Communication; Non-verbal Communication; Activity Level; Level and Consistency of Intellectual Response and General Impressions. Each item is graded between 1-4 with half point-scoring. The lowest score which can be received from the scale is 15, and the highest score is 60. The scale’s validity and reliability studies were performed by Gassaloğlu et al. The cutoff score of the scale was determined as 29.5. It is stated that as the score increases, the intensity of ASD increases as well [21].

Denver developmental screening test- II (DDST-II)

Denver developmental screening test- II was developed in 1967 by Frankenburg and Dodds, and it was reviewed, and DDST-II was created [22, 23]. The test’s standardization for Turkish was carried out by Yalaz et al. [24]. The test was developed with the purpose of observing the development of language, personal-social, fine motor-adaptive, and gross motor skills and identifying all children who are under risk developmentally. In our study, DDST-II developmental screening test was applied by certified early childhood specialists.

Statistical analysis

The SPSS 17.0 (Chicago, IL, USA) software was used for statistical analysis [25]. The suitability of the data for normal distribution was evaluated with the Shapiro-Wilk test. The data were defined as number, percentage, mean, standard deviation, median, min., max. And 25-75 percentile (Q1-Q3). The positive predictive value (PPV) was calculated as the subject who received accurate diagnosis/referred subjects (true positive+false positive). In the comparison of the two groups, an independent group t-test was used. The significance level was accepted as p<0.05.


For a duration of one year, a total of 18,678 children were screened in Mugla city centers and districts. 52.6% of the children who were screened were male (n=9743), and 47,83% were female (n=8935). Three hundred twenty of the children who were screened were identified as being at risk. Among the children identified as being at risk, 264 of them showed up for examination by CAMH specialist (82.5%), and 56 did not show up for the examination (17.5%) (Figure 1). 22.5% of all subjects (n=72) were female and 77.5% were male (n=248). The mean age was 40.72±20.73 (min 12, max 150 months). The sociodemographic and clinical characteristics of the children who showed up and did not show up for examination are given in Table 1. It was determined that 299 of the subjects were referred with the ASD screening and follow-up screening form, and only 21 subjects were referred without that form. 94 of the 264 were diagnosed with ASD (35.6%). While 6.8% of the referred patients (n=18) were subjects who had received ASD diagnosis previously, 77 subjects received ASD diagnosis for the first time. It was determined that one patient had lost diagnosis during the examination. Therefore, it was determined that 12 of the diagnosed patients were female, and 65 were male. It was determined that only 1 of the subjects out of 8 whose ASD diagnosis would be decided during the course of the follow-up period came for follow-up. The other diagnoses put on the patients are given in Table 2.

When the mean CARS total scores of the subjects diagnosed with ASD (28.06±5.79) and not diagnosed with ASD (16.13±1.60) were compared, the scores of the diagnosed with ASD were found statistically significantly higher (p<0.001). It was seen in the screening scale applied by the family physicians that as the number of items marked as risky increased in the screening forms of the subjects diagnosed with ASD, their CARS scores also increased (Figure 4).

It was seen in the history of 96 of the 264 subjects who showed up for examination were examined in different branches such pediatric or otorhinolaryngology specialists previously, and it was determined that 30 of the 77 subjects who were diagnosed for the first time with ASD (39%) were evaluated by a specialist other than a child psychiatrist but were not referred with the suspicion of ASD. In addition, 15 of the parents of the newly diagnosed children (19.5%) had suspected in the past that their children could have ASD.

When the most common referrals made by family physicians were analyzed, it was seen that 90.7% of the children had speech delay; 35.9% had repetitive behaviors; 31.1% did not establish eye contact, 25.3% did not have joint attention, and 24.1% did not receive a response to their names (Figure 2). It was determined that all 5 of the symptoms in the referral form of 31 subjects were positive, 2 of the subjects did not show up for examination, 26 were diagnosed with ASD, and one subject was identified as being under ASD risk. It can be seen that as the markings in the screening form as risky increased, the rate of being diagnosed with ASD also increases (Figure 3). The positive predictive value of the ASD screening and follow-up screening form was determined as 35.6%.


In our study, the results of the ASD screening and follow-up program’s at the end of one year were evaluated within the framework of the national action plan on individuals diagnosed with ASD. 320 children out of 18,678 who were detected as risky based on the 5-question screening form were referred by family physicians.

Considering the distribution of diagnoses children received in our study, autism, language disorder, and retardation in cognitive development were found, respectively. It was considered that since the speech delay item’s being marked as risky by itself is a reason for referral, the number of subjects who were diagnosed with language disorder was high. On the other hand, although the variety was seen in the distribution of diagnoses other than ASD, we believe that it is an added advantage for patients that these diagnoses were put in the early age period, and the required interventions were carried out by the CAMH specialist. The DDST-II results of only 7.2% of the subjects who were identified to be at risk and referred were found normal. It was seen that besides the subjects diagnosed with ASD or subjects with intellectual disability, subjects with symptoms such as speech deficiency cause the result of the DDST-II test to turn out abnormal. Therefore, the development test should be applied within certain intervals in the follow-ups. In terms of the other identified diagnoses, it was reported that 48 out of 3999 children screened in Sweden were diagnosed with ASD and 3 with language disorders similar to the results of our study [16]. It was reported that 18 of the 31,724 children aged between 14-15 months in the Netherlands were diagnosed with ASD, 18 were diagnosed with a language disorder, and 13 with mental retardation [17]. In our study, although the ratio of female to male children out of all screened children is close to each other, it was determined that children identified as risky were mostly male. This is consistent with the fact that diagnoses are more common in males. Therefore, it is thought that most children at risk in screening programs will be males.

Due to the fact that 94 out of 264 children identified and evaluated as risky were diagnosed with ASD, the PPV value of the screening scale consisting of 5 questions applied in our study was determined as 0.35. Similar values were found in ASD screening programs [4]. However, it needs to be taken into consideration that PPV values should be evaluated together with autism prevalence, and since the negative predictive value could not be calculated in our study, the specificity and sensitivity values of the screening test could not be calculated either. In our country, the ASD prevalence is not known. In addition, research has not been carried out within the scope of this study to determine the false negativity rates for those who were screened but whose results were negative. What is more, the PPV value we presented reflects the values obtained from only a single city of the autism screening program w, which is implemented in the whole country. We believe that a reevaluation to be made with the values to be obtained from different regions covering the whole country will be appropriate. As the number of risky symptoms in the screening form increases, the increase in the rate of ASD indicates the sensitivity of the survey (Figure 3). For instance, 89.6% of the subjects for whom all of the five questions were marked as risky were diagnosed with ASD. Among the screening programs implemented in Europe, there are programs with a high PPV value. Usually, multiple screening surveys are used together, or a single scale with a higher number of questions is used in these programs [4]. Meanwhile, the screening form used in Turkey within the scope of the current program has been derived from the questions in M-CHAT. The greatest advantages of the screening form are that it is practical to implement, and its implementation duration is short. Due to the work systems of primary care physicians in Turkey, the applicability of M-CHAT or similar screening scales as a whole is low. On the other hand, it is an expected result that the CARS scores of the children diagnosed with ASD are statistically significantly higher compared to the children who were not diagnosed with ASD. As the number of risky items increases in the screening scale, the increase which takes place in both the number of ASD diagnoses and CARS scores can give an idea about the sensitivity of the questions of the scale.

In our study, it was determined that only 19.5% of the parents of children newly diagnosed with ASD had noticed symptoms which could be related with ASD in their children and that 39% of the children were evaluated by a specialist other than a CAMH specialist, but these children were not referred with the suspicion of ASD. In our study, the screening programs as expected allow the child to be diagnosed even if the parents do not notice an ASD related symptom. In addition, it can be made possible for children to be referred with the suspicion of ASD by other branches who have contact with the children in early ages by increasing awareness of the other specialized fields on ASD. The current screening and follow-up program is only being carried out with primary health care workers. In the literature, there are studies that show that the ASD knowledge and rate of referral with the suspicion of ASD by different branches such as pediatricians, pediatric nurses, and otorhinolaryngology assistants is low [26-29].

The mean age of the children who were identified as risky in our study and were diagnosed with ASD (n=77) for the first time was determined as 34.53±14.82 months. In the literature, it is reported that the mean age for ASD diagnosis is 4-6 years [30-32]. The most important advantage of the screening test is that the risky child can be identified at an early age by a health professional and his/her contribution to early diagnosis. The necessity for the implementation of the screening programs arose from the fact that the age of diagnosis has fallen. In addition, it was also underlined that children under risk need to be screened periodically with screening scales unique to ASD [33]. Insomuch that it was suggested that the siblings of children diagnosed with ASD should also be examined due to increased risk [34].

The person giving the ASD screening and follow-up training and evaluating the patients being the same doctor can be interpreted as the limitation of the study.

It was seen that primary health care providers such as family physicians, midwives, and nurses who received training on ASD screening and follow-up programs learned how to apply the screening form, question the symptoms, and refer the risky children in a routine manner. The family physicians of the diagnosed children having been given feedback about the diagnoses also come to the fore as a process that solidifies learning. The screening form containing a small number of questions and established chain of referral of the children from the first step to the child and adolescent psychiatry is considered to be effective in the sustainability of the program as well. Based on the findings of our study, it is considered that the screening scale used in this program is productive in terms of time-benefit balance. To overcome the negative aspects of screening scales such as long application time and application difficulty used in different countries, using our program and scale after cultural adaptations may be an option. We believe that the most important contribution of the program is to make it possible to diagnose children with ASD at an early age.


We would like to express our sincere gratitude to primary health care workers who helped us make this study possible. In addition, we would like to thank Nahit Motavalli Mukaddes and Onur Burak Dursun, who contributed to the development of this screening program. We are grateful that Zumran Ozdemir and Cigdem Ekinci applied development tests.




[1]       American Psychiatric Association. DSM-5 Task Force. Diagnostic and statistical manual of mental disorders: DSM-5. 5th ed. Washington, D.C.: American Psychiatric Association; 2013. p. 50-51.

[2]       Dawson G. Early behavioral intervention, brain plasticity, and the prevention of autism spectrum disorder. Dev Psychopathol. 2008;20(3):775-803.

[3]       Reichow B, Hume K, Barton EE, Boyd BA. Early intensive behavioral intervention (EIBI) for young children with autism spectrum disorders (ASD). Cochrane Database Syst Rev. 2018;5:CD009260.

[4]       Garcia-Primo P, Hellendoorn A, Charman T, Roeyers H, Dereu M, Roge B, et al. Screening for autism spectrum disorders: state of the art in Europe. Eur Child Adolesc Psychiatry. 2014;23(11):1005-10021.

[5]       Kılınç Ç, Bağlama B, Akçamete G. Investigation of the measurement tools used in the diagnosis and evaluation of autism spectrum disorder in early childhood. Cyprus Turkish Journal of Psychiatry and Psychology 2019;1(3):200-205.

[6]       Towle PO, Patrick PA. Autism spectrum disorder screening instruments for very young children: A systematic review. Autism Res Treat. 2016;2016:4624829.

[7]       Myers SM, Johnson CP, American Academy of Pediatrics Council on Children With D. Management of children with autism spectrum disorders. Pediatrics. 2007;120(5):1162-1182.

[8]       Bakare MO, Ebigbo PO, Agomoh AO, Menkiti NC. Knowledge about Childhood Autism among Health Workers (KCAHW) questionnaire: description, reliability and internal consistency. Clin Pract Epidemiol Ment Health. 2008;4:17.

[9]       Altay MA. Family physicians’ Awareness of autism spectrum disorder: Results from a survey study. Open Access Maced J Med Sci. 2019;7(6):967-972.

[10]     Sabuncuoglu M, Cebeci S, Rahbar MH, Hessabi M. Autism spectrum disorder and attention deficit hyperactivity disorder: Knowledge and attitude of family medicine residents in Turkey Turkish Journal of Family Medicine and Primary Care. 2015;9(2):46-53.

[11]     Salama HM. Assessment of family physicians’ knowledge of childhood autism. Family Medicine and Community Health. 2017;5(4):266–274.

[12]     Hacıibrahimoğlu B, Ustaoğlu A. An investigation of preservice preschool teachers’ knowledge and attitudes toward children with autism spectrum disorders. Inonu University Journal of the Faculty of Education. 2019;20(2):491-506.

[13]     Er-Sabuncuoğlu DM. Autism knowledge; preschool educators. HÜ Sağlık Bilimleri Fakültesi Dergisi 2016;3(1):28-52.

[14]     Hidiroglu S, Luleci NE, Karavus M, Tanriover O, Bayar ES, Karavus A. The awareness of childhood autism among residents of neuropsychiatric and other disciplines of a research and training hospital in Istanbul, Turkey. J Pak Med Assoc. 2018;68(2):247-251.

[15]     Hartley-McAndrew M, Doody KR, Mertz J. Knowledge of autism spectrum disorders in potential first-contact professionals. N Am J Med Sci. 2014;7(3):97-102.

[16]     Nygren G, Sandberg E, Gillstedt F, Ekeroth G, Arvidsson T, Gillberg C. A new screening programme for autism in a general population of Swedish toddlers. Res Dev Disabil. 2012;33(4):1200-1210.

[17]     Dietz C, Swinkels S, van Daalen E, van Engeland H, Buitelaar JK. Screening for autistic spectrum disorder in children aged 14-15 months. II: population screening with the Early Screening of Autistic Traits Questionnaire (ESAT). Design and general findings. J Autism Dev Disord. 2006;36(6):713-722.

[18]     Resmi Gazete TC. Otizm spektrum bozukluğu olan bireylere yönelik ulusal eylem planı (2016-2019). Ankara: Başbakanlık Basımevi; 13.4.2016.

[19]     Robins DL, Dumont-Mathieu TM. Early screening for autism spectrum disorders: update on the modified checklist for autism in toddlers and other measures. J Dev Behav Pediatr. 2006;27(2 Suppl):S111-119.

[20]     Schopler E, Reichler RJ, DeVellis RF, Daly K. Toward objective classification of childhood autism: Childhood Autism Rating Scale (CARS). J Autism Dev Disord. 1980;10(1):91-103.

[21]     İncekaş-Gassaloğlu S, Baykara B, Avcil S, Demiral Y. Validity and reliability analysis of Turkish version of Childhood Autism Rating Scale. Türk Psikiyatri Dergisi. 2016;27(4):266-274.

[22]     Frankenburg WK, Dodds J, Archer P, Shapiro H, Bresnick B. The Denver II: a major revision and restandardization of the Denver Developmental Screening Test. Pediatrics. 1992;89(1):91-97.

[23]     Frankenburg WK, Dodds JB. The Denver developmental screening test. J Pediatr. 1967;71(2):181-191.

[24]     Yalaz K, Anlar B, Bayoğlu B. Denver II gelişimsel tarama testi, ‟Türkiye standardizasyonu”. Ankara: Gelişimsel Çocuk Nöroloji Derneği; 2011.

[25]     SPSS I. SPSS Statistics for Windows. Version 17.0 ed. Chicago: SPSS Inc.: SPSS Inc.; 2011. p. released 2008.

[26]     Karavuş M, Karaketir ŞG, Hıdıroğlu S, Lüleci NE, Çetinkaya CY, Hoşgör EB, et al. A Cross-sectional study evaluating childhood autism awareness of residents otorhinolaryngology department in İstanbul. İstanbul Med J. 2019;20(5):450-7.

[27]     Erden G, Akçakın M, Gümüş-Doğan D, Öztürk-Ertem İ. Pediatricians and autism: Difficulties in diagnosis. Turkiye Klinikleri J Pediatr. 2010;19(1):9-15.

[28]     Eseigbe EE, Nuhu FT, Sheikh TL, Eseigbe P, Sanni KA, Olisah VO. Knowledge of childhood autism and challenges of management among medical doctors in Kaduna state, Northwest Nigeria. Autism Res Treat. 2015;2015:892301.

[29]     Sampson WG, Sandra AE. Comparative study on knowledge about autism spectrum disorder among paediatric and psychiatric nurses in public hospitals in Kumasi, Ghana. Clin Pract Epidemiol Ment Health. 2018;14:99-108.

[30]     Shattuck PT, Durkin M, Maenner M, Newschaffer C, Mandell DS, Wiggins L, et al. Timing of identification among children with an autism spectrum disorder: findings from a population-based surveillance study. J Am Acad Child Adolesc Psychiatry. 2009;48(5):474-483.

[31]     Valicenti-McDermott M, Hottinger K, Seijo R, Shulman L. Age at diagnosis of autism spectrum disorders. J Pediatr. 2012;161(3):554-556.

[32]     Kurasawa S, Tateyama K, Iwanaga R, Ohtoshi T, Nakatani K, Yokoi K. The Age at diagnosis of autism spectrum disorder in children in Japan. Int J Pediatr. 2018;2018:5374725.

[33]     Zwaigenbaum L, Bauman ML, Fein D, Pierce K, Buie T, Davis PA, et al. Early screening of autism spectrum disorder: Recommendations for practice and research. Pediatrics. 2015;136 Suppl 1:S41-59.

[34]     Zwaigenbaum L, Bauman ML, Choueiri R, Fein D, Kasari C, Pierce K, et al. Early Identification and interventions for autism spectrum disorder: Executive summary. Pediatrics. 2015;136 Suppl 1:S1-9.