Autism Spectrum Disorders in Mali: Parental concerns about child developmental milestones and late diagnostic age

Background In Mali, the health facility-based prevalence of autism spectrum disorders (ASD) was 4.5% (105/2,343) and the age of first medical visit for autistic children was around 7 years old in 2018. Parental recognition of developmental abnormalities initiates the early autism detection and diagnosis. Our aim was to investigate if parental concerns were associated with early first medical visits and subsequent ASD diagnostic in Mali. Methods We conducted a pilot study from November 2017 to July 2019. We surveyed parents of 57 autistic children aged 3-14 years old. Results Parents were concerned over verbal communication in 54.1% and reciprocal social interaction in 43.8%. Children with ASD had their first medical visit after 18 months old in 66.7%, ≥ two medical visits in 87.7%, and were identified after 36 months old in 76.8%. Conclusion: Parental concerns were not significantly associated with early first medical visit and ASD identification. A full scale study will be conducted for more consistent results. To determine the factors associated with a delayed first medical consultation for autistic children in our cohort, we used chi square or Fischer exact test to compare the proportions (before and after a cutoff age of 18 months old for early ASD risk screening and 36 months old for early ASD diagnosis) of the presence of a targeted factor (male gender, Bambara ethnicity, impaired reciprocal social interaction, impaired verbal communication, repetitive behaviors, other disorders). Other disorders represented the presence of epilepsy, presence of strabismus and the absence of crying at birth. P -value < 0.05 was considered statistically significant.


Background
Autism spectrum disorders (ASD) affects 1 in 59 children aged eight years old in The U.S [1]. ASD is recognized as a global public health concern, yet almost everything we know about ASD comes from high-income countries. Prevalence of ASD is still unknown in sub-Sahara Africa [2]. In Mali, the Health facility-based prevalence of ASD was 4.5% (105/2,343) in 2018 [3].
Early ASD diagnosis is very important, because early intensive treatment greatly improves its prognosis [4]. Early intervention in ASD requires early detection in which screening is crucial. Various early autism screening tools have been developed in the U.S to promote early autism risk detection [5]. In developed countries, the modified checklist for autism in toddlers-Revised/follow up (M-CHAT-R/F) and the social communication questionnaire (SCQ) are available online to parents who suspect their child at risk of ASD [6]. In Mali, we have validated the M-CHAT-R and SCQ in 2017-2018 [7]. Timely ASD diagnosis requires early detection, which greatly depends on parental recognition of developmental abnormalities [8]. Many studies were conducted to investigate how parental concerns of childhood developmental abnormalities associate with early or late ASD screening and diagnosis as well as the overall well-being of autistic children. In general, parents are the first to bring up concerns about their autistic child's development in up to 80% [9]. Developmental motor delays during early childhood may represent an important predictor for ASD diagnosis [10,11]. In high risk children, parentreported concerns can improve earlier recognition of ASD [12]. However, research has demonstrated that socioeconomic ranking of the parents correlate with the specific of the parental concerns in children suspected with ASD. For instance, parents with high socio-economic status were 8.5 times more concerned about the children being at risk of ASD as compared to those with low social economic status [13].
Mali is one of the poorest countries in the world, where most women struggle for their daily living [14]. The literacy rate is also one of the lowest in the world (33.1%) in people aged 15 years old and above [15]. In addition, early child visits and autism screening are neither mandatory nor systematic in the Malian health system. After the validation of the M-CHAT-R/F and SCQ in Mali, we wondered how effective these tools are to the early ASD screening and diagnosis in our socio-cultural setting. We hypothesized that parental concerns on childhood developmental abnormalities could guide our ASD awareness campaigns and promote the use of M-CHAT-R and SCQ for early autism risk screening in Mali. This original work was aimed to investigate the association between the child age at the first medical visit, as an indicator of the age of ASD diagnosis, and the parental concerns about their children's development abnormalities in Mali.

Methods
We conducted a descriptive cross-sectional pilot study from November 2017 to July 2019 in Bamako. In this study, we included 57 autistic children aged 3-14 years old who were already enrolled into our autism genetic family study in Bamako. Our study was conducted in three phases. First, potential study participants were screened using the social communication questionnaire (SCQ) at a private medical clinic "Kaidara" downtown Bamako, the capital city and the cutoff score of > 15 was used as the main inclusion criterium. Then, screened potential study participants underwent a thorough medical visit by a multidisciplinary team (a senior clinical psychologist, a senior child psychiatrist, a neuro-pediatrician, a neurologist, a resident in psychiatry and a resident in pediatrics). Those who did not have a brain CT scan at inclusion (n = 40) underwent brain CT scan by an experienced neuroradiologist. All the potential study participants did eye and ear exams by a senior eye, nose & throat (ENT) specialist and a certified junior ophthalmologist to exclude any medical condition, which could explain the child's deficit in social interaction. Finally, study participants were selected based on presence of the signs and symptoms of autism spectrum disorders (dyad of ASD in DSM-V) and the normal brain CT Scan exam (or at least the absence of a structural cerebral lesion that can explain the child's condition). We interviewed parents of our study participants during home visits to administer our study questionnaire (see supplemental material). After informed consent, we collected information on the age of developmental milestones of their autistic children. Autistic children and their parents were compensated for their study participation at an approved rate by the Malian ethical committee at the Faculty of Medicine, Pharmacy and Odontostomatology (FMPOS), Bamako, Mali. Our study protocol, consent form and questionnaire were approved by the ethical committee at the FMPOS. Completed questionnaires were coded without any identifiable personal information. In the database created, data were analyzed with SPSS version 25. We determined the median and mean age for each childhood developmental milestone with a standard deviation and the extreme ages (minimum and maximum). To diagnose ASD, we used the criteria of DSM-IV [16]. To determine the factors associated with a delayed first medical consultation for autistic children in our cohort, we used chi square or Fischer exact test to compare the proportions (before and after a cutoff age of 18 months old for early ASD risk screening and 36 months old for early ASD diagnosis) of the presence of a targeted factor (male gender, Bambara ethnicity, impaired reciprocal social interaction, impaired verbal communication, repetitive behaviors, other disorders). Other disorders represented the presence of epilepsy, presence of strabismus and the absence of crying at birth. P -value < 0.05 was considered statistically significant.

Results
Among 57 surveyed autistic children aged 3-14 years old, the sex-ratio was 1.5 in favor of male. The first medical visit was after the age of 18 months old in 66.7%. The diagnosis of ASD was made after 36 months old in 76.8%. Autistic children had at least two medical visits before the ASD diagnosis in 87.7% (Table 1).  The median age of imitation was 24 months old with the extremes of 4 and 168 months old (Table 3). To hold his/her head 6.00 8.9 ± 8.0  To reach for an object 7.00 16.1 ± 23.8  To move in bed by rotation 7.00 11.1 ± 10.7  To make pedaling movements 6.00 10.5 ± 9.6  To sit with support 6.00 12 ± 16.8  To turn around on his/her back 8. 50 14.3 ± 14.6  To put his/her feet in his/her mouth 7.50 14.7 ± 16.3  To sit without support 7.00 11.7 ± 10.7  To know how to stretch his/her arms 7.00 13.6 ± 13.9  To sit from flat back position 9.00 17. To button his/her cloth 60.00 64.8 ± 41.5  None of our targeted factors was significantly associated with a delayed age of first medical visit at the 18 months old (Table 4) and 36 months old time points (Table 5) p > 0.05.

Discussion
Children aged 18 to 24 months are screened for ASD to assist in early detection, consistent with current American Academy of Pediatrics' recommendations [17]. Despite the advances in ASD screening and evaluation, the mean age of diagnosis is still 4-5 years [18]. The age of ASD diagnosis depends on many factors and to label a child "autistic" may require a multidisciplinary team [19] or multiple medical visits (4-5 on average with the extremes of 1 and 29) [20]. While the high demand of services due to a recent increase in ASD prevalence delays its diagnosis in developed countries, the lack of trained health professionals and the socio-cultural representation of the disorder may prevail in Mali. Similar to sub-Sahara Africa, in the Arab world, culture may significantly influence the age of noticing abnormality and the ways of investigating and treating autism [21].
The age of ASD identification was after 36 months old in 76.8% (Table 1). In two separate U.S national surveys, the 2011-2012 National Survey of Children's Health (n = 95,677) and the 2009-2010 National Survey of Children with Special Health Care Needs (n = 371,617), many parents of children with ASD reported identification after 3 years old with 1/3 to 1/2 of cases after 6 years old [21].
In our cohort, parental concerns on their children's development were mostly about verbal communication (54.1%) and reciprocal social interaction (43.8%) ( Table 2). This result corroborates the findings of Richards et al. They reported that over 90% (n = 532) of the parents brought up concerns during well child visits and 78.6% were about speech and verbal communication [22]. Children whose parents expressed concerns about their child's verbal communication experienced earlier ages for all outcomes when compared to children of parents who did not have verbal communication concerns [23]. Such finding highlights the evolution of parental concerns over time. In mid-2000s, parents of autistic children used to be concerned over either a delayed diagnosis of physical disabilities, such as hearing impairment, cerebral palsy [24], learning difficulty, being bullied, stress-coping, or achievement [25].
The median age of being able to smile (social and non-social) for the first time was 4 months old and the mean age was 7.7 months old ± 7.9 with the extremes of 2 and 36 months old (Table 3). Anticipatory smiling levels in the first year may predict ASD diagnosis or continuous ASD severity outcomes [26]. The median age of imitation was 24 months and the average age of 45.23 months ± 44.8 with the extremes of 4 and 168 months old (Table 3). Immediate and deferred imitations of adults by children with ASD were strongly associated with language ability at age 3-4 years and communication development from age 4 to 6.5 years [27], but the loss of social communication skills is highly variable (its rate, timing and severity) in ASD and it usually occurs in the period between 9 and 24 months [28,29].
The median age of storytelling in our cohort was 49 months and the mean age was 50.4 months ± 21.5 with the extremes of 36 and 99 months (Table 3). Difficulties in storytelling have been reported in children with ASD [30,31]. While the contribution of problematic use of subject pronoun has been demonstrated in the difficulties in storytelling in school-aged children with ASD [32], speculation has been on whether or not difficulties in orientation in time and space play an important role in the process [33].
Motor development is critical to the overall development of the child. In our cohort, the age of motor skill acquisition was very variable. For instance, the mean age of reaching out for an object was 16.1 months ± 23.8 with the extremes 3 and 144 months ( Table 3). The gross and fine motor skills of young children with ASD are delayed and become progressively more delayed with age [34,35], but the great variability in motor skill acquisition makes it challenging the universality for health professionals [36]. The median age of walking (AOW) in our cohort was 14.50 months and the mean age was 19.5 months ± 44.8 with the extremes of 7 and 72 months (Table 3). Reindal et al. reported a mean AOW of 15.3 months ± 5.5 in children with ASD versus 14.1 months ± 3.4. AOW is reported to be later in autism spectrum disorder (ASD) compared with typical normal development [37].
None of our targeted factors (male gender, Bambara ethnicity, impaired reciprocal social interaction, impaired verbal communication, repetitive behaviors, the presence of epilepsy, presence of strabismus, and the absence of crying at birth) was significantly associated with the age of first medical visit (the cutoff was 18 months) or the age of identification of autistic disorder (the cutoff was 36 months (Tables 4-5). This finding might have resulted from the small size of our sample. Otherwise, evidence from the literature are compelling for the association of some (if not all) of these studied factors with early detection and diagnosis of autism spectrum disorders.
Parental concerns are more pronounced in males and females with ASD. This may have many plausible explanations. First, even with higher ASD prevalence in both sexes, core ASD symptomology are easily presentable in males, and females are too good in camouflaging their symptoms, which may result in sex differences in parental concerns after age 5 years [38][39][40][41][42][43]. Second, females experienced less unusual stereotyped and repetitive behaviors than males due to genetic (Y chromosome) and hormonal factors (fetal testosterone) and they have an increased prevalence of internalizing problems [40,44]. Third, females were found to receive lower scores than males particularly on modules 2 and 3 of the Childhood Autism Rating Scale scores (CARS) [45]. Finally, Baron-Cohen's extreme male theory may induce a stronger examiner bias toward males than females with ASD [46].
Black parents reported significantly fewer autism concerns and fewer social and restricted and repetitive behavior concerns as compared to White parents [47]. The Bambara are the largest ethnic group in Mali with up to 36.5% of the total population. They also present in significant percentages in Guinea Conakry, Burkina Faso, Niger, Ivory Coast, and Mauritania [48]. In addition, the other disorders in Tables 3-5 (epilepsy, strabismus, and absence of crying at birth) are very frequent in autism [49][50][51]. The absence of crying at birth and seizure episode in childhood were highly associated with autism risk in Brazil (OR 5.75; 95%CI 3.37-9.81) [5]. Even though, strabismus has a low prevalence in Africa [52], it has a strong cultural representation in the West African society in general and in Mali in particular. It is referred to as "the hen is looking at the cloudy sky" in Mali and it is a distinctive and particular trait which, a child can be labeled with among his/her peer.
Which one of the impaired reciprocal social interaction, afflicted verbal communication and repetitive behaviors in children with ASD draws first the attention of parents? The answer to this question depends on multiple factors. Truly, ASD is a spectrum in the real sense of the word. When all domains of the child's development are affected moderately or severely at the same time, parents usually notice as early as possible, especially when they have another child with normal developmental for comparison [53]. When language skills lag far behind, parental concerns habitually raise between 18 and 24 months old of age [22,53].
Altogether, assessing parental concerns about children with ASD may be more challenging and more complex than one may initially think. Core symptoms of ASD alone may not explain at which extent parents are concerned and stressed out their child development and well-being. A more comprehensive assessment should consider other aspects such as sleep and eating problems, parenting stress, the specific burden on mothers with subsequent poor marital relationship [46,[54][55][56].

Conclusion
After this pilot study, we surveyed parents of 57 children with ASD aged 3-14 years. Parental concerns on their children's development were mostly about verbal communication (54.1%) and reciprocal social interaction (43.8%). The first medical visit was after the age of 18 months old in 66.7%; autistic children had at least two medical visits before the ASD diagnosis in 87.7%. The age of ASD identification was after 36 months old in 76.8%. No investigated factor was associated with the age of first medical visit and identification of ASD due to the small size of our sample. In the future, we will conceive an online version of our study questionnaire for a larger online survey across West Africa. • Our study protocol, consent form and questionnaire were approved by the ethical committee at the FMPOS. "All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee (include name of committee + reference number) and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards." Informed consent was obtained from all individual participants in the study.

ASD
• Consent for publication: • "Not applicable" • Availability of data and materials: • The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Competing interests:
None. The authors declare that they have no conflict of interest.