Differential Diagnosis of Autism and Mental Retardation: Issues and Possible Solutions

Abstract

Co-morbidity of Autism and Mental Retardation (MR) makes it difficult to differentiate the two disorders. The current study was aimed at identifying the key features that separate the two conditions and those which create confusion. To explore the phenomenology, ten parents of children with each disorder were interviewed on 9 major domains (social, communication, stereotypical movements, cognitions, self care, behavioural problems, interests/routines/sensory issues, neurotic traits and developmental milestones). After validation from 10 experts, the final list of 66 features was administered on a sample of 40 children with Autism and MR (20 each). Mann Whitney U test found that 35 features were significant in differentiating the two disorders, whereas 31 features were not statistically significant. The findings were discussed within diagnostic relevance and suggestions were made to improve the efficiency of diagnosis by utilizing the dubious features appropriately. [PUBLICATION ABSTRACT]

Headnote

Co-morbidity of Autism and Mental Retardation (MR) makes it difficult to differentiate the two disorders. The current study was aimed at identifying the key features that separate the two conditions and those which create confusion. To explore the phenomenology, ten parents of children with each disorder were interviewed on 9 major domains (social, communication, stereotypical movements, cognitions, self care, behavioural problems, interests/routines/sensory issues, neurotic traits and developmental milestones). After validation from 10 experts, the final list of 66 features was administered on a sample of 40 children with Autism and MR (20 each). Mann Whitney U test found that 35 features were significant in differentiating the two disorders, whereas 31 features were not statistically significant. The findings were discussed within diagnostic relevance and suggestions were made to improve the efficiency of diagnosis by utilizing the dubious features appropriately.

Keywords: Autism, Autistic Spectrum Disorder, Mental Retardation, Differential Diagnosis,

Autism is known as a complex disorder and to coexist with MR (American Psychiatric Association, APA, 2000; Kraijer, 2006). Research in recent years has questioned this assumption saying that the problem might not be with the ability of the child but with instruments and certain difficulties that children with autism face (Edelson, 2006). Being disorders of development, there are few similarities between the two disorders that pose problem in their differntial diagnosis. For expamle, Skuse (2007) highlighted the possiblity that association between MR and autism might not be because of their common etiology, but because the presence of both will make it more likely to be diagnosed together (perhaps due to the convenience and confusion in their symptoms). Vig and Jedrysek (1999) argued that in severe and profound MR, the symptoms become more identical with that of Autism, posing a great problem in the differential diagnosis between the two disorders. Although the Diagnostic and Statistical Manual Fourth Edition Text Revised (DSM-IV-TR) emphaíses on the presence of stereotyped behaviour as the distinguishing feature between Autism and MR (APA, 2000). But stereotped behaviours were also found in persons with and without Autism, which made it difficult to decide if it was an indication of the presence of autistic disorder or due to low level of functioning (Bodfish et al. 1995). Likewise language and speech difficulties, present in Autism might not be present in MR giving rise to the issues of diagnostic overshadowing and diagnostic under-representation (Kraijer & Bildt, 2005).

Differentiating between the two cooccurring disorders is important as it has implications for the management of Autism with and without MR. Verheij and Doom (1990) as well as Society for the Autistically Handicapped (STAFH , 2011) mentioned that when Autism and MR co-occur, the therapeutic environment must take full account of both the disorders.

Early identification of Autism is crucial and has great effects on the progranosis of the child (Dawson & Osterling, 1997; Howlin, 1998; Rebecca, 2007; Rogers, 1996, 1998). The confusion in differnetial diagnosis delays the correct identification of the problem, thus wasting the most crucial time of the child's life. Volker and Lopata (2008) declared that the diagnostic complexity and heterogeneity of austim were a challange in school placements of such children, posing a problem in their early management.

Whereas the problem of differential diagnosis between the two disorders is crucial but it is not new. A number of tests are available to assess autism features in children and adults such as Childhood Autism Rating Scale - CARS (Schopler, Reichler & DeVellis, 1980), Pervasive Developmental Disorders in Mental Retardation Scale -PDD-MRS (Kraijer, 1990; Kraijer & Bildt, 2005) and Social Communication Questionnaire -SCQ (Rutter, 2003). Let us have a close look at some of them.

Childhood Autism Rating Scale (Schopler et al, 1980) is the most widely used standardized instrument specifically designed to aid in the diagnosis of Autistic Spectrum Disorder (ASD) for use with children as young as 2 years of age. The 15 item rating scale covers different aspects of behaviours of a child with ASD. It effectively discriminates between persons with autism and those who are trainable mentally retarded (Morgan, 1988; Teal & Wiebe, 1986). According to the New York Department of Health, Early Intervention Program (2005) declared CARS to possess an acceptable combination of practicality and research support. However the report also highlighted its limits in possibility of getting false positive (incorrectly identify autism in children with severe MR) or false negative (may not identify milder presentations of autism) test results. The report further claimed that CARS may not fully reflect current information about cognitive and social development in young children.

The PDD-MRS (Kraijer & Bildt, 2005) assesses the presence of Autism in persons with MR between the age-range of 2-70 years. However it covers the whole range of Pervasive Developmental Disorder.

Social Communication Questionnaire (Rutter, 2003) is another instrument that screens the children who need further assessment for autism spectrum disorder. It was previously known as the 'Autism Screening Questionnaire1. This brief instrument can be used for children over 4 years of age (if child's mental age is above 2 years). Although it is suitable for screening and monitoring purposes, it is not suitable for diagnosis as it does not provide information about the onset, course, and pervasiveness of the symptoms across different contexts (Hogrefe, 2010). Some of the other studies (Eaves, Wingert, Helena & Mickelson, 2006; Oosterling et al, 2010) reported that SCQ was likely to have a number of false positive cases.

Relatively little is known about the the presentation of persons with autism in developing countries (Salehi, Al-Hifthy & Ghaziuddin, 2009). In a country like Pakistan, where history of autism awareness is only about a decade old and only a couple of published articles are available, that too related to the incidence of autism in special education institutes (Suhail & Zafar, 2008) and behaviour management of such children (Ashfaq, Saeed & Jahangir, 2009).

The objectives of the current study were to identify the features that differentiate Autism from MR and those that might be found in both and create diagnostic confusion.

It was hoped that identification of such features will lessen the chances of misdiagnosis of the children especially those of Autism thus enhancing the person's prognosis and possibility of correct management plan. It also aimed at identifying the possible issues in the diagnosis. Since no such work was available in Pakistan, so it was wise to start from the scratch.

METHOD

The study was conducted in three phases:

Exploring Phenomenology- The aim of this phase was to generate features of Autism and MR from the parents of persons firmly diagnosed with Autism and MR.

Participants and Procedure

The purposive sample comprised of 20 firmly diagnosed children (10 with Autism and 10 with MR) with the age range of 3 years onwards.

In order to generate features from the parents of diagnosed persons with Autism and MR (which also included Down's Syndrome), phase-1 was carried out.

Principals of four Special Education Institutions of Lahore were contacted. They were explained about the purpose of the research and confidentiality measures. Twenty students were selected who were firmly diagnosed by clinical psychologist or psychiatrist (the two professionals usually involved in diagnosing such children in Pakistan) with Autism and MR based on criteria from DSM-IV-TR (APA, 2000). The parents were contacted through schools and after taking the consent from parents, they were interviewed in depth on a semi-structured questionnaire on 9 major domains (social, communication, stereotypical movements, cognitions, self care, behavioural problems, interests/routines/sensory issues, neurotic traits, and developmental milestones). The questionnaire contained open ended questions studying the developmental and behavioural aspects of the two disorders. Each parent (mostly both parents or in few cases mothers only) was interviewed individually and their responses were recorded in verbatim. Persons with multiple disorders were excluded. This elicited a list of 92 features. In order to validate the symptoms, the phase 2 was carried out.

Empirical Validation

Participants and Procedure

In the phase 2, the list of 92 features was validated through experts' ratings. Ten experts (six clinical psychologists and four psychiatrists) with a minimum four years of experience in working with persons with Autism and MR were selected. The experts were asked to rate each feature indicating the degree to which it denotes Autism or MR. A five point scale (0-4) was used in which O meant "Not at AH" and 4 denoted "Very much." Based on the expert's agreement, the features that were identified as belonging to Autism or MR by less than 30% of the experts were removed from the list. Also the features which overlapped each other in both the lists were removed, resulting in a list of 68 final features. This also reduced the major domains to 8 (neurotic traits was removed). The features related to the child's milestones were given three options i.e. normal, delayed and don't remember. The upper age limit was taken into account for each milestone (based on the literature for normal development). Whereas for the behavioural features, a four point scale (0-3) was used was used. The options included "Not at All (O)", "A Little Bit (I)", "Somewhat True (2)" and "Very Much True (3)." The four point scale gave more freedom to the respondents in deciding about the deviation of the person's behaviour from the normal child of the same age in terms of frequency, intensity, and peculiarity. Since we were talking about the differential diagnosis, it was wise to see the degree rather than just yes or no options. The features were worded to assess the positive symptoms of the disorder. The final list of 68 features was used to collect data for main study at phase - III.

Main Study

The purpose of this phase was to assess the frequency of features found in persons with Autism and Mental Retardation and their significance in discriminating the two disorders.

Participants and Procedure

The sample of the study consisted of 40 firmly diagnosed persons (20 each with Autism and MR) from five Special Education Institutions of Lahore. Their age range of the children was 3 years onwards. The children below 3 years of age were excluded for this study, as going below 3 years makes it difficult to give a firm diagnosis to the child for Autism. Also the children other than Autism, MR and Down's syndrome, were excluded for the study. The final list of 66 features was administered on children with Autism and MR through parental or teacher's interview. In case of teachers, the minimum criteria were that the teacher should have at least 1 year of work experience in a special education setting and should be working with the index child at least for the last four months.

In phase 3, the sample was collected from five Special Education Institutes of Lahore. Institutional and individual consents were taken. Then the parents (where available) or teacher (in case of non-availability of the parents) were interviewed individually. After filling in the demographic data, the parents or teachers were asked to rate the child's behaviour on a four point scale (0-3) from not all at (O) to very much true (3) for the child. They were asked to keep in mind the behaviour of the child, and rate each feature the degree to which it deviated in terms of frequency, intensity, and peculiarity from the behaviour for the normal child of that age.

RESULTS

Descriptive statistics was applied to demographic variables whereas descriptive as well as inferential statistics were applied to compare the results of each behaviour in autism and MR. The significance of each feature in discriminating the two disorders was computed through Mann Whitney U test at 0.05 levels of significance. Mann Whitney was used as it tests whether two independent samples were from the same population without relying on a normal distribution curve.

Table 1, provides the demographic information of the sample (n=40) in the main study. Most of the students were male (67.5%) and age range of 3-10 years (55%). There was equal number of children from nuclear and combined family systems (50% each). Regarding the respondents qualification, most of the parents (57%) of persons with autism were graduates or above qualified, whereas none of the parents of persons with MR was graduate or above. However teachers were almost equally qualified for both the groups (Autism mean was 16 years, whereas MR mean was 14.8 years of formal education). However, the mean for overall parent's qualification was 11.22 years of formal education. The parents of both the groups had no formal training in special education. However, the teachers who rated students with autism had an average of 9 years of experience in special education whereas it was 12.13 for teachers of persons with MR. All the parents knew their children since birth, whereas the teacher's knew the child with a mean duration of 1.43 years for Autism (ranging from 7 months to 2 years) and 1.16 years for MR (with a range of 4 months to 5 years).

The table 2 shows that 28 features were found significant and are more related to autism (based on their percentage of occurrence. Out of 28 features, 12 were related to social skills, 6 with self help skills, 4 with routines/sensory issues, 2 with stereotypical movements and one each with motor skills and behavioural problems. Fifteen of these behaviours were significant at <0.001 levels (9 related to social skills, 5 with self help skills whereas one belonged to change of routine). Five of the features were significant at <0.01 levels (2 related to changes in routine and one each related to stereotypical movements, communication skills, and self help skills). The rest of 8 features were significant at <0.05 levels (2 related to social skills and sensory issues/limited interests, and one each related to motor stereotypical movements, communication skills, motor skills and behavioural problems.

Table 3 shows that 7 features were significantly found in MR. Six of them were related to developmental milestones whereas one was related to social skills. One was significant at <0.01 levels, whereas the rest of 6 features were significant at <0.05 levels.

Table 4 indicates that 31 features that were validated by experts were not found significant in discriminating between autism and MR. Eight of the features were related to communications skills, 5 to cognitions, 4 each from stereotypical movements, and behavioural problems and 3 each from social skills, limited interests/sensory issues and other skills. One of the features was also related to developmental milestones.

DISCUSSION

Autism is still a mystery which needs to be fully understood. The co-morbidity with any other developmental disorder especially MR makes it even difficult to give the proper and timely diagnosis to such children. The objectives of the study were to generate features that clearly differentiate between autism and MR and those which create confusion between the two disorders. A list of 66 features was generated after interviewing 10 parents each of persons with autism and MR and then validated by ten experts.

The current study has replicated some of the already existing knowledge. For example, it confirms the early findings by Ozonoff et al (2010) that social domain is one of the most important distinguishing factors between the two disorders. The study found that stereotypical movements are not the primary distinguishing feature among autism and MR in our culture, contrary to the recommendations of DMS-IV-TR (APA, 2000). This confirmed the findings of Bodfish et al. (1995) who stated that sterepotypical movements were also found in other disorders as well. There was hardly any distinugishing features to discriminiate the two disorders based on the communcation skills (only one factors was found significant in this domain in discriminating the two disorders i.e. loss of learned speech after two years of age). This confirmed the findings of Kraijer and Bildt (2005) who advocated that speech and language problems give rise to issues of diagnostic overshadowing and diagnostic under-representation.

The study has added to the existing knowledge in a number of ways. It emphasizes that while looking for the differentiation between Autism and MR, one should look in to the social skills, self help skills and interest patterns of the person. Whereas while looking for the possibility of MR, one should focus more on the developmental profile of the child. The interesting finding is that communicaiton skills and stereotypical movements don't distinguish signifcantly between the two disoders. Where as some of the features that were considered as red flags for autism by Autism Speaks (2010) like babbling, social smile were not found significant in differenting the two disorders.

The study hoighlighted the issues in the way the experts look into the problem. For example, some of the features highly rated by experts as belonging to Autism (spinning and handflapping behaviours, does not like to be picked up by others and lining up things) were not found statistically significant in differentiating Autism from MR, confirming the findings of Bodfish et al. (1995) and Kraijer and Bildt (2005) who referred the prbolem as overshadowing and under representation of a diagnosis. It also emphasizes on the importance of having a comprehensive approach towards diagnosis which should include developmental as well as behavioural aspects (Charman & Baird, 2002; Autism in Action, 2011) rather than relying only on behavioural aspects.

It has raised questions in the way information is elicited from the parents and high reliance on checklist and rating scales approach, where presence or absence of a symptom is seen only on the its face value, without going into the details of the circumstances which might have a crucial role in deciding for or against the presence or absence of a symptom. For elicting information, a checklist approach is used which overlooks many of the symptoms that might be crucial in making a diagnosis, like differentiating between the lack of social skills because of shyness or because of autism. The checklist approach usually does not go into details. The most widely used instrument like CARS also states that the deviation should be noted down irrepective of the causes, which limits its ability to differentiate between the two disorders.

The study also identified the features that were dubious in nature and possibley pose the acutal problems in differentiating the two disorders. It suggests that persons involved in diagnosis of Autism and MR, should pay special attention to the features which were found in both the disorders as they might be the deciding factor in differentiating the two disorders. While diagnosing a developmental disorder like Autism or MR, one should pay close look to each and every symptom. The importance should not only be given to the quantity of deviation but the quality of deviation as well, to improve the diagnosis.

Conclusion

Majority of the results were consistent with finding of the previous researches. It not only adds information to our culture regarding the early identification of the developmental problems like Autism and MR, but also identifies that the social domain as the primary area of differentiation between the two disorders. Based on the results of the study, we can conclude

1. A comprehensive approach for the diagnosis of autism and discriminating it from MR should be used which should include the exploration of developmental as well as behavioural features.

2. A semi structured interview approach should be used to elicit the information rather than checklist approach.

3. The features should be explored qualitatively and the context of its occurrence should also be considered while deciding its presence or absence.

4. While differentiating between Autism and MR, special attention should be paid to the dubious feature, which play crucial role in improving our diagnosis of Autism.

Further suggestions

* The findings of the study could lead to the development of an indigenous tool for differential diagnosis of Autism and MR.

* The study gives a guideline to the professional involved in the diagnosis of developmental disorders by adopting a semi structured approach, and paying close attention to the apparently dubious looking features, and claims that by utilizing these features wisely, we can improve the diagnosis of Autism and MR.

* The study could lead to further work in the area and expanding its link to the comorbidity of other developmental disorders.

* This study highlights the importance of using a phenomenological approach towards the developmental disorders and it can be used in future to explore other aspects.

References

References

American Psychiatric Association. (2000). Dignostic and Statistical Manual for Mental Disorders 4h Ediction, Text Revised. Washington DC: APA.

Ashfaq, S., Saeed, T. & Jahangir, F. (2009). Behaviour Management of Children with Autism/Learning Disability through Parents Training: An Exploratory Study. Frontier Women University. Journal of Social Sciences, 3 (1), 315.

Autism in Action. (2011, February 20). Autism in Action. Retrieved February 11, 2011, from http://www.auti5minaction.com/faqs.html

Autism Speaks. (2010). Retrieved March 10, 2011, from What is Autism: http://www.autism5peaks.org/whatisit/learn5igns.php

Bodfish, J. W., Crawford, T. W., Powell, S. B., Parker, D. E., Golden, R. N. & Lewis, M. H. (1995). Compulsions in adults with mental retardation: prevalence, phenomenology, and comorbidity with stereotypy and selfinjury. American Journal on Mental Retardation, 100, 183-192.

Charman, T. & Baird, G. (2002). Practitioner Review: Diagnosis of autism spectrum disorders in 2 and 3-year old children. Journal of Child Psychology and Psychaitry, 43(3), 289-305.

Dawson, G. & Osterling, J. (1997). Early intervention in autism: Effectiveness and common elements of current approaches. In Guralnick (Ed.) The effectiveness of early intervention: Second generation research, (pp. 307-326) Baltimore: Brookes.

Eaves, L C, Wingert, H. D., HO, Helena, H. & Mickelson. E. C.R. (2006). Screening of Autism Spectrum Disorders with the Social Communication Questionnaire. Journal of Developmental and Behavioral Pediatrics, 27 (2), 95-103.

Edelson, M. G. (2006). Willamette University. Retrieved August 16, 2010, from http://www.willamette.edu/dept/comm/reprint/edelson/

Hogrefe. (2010). Retrieved February, O, 2011, from http://www.hog refe.co.uk/?/test/show/64/

Howlin, P. (1998). Practitioner Review: Psychological and Educational Treatments for Autism. Journal of Child Psychology and Psychiatry, 39(3), 307-322

Kraijer, D. (1990). Psychological Instruments, Tests and Services. Retrieved December, 29, 2010, from http://www.pits-online.nl/en/PDD-MRS.html

Kraijer, D. (2006). Psycholgical Instruments, Tests and Services. Retrieved August 16, 2010, from http://www.pits-online.nl/en/PDD-MRS.html

Kraijer, D., & Bildt, d. A. (2005). The PDD-MRS: an instrument for identification of autism spectrum disorders in persons with mental retardation. Journal of Autism and Developmental Disorders , 35(4), 499-513.

Morgan, S. (1988). Diagnostic assessment of autism: A review of object scales. Journal of Psychoeducational Assessment, 6(2), 139-151.

New York Department of Health. Early Internvetion Program (2005). Clinical Practice Guideline. Report of the Guideline Recommendations Autism/Pervasive Developmental Disorders Assessment and Intervention for Young Children (Age 0-3 Years). New York: Deparment of Health. Early Intervention Program.

Oosterling et al. (2010). How useful is the Social Communication Questionnaire in toddlers at risk of autism spectrum disorder? Journal of Child Psychology and Psychiatry, 51(11), 1260- 1268.

Ozonoff et al. (2010). A prospective study of the emergence of early behavioural signs of autism. Journal of American Academy of Child and Adolescence Psychiatry, 49 (3), 256-266.

Rebecca, L. (2007). Early communication development and intervention for children with autism. Mental Retardation and Developmental Disorders Research Rev/ews,13C\.)f 16-25

Rogers, S. (1996). Brief report: Early intervention in autism. Journal of Autism and Developmental Disorders, 26, 243-246.

Rogers, S. (1998). Empirically supported comprehensive treatments for young children with autism. Journal of Clinical Child Psychology, 27, 168-179.

Rutter, M. (2003). Social Communication Questionnaire. USA: Western Psychological Services.

Salehi, S. M., AI-Hifthy, E.H. & Ghaziuddin, M. (2009). Autism in Saudi Arabia: Presentation, clinical correlates and comorbodity. Transacational Psychiatry, 46 (2), 340-347.

Schopler, E., Reichler, RJ., DeVellïs, R.F., & DaIy, K. (1980). Toward objective classification of childhood autism: Childhood Autism Rating Scale (CARS). Journal of Autism and Developmental Disorders, 10, 91 - 103.

Society for the Austically Handcanpped -SFTAH. (2010). Autism Independent Retrieved January, 26, 2011, from http://www.autismuk.com/indexlO.htmtfWHATIS5

Skuse, D. (2007). Rethinking the nature of genetic vulnerability to austistic spectrum disorders. Trends Genet, 23(8), 387-95.

Suhail, K. & Zafar, F. (2008). Prevalence of Autism in Special Education Schools of Lahore. Pakistan Journal of Psychological Research, 23(3-4), 45-64.

Teal, N.B., & Wiebe, MJ. (1986). A Validity Analysis of Selected Instruments Used to Assess Autism. Journal of Autism and Developmental Disorders, 16, 485-494.

Verheij, F. & Doom, E.G. (1990). Autism and mental retardation: the planning of therapeutic environment. International Journal of Rehabilitation Research, 13(2), 127-36.

Vig, S. & Jedrysk, E. (1999). Autistic features in young children with significant cognitive impairment: austim or mental retardation. Journal of Autism and Developmental Disorders, 29, 235-248.

Volker, MA & Lopata, C. (2008). Autism: A Review of Biological Bases, Assessment, and Intervention. School Psychology Quarterly, 23(2), 258270

Received: 18.03.2011