Prevalence of Specific Learning Disorders (SLD) Among Children in India: A Systematic Review and Meta-Analysis

Abstract

Background:

Specific learning disorders (SLD) comprise varied conditions with ongoing problems in one of the three areas of educational skills–reading, writing, and arithmetic–which are essential for the learning process. There is a dearth of systematic reviews focused exclusively on the prevalence of SLD in India. Hence, this study was done to estimate the prevalence of SLD in Indian children.

Methods:

A systematic search of electronic databases of MEDLINE, Embase, PsycINFO, and CINAHL was conducted. Two authors independently assessed the eligibility of the full-text articles. The third author reassessed all selected studies. A standardized data extraction form was developed and piloted. The pooled prevalence of SLDs was estimated from the reported prevalence of eligible studies, using the random-effects model.

Results:

Six studies of the systematic review included the diagnostic screening of 8133 children. The random-effects meta-analysis showed that the overall pooled prevalence of SLD in India was 8% (95% CI = 4–11). The tools used to diagnose SLD in the studies were the National Institute of Mental Health and Neurosciences (NIMHANS)-SLD index and the Grade Level Assessment Device (GLAD).

Conclusions:

Nearly 8% of children up to 19 years have SLD. There are only a few high-quality, methodologically sound, population-based epidemiological studies on this topic. There is a pressing need to have large population-based surveys in India, using appropriate screening and diagnostic tools. Constructing standardized assessment tools, keeping in view the diversity of Indian culture, is also necessary.

Specific learning disorders (SLD), often referred to as learning disability, is a neurodevelopmental disorder (NDD) and refers to ongoing problems in one of the three basic skills–reading, writing, and arithmetic–which are the essential requisites for the learning process. ¹ These difficulties, namely dyslexia, dysgraphia, dyscalculia, dyspraxia, and developmental aphasia, ² can occur alone or in different combinations ranging from mild to severe difficulties. ³

Dyslexia, the reading disability, is the most common condition, accounting for about 80% of all SLDs. ⁴ Dysgraphia is generally characterized by distorted writing despite thorough instructions. The significant characteristic of dyscalculia is the problems in understanding or learning mathematical calculations. About 30% of children with SLD have behavioral and emotional problems, and they are at increased risk for hyperactivity and other comorbidities. ⁵

Although SLD cannot be cured, there are interventions for underlying conditions so that children with SLD can adapt, accomplish academic achievements, and live productive and fulfilling lives. ³ Diagnostic and Statistical Manual of Mental Disorders (DSM-5) estimates the prevalence of all learning disorders (including impairment in writing, reading, and mathematics) to be about 5% to 15% worldwide. ⁶ The lifetime prevalence of learning disability among children in the USA was 9.7%. ⁷ In India, the prevalence of SLD is reported to vary from 3% to 10%. ⁸

In India, although SLD is included as one of the disabilities according to the Rights of Persons with Disability Act of 2016, the screening and diagnosis of SLD are complicated. Various tools are used for the assessment, with their own merits and demerits. Some tools like the AIIMS SLD: comprehensive diagnostic battery ⁹ and the National Institute of Mental Health and Neurosciences (NIMHANS) index for SLD ¹⁰ are commonly used for assessment, but there is a dearth of well-established norms for the subtypes of SLD. There is no screening tool available for teachers to identify SLD, and various education boards (central and state boards) have different levels of academic curriculum. Some tools like the NIMHANS index for SLD can only be administered in English-medium schools, whereas in India, about 42% of students are studying in Hindi-medium schools. ¹¹ Although many tools are developed in regional languages like Tamil, Kannada, and Marathi, ¹² there is no nationwide acceptability of these tools to certify children with SLD.

It is crucial to have a review to know the depth and breadth of the problem and the differences in the diagnostic criteria used in the studies. There is a lacuna in the evidence regarding the prevalence of SLDs, and usually, they go undetected.^13,14 Early diagnosis and assistance for a child with SLD is the need of the hour, and thus it is also essential to know about the diagnostic methods used. There is a lack of systematic reviews focused exclusively on the prevalence of SLD in India. Estimating the prevalence of SLD in India is valuable in planning diagnostic and intervention services. Information regarding the overall estimate of SLDs in the country will help develop a school-based policy for early identification, referral, and management of children with SLDs. Hence, this study was designed to perform a systematic review and meta-analysis to estimate the prevalence of SLD in Indian children and review the tools used for diagnosing SLD.

Materials and Methods

The protocol for the review was registered with PROSPERO (registration number- CRD42020154690).

Data Sources and Search Strategy

Two investigators (LMS and DB) searched the electronic databases of MEDLINE, Embase, PsycINFO, and CINAHL. Data search was carried out between June and August 2021. Because SLD prevalence studies were published since 1990, the authors selected 30 years to review articles (1990–2020). Additional searches were conducted in Google Scholar and grey literature sources such as documents of conferences and government websites. Hand searching and retrospective searching of relevant published literature was also done. All English-language studies containing information on SLD prevalence among children and adolescents aged 6 to 19 years were retrieved. To select the upper age limit, the WHO definition of adolescents as 10 to 19 years was adopted. ¹⁵ From the selected studies having information on the prevalence of SLD, information on screening criteria and tools used to diagnose SLD was identified and reviewed. A search strategy that included the combination of subject terms and free-text terms was employed using the operators “OR” and “AND.” The Medical Subject Headings (MeSH) terms were SLD, learning disability, learning disorder, dyslexia, dysgraphia, dyscalculia, prevalence, and India. All MeSH terms were exploded where necessary (Table 1).

Table 1.

Search Strategy Used in MEDLINE Database (1989-2020)

Number	Search Terms
1	prevalence/or incidence/or prevalence [MeSH Terms]/or prevalence*
2	AND
3	learning disability/or learning disabilities/or learning disorder [MeSH Terms] or learning disorder*/or dyslexia [MeSH Terms]/or dysgraphia [MeSH Terms]/or dyscalculia [MeSH Terms]
4	AND
5	children/OR child*/or child aged less than 18 years
6	AND
7	India/OR Indian/OR Indian studies

Population

The population of interest was school-going children residing in India aged 6 to 19 years who were assessed for SLD using different existing tools for diagnosing SLD.

Inclusion Criteria

Observational studies, including cross-sectional, cohort, or case-control studies, of children with SLDs, using validated or nonvalidated tools, published in the English language and conducted in community settings, were included. Where multiple publications were generated from the same data with the same outcome, only the most relevant study was included.

Exclusion Criteria

Studies that discussed therapy, management, and comorbidities of SLDs were excluded. Studies conducted in hospitals were excluded because the children are likely to be highly selected (i.e., selection bias), resulting in inaccurate estimations of the true prevalence of SLDs. Studies were excluded if children were not screened for intelligence quotient (IQ). Editorials, letters, opinion articles, narrative or systematic reviews, brief communications, and posters were excluded.

Screening Strategy

Two authors reviewed the titles and/or abstracts of studies identified using the search strategy and those from additional sources. They independently assessed the eligibility of the full-text articles. The third author (BG) reassessed all selected studies. Any disagreement between the reviewers was resolved through discussion with the third author.

Quality Analysis

The quality of reporting in the selected articles was checked using Strengthening the Reporting of Observational Studies in Epidemiology (STROBE). The STROBE checklist for cross-sectional studies was used to evaluate the relevant information from each article. LMS and DB independently assessed studies’ reporting quality. In case of any disagreement on this assessment, the issue was resolved by discussion or consensus with the third investigator (BG).

Data Extraction

A standardized data extraction form was developed and piloted based on the Cochrane good practice data extraction form template to extract data from the selected studies. Extracted information included study design and methods, study settings, participant characteristics, study outcomes, results, conclusions, and study funding sources.

The pooled prevalence of SLDs was estimated from the reported prevalence of eligible studies, using the random-effects model. Analyses were performed using STATA 16 (College Station, Texas, USA) software. Forest plots were generated displaying prevalence with the corresponding 95% confidence intervals (asymptotic Wald) for each study. The I-squared (I2) test was used to assess heterogeneity. The tools used to diagnose SLD were identified from the selected articles and reviewed.

Results

Literature Search

The preferred reporting items for systematic reviews and meta-analyses (PRISMA) statement flowchart ¹⁶ in Figure 1 describes the literature screening, study selection, and reasons for exclusion. Out of 17 studies assessed for eligibility, 11 were excluded for the following reasons: management/interventional/risk factor studies,^17,18 no diagnosis done/only screened for different SLDs,^19–22 the prevalence of SLD was not assessed,^23,24 the study did not screen for the intelligence of the participant children, ²⁵ and studies assessed only dyslexia.^26,27 A total of six studies met the inclusion criteria for this review and were finally included in the meta-analysis (Table 2).^28–33

Figure 1. PRISMA Flow Diagram of the Review Process and Study Selection.

Table 2.

Characteristics of Selected Studies

Author	Year	Region	Study Setting	Age in Years	Number of Children with SLD	Total Number of Children Surveyed	Male Female Ratio	SLD Studied
Mogasale et al.	2012	Belgaum, Karnataka	School	8–11	165	1088	1.69	SLD–total and Dyslexia Dysgraphia Dyscalculia
Arun et al.	2013	Chandigarh	School	12–19	38	2402	1.33	SLD–total
Arora et al.	2018	Himachal Pradesh, Haryana, Odisha, Andhra Pradesh, Goa	Community	6–9	32	1970	1.01	SLD–total
Sharma et al.	2018	Gwalior, Madhya Pradesh	School	NA (third –sixth standard)	23	800	0.97	SLD–total and Dyslexia Dysgraphia Dyscalculia
Shah and Buch	2019	Jamnagar city, Gujarat	School	7–12	38	393	0.87	SLD–total and Dyslexia Dysgraphia Dyscalculia
Chacko and Vidhukumar	2020	Ernakulam, Kerala	School	8–12	244	1480	1.03	SLD–total

SLD, specific learning disorders.

Description of Included Studies

The studies included in this review were conducted in different states of India, including Andhra Pradesh, Chandigarh, Goa, Gujarat, Haryana, Himachal Pradesh, Karnataka, Kerala, Madhya Pradesh, and Odisha. All were cross-sectional studies done among children aged 6 to 19 years. The studies assessed children at a younger age itself, except for Arun et al., for which the age group was 12 to 19 years. Three studies assessed the subcategories of SLD separately along with the total prevalence of SLD^31–33; all other studies assessed SLD in total and not the subtypes. The study by Arora et al. was done in the community setting, ²⁹ and all the other studies were conducted at schools. Of the studies conducted in schools, the study setting of four included both private and government schools. Three studies were conducted in urban areas alone.^31–33 The grade in which the students were studying ranged from Class II to Class XII. The articles by Mogasale et al., Sharma et al., and Shah and Buch assessed students of Classes III to IV, III to IV, and II to VI, respectively.^31–33

SLD was diagnosed with different diagnostic tools in different studies. The tools used to screen and diagnose SLD were the NIMHANS-SLD index and Grade Level Assessment Device (GLAD ³⁴ ; Table 3). All the studies except Arora et al. used the NIMHANS-SLD index to diagnose SLD. The tool is available for English-medium students, and while using this tool, the authors used local language textbooks of lower grades for assessments.

Table 3.

Methodological Details of Specific Learning Disorders’ Screening and Evaluation Done

Criteria	Mogasale et al.	Arun et al.	Arora et al.	Sharma et al.	Shah and Buch	Chacko and Vidhukumar
IQ test	Yes	Yes	Yes	Yes	Yes	Yes
Hearing test	Yes	Not mentioned	Yes	Yes	Yes	Yes
Vision assessment	Yes	Not mentioned	Yes	Yes	Yes	Yes
Other NDDs excluded	Yes	ADHD not excluded	Yes	Yes	Yes	Locomotor impairment
Diagnostic measure	NIMHANS index	NIMHANS index	GLAD	NIMHANS index	NIMHANS index	NIMHANS index
Case ascertainment	Screened by pediatric postgraduates diagnosed by a clinical psychologist	Screening by teachers based on a six-item proforma diagnosed by psychologist	The diagnostic team comprised a physician, audiologist/speech therapist, and psychologist	Initial screening using academic performance screened for vision and hearing from the pediatric outpatient department	Team of developmental pediatricians, special educators, and psychologists	Screened by parents or teachers using a learning disorder screening tool diagnosed by a psychiatrist

NDDs, neurodevelopmental disorders; ADHD, attention deficit hyperactivity disorder; GLAD, grade level assessment device; NIMHANS index, National Institute of Mental Health and Neurosciences index for SLD; SLD, specific learning disorders; IQ, intelligence quotient.

The highest prevalence rate of SLD from individual studies was reported as 16.49% by Chacko and Vidhukumar, ³⁰ followed by Mogasale et al., who reported a prevalence rate of 15.17%. ³¹ The least prevalence was reported as 1.58% by Arun et al. ²⁸ Mogasale et al. reported 12.5%, 11.2%, and 10.5% prevalence of dysgraphia, dyslexia, and dyscalculia, respectively, ³¹ while the prevalence of SLD subtypes–dysgraphia, dyslexia, and dyscalculia–reported by Shan and Buch was 7.4%, 8.6%, and 7.1%, respectively. ³²

The six studies of this systematic review have included the diagnostic screening of 8133 children. The random-effects meta-analysis showed that the overall pooled prevalence of SLD in India was 8% (95% CI = 4–11, Figure 2). In this meta-analysis, a high level of heterogeneity (98.72%) was observed between the studies. The diamond in the result represents the point estimate of 7.7% from all the individual studies together. The horizontal point of the diamond represents the 95% confidence interval of this combined point estimate.

Figure 2. Prevalence of Specific Learning Disorders Among Children Aged 6 to 19 Years in India (Random Effect Model).

Subgroup analysis and meta-regression were not attempted because the studies did not mention urban-rural differences or gender differences. Also, characteristics such as age group, type of study, and the diagnostic measure did not vary much among the studies.

In this meta-analysis, because the outcome measure is the prevalence and the probability that significance levels that may have biased publications are less, publication bias may not be applicable. The reasons for nonpublication are more likely small studies not using appropriate methodology. All the selected articles satisfied the STROBE criteria for reporting.

Discussion

This systematic review reports an 8% prevalence of SLD in India. All the enrolled studies were recently published from 2012 to 2020. However, the six studies included in this review used a spectrum of tools for screening and diagnosis of SLD.

There is no single screening and diagnostic tool that may be considered specific for the diagnosis of SLD. The NIMHANS index for SLD was developed in 1991 in the Department of Clinical Psychology, NIMHANS, Bangalore. The NIMHANS index for SLD is a curriculum-based assessment that can be used to confirm the diagnosis of SLD. ¹⁰ It includes tests of reading, writing, spelling, and arithmetic abilities to detect children with disabilities in these areas. There are norms for children in Standards I to V. This battery can be used not only for confirming an initial diagnosis of SLD but also for certification of SLD in India. The Gazette of India (No. 61, dated January 6, 2018) states that the NIMHANS-SLD index shall be used to diagnose SLD in children. The tool can also be used for the assessment of improvement after remediation. However, the different types of SLDs cannot be picked up using this battery. ³⁵ Besides, since the tool is in English and India is a multilingual country, professionals find it challenging to assess SLD in a child’s mother tongue.

The assessment using GLAD includes the level of functioning and process of learning. In developing this tool, the National Council of Educational Research and Training (NCERT)-prescribed minimum levels of learning (MLL) were taken as standard. English, Hindi, and Mathematics textbooks from Class I to Class IV of the Central Board of Secondary Education (CBSE), Indian Certificate of Secondary Education (ICSE), and the state board in Andhra Pradesh were used to develop the tool. Items were taken from all the syllabi based on the MLL fixed based on NCERT stipulations. ³⁴

There is a dearth of acceptable tools that are developed and validated in regional languages, particularly in rural parts of the country and the Adivasi population, where the dialects are different. The tools accepted for diagnosis of SLD are developed for students of English-medium schools, whereas in India, only one-fourth of the students study in English-medium schools. ³⁶ The content used in the tools is not standardized. Existing tools have not included all the age groups for assessment, which makes assessment difficult, especially when the student is to be assessed in tenth or twelfth classes to issue a certificate for availing benefits. ³⁷

In a population-based prevalence estimate from the USA, the prevalence of SLD reported was 9.7% in children aged 3 to 17 years. ⁷ Nearly 5% of the US school-age population have learning disabilities that have been formally identified. ³⁸ Our study reports that nearly one in twelve Indian children have SLD. In Brazil, recent estimates show that the prevalence rate was 7.6% for global impairment, 5.4% for writing, 6.0% for arithmetic, and 7.5% for reading impairment. ³⁹ Also, an epidemiological study from Turkey found the prevalence rate to be 13.6%. ⁴⁰ A recent estimate from Pakistan showed a similar prevalence of 7.7% among primary school children. ⁴¹

SLDs are challenging to diagnose and are often not well understood as a group of disorders. There is a gap of nearly four years between the child’s age at SLD diagnosis and the mother’s first suspicion of a problem. ⁴² The treatment of SLD focuses on educational interventions, and early interventions are most desirable. ⁴³ Therefore, it is crucial to identify SLD as soon as possible.

Lack of appropriate resources, tools, and support and lack of awareness among parents and school teachers are significant issues in the Indian context. ⁴⁴ The multiple curriculums at schools, varying standards, and multilingualism prevent a unifying standardized approach. ⁴⁵ Regional adaptations in protocols and universal screening of children are the vital components. Prospective studies (across different states and vernacular languages), multicenter collaborations, and longitudinal research with a large sample and a single comprehensive test battery are needed to understand the situation better and make the children achieve their maximum potential. Also, SLD epidemiology needs to develop into the arenas of operational research to study the utilization pattern of services as well, thereby making care available to those in need.

The high prevalence of SLD among children in India implies the need for awareness generation among parents and teachers. Adopting community sensitization programs will be beneficial for early identification and improving access to remedial education programs. Advocating and strengthening the integrated education system, management of comorbidities, and prevention of mental health problems will improve the quality of life of children with SLD.

This systematic review had a few limitations. There was heterogeneity in the methodology among the applied screening and diagnostic tools used in the included studies, which might have led to under‑ or over-estimation of the prevalence data. The prevalence rate trend analysis was not done because the studies were published recently within ten years. Subgroup analysis on rural versus urban population and male and female sex could not be done because the articles did not mention the required data. Prevalence in the subgroups of ages could not be assessed because data were not available for different age groups.

Conclusion

This review systematically analyzed data from Indian studies to determine the prevalence of SLD in India. This is the only systematic review on the topic so far, and it demonstrated that nearly 8% of children have SLD. This may include mild, moderate, and severe cases. The conclusion shall be inferred taking care of the study’s limitations. The study also highlights that there are only a few high-quality, population-based epidemiological studies on this topic. This review has contributed to explaining the prevalence estimates of SLD in India. The impact of factors such as urban or rural location, age, diagnostic tools, and medium of instructions on SLD prevalence needs to be further investigated. As India is a vast country, there is a pressing need to have extensive population-based surveys using appropriate screening and diagnostic tools. Constructing standardized assessment tools, keeping in view the diversity of Indian culture, is an enormous task. Similarly, regional-language-based screening and diagnostic tools must be developed for easy identification and reporting. Since SLD is included as one of the disabilities in the RPWD Act 2016, diagnosis and certification are warranted. Early diagnosis and disability certification are essential requirements for providing equal opportunities, equal rights, and equal participation of the children in the community.