Assessing the Strengths and Gaps of PRASHAST in Screening for Mental Disabilities Among School-Going Children

The Government of India previously estimated that by 2022, there would be 22.7 lakh children with special needs in schools, accounting for about 10% of 26.7 crore children in pre-primary to higher secondary schools. ¹ Section 16 (vi) of the Rights of Persons with Disabilities (RPwD) Act (2016) mandates the appropriate government to detect disabilities in school children and directs the use of necessary pedagogical and other supportive measures for children identified with such disabilities. ² This Act describes 21 conditions as disabling and has laid down the guidelines for medical boards to certify benchmark disability, which is defined as not less than 40% disability. ² The Right to Education Act of 2009 also guarantees free and compulsory education to all children, including those with special needs (CWSN). ³ From these, we can understand that the responsibility of the government to every school-going child is to screen for any special needs or disability as early as possible, and school teachers are in a key position to identify children who may have a disability and to refer them to appropriate medical boards for a thorough evaluation and certification.

The Department of Education of Groups with Special Needs, Central Institute of Educational Technology-National Council of Educational Research and Training (CIET-NCERT) created a screening tool titled ‘PRASHAST’, ⁴ to assist regular teachers and special educators in screening and tentatively categorising children under the 21 different disability conditions listed in the RPwD Act (2016) ² at the school level and refer them to appropriate medical boards. The PRASHAST helps regular teachers in identifying the children with any of the 21 disabilities described under the RPwD Act 2016, including four mental disabilities: autism spectrum disorder (ASD), specific learning disability (SLD), intellectual disability (ID) and mental illness (MI). This tool is expected to be used by lakhs of regular teachers and hundreds of special educators for screening the presence of disabilities in school-going children, including four mental disabilities: ASD, ID, SLD and MI. However, PRASHAST is not published in any peer-reviewed journal. A critique of PRASHAST may help to know more about its strengths, limitations, if any, and its potential usefulness to screen four mental disabilities as described above, which, at times, may go unnoticed by teachers in schools.

Brief Description of the Tool PRASHAST

The PRASHAST tool is made available to all as a soft copy and electronic (E) book with a few details about its development, field testing and people involved in its expert validation. ⁴ This tool is meant for students studying from pre-primary to standard XII. Therefore, it covers all age groups of students studying in school. PRASHAST contains two parts. ⁴ Part 1 contains 63 questions and is intended to be used by regular school teachers after observing the student for at least 2 weeks or more in different contexts. Children that found positive for any of the 21 disabling conditions would be referred to special educators, counsellors and school heads who would use PRASHAST ⁴ Part-2 containing 150 questions. All of these questions have to be answered either Yes or No. In the PRASHAST E-book, special instructions are included on how to use Parts 1 and 2. ⁴ The special educator/resource teachers/counsellors will orient the general teacher through a meeting/workshop for using PRASHAST Part-1, and the general teacher should read the document carefully before its use. ⁴

Online Availability

The CIET-NCERT also launched an Android application for PRASHAST, which is available on the Google Play Store. ⁴ We tried to download it to test it for our understanding, but it is only for teachers/special teachers/principals of schools. Only persons with a valid Unified District Information System for Education, an 11-digit number of their school allotted to them by the appropriate government, are eligible to use it. We observed that as of 24 March 2025, from the Google Play Store, about 3,300 downloads of this tool happened, and it has a 3.3 rating out of five stars with variable feedback from teachers. We observed that some of the negative remarks are about the difficulties in registration and using the tool rather than the content. Both the E-book and mobile app versions of PRASHAST are freely available on the Central Institute of Educational Technology website. ⁴ In the PRASHAST Part-1, there are 4 items for ID, 10 items for SLD, 10 items for ASD and 8 items for MI. In Part 2, there are 10 items for ID, 10 items for SLDs, 13 items for ASD and 14 items for MI. A total of 32 items in Part-1 and 47 items in Part-2 serve as a checklist to ensure that no child with any one of these four disabilities is missed. This is a strength of this tool.

Guidelines and Instructions for PRASHAST Usage

It is explicitly mentioned in the PRASHAST E-book that it should be used only for screening the child for disability and for referring to a specialist for further evaluation, but not to diagnose and label any disability. ⁴ A separate set of guidelines for regular teachers, special educators and heads of schools/principals on how to use PRASHAST, ⁴ which we feel are essential and improve the ease of using this tool. In these guidelines, developers suggest that teachers/special teachers and principals have a better understanding by interacting with relevant stakeholders, by keen observation of children for a reasonable period and by attending orientation programs on how to use PRASHAST. ⁴ An explicit mention is also made that observations and findings through the screening should be confidential and shared only with relevant stakeholders.

Discussion

Strengths of PRASHAST

A few questions like in Part 1, SLD category, item no. 12.10 reads as ‘This student has difficulty in understanding mathematical symbols such as +, −, ×, ÷’, and in the ASD category, item no. 13.4 reads as ‘This student finds it difficult to deal with sudden changes in routine, for example, change in class teacher/change in classroom/timetable/or seating arrangement’. We feel that these kinds of questions are easy for regular teachers to understand, even if they lack or have inadequate training about learning difficulties that children may have.

At the beginning of the PRASHAST E-Book under ‘Why PRASHAST and Development of PRASHAST (Page 1–5)’, authors report that they did a status survey in 2019 in 34 states and or union territories and developed an in-house draft of PRASHAST by August 2019. ⁴ Authors report that following this, they conducted a 3-day workshop to vet with several stakeholders and finalised for a field trial. Authors further state that field testing was conducted initially in 29 schools of three states (Karnataka, Odisha and Meghalaya) in the year 2019 and subsequently in 100 schools in Chandigarh. The authors state that 3,482 children were screened with Part-1, and 637 children were administered Part-2, with 18.2% positivity on Part-1. Authors further mention that there is 95.7% similarity in the suspected disability as per PRASHAST assessment and subsequent disability certification. The PRASHAST developers report a Cronbach’s alpha of 0.801 for 63 items in Part-2, indicating good internal consistency. ⁴ Further to the field test, developers report that PRASHAST was vetted by national and international experts in 2021 before its release in 2022. ⁴

Limitations of PRASHAST

Incomplete Details About the Development and Validation

Explicit details regarding the development of initial items for the PRASHAST draft version, what the initial items were, and how items were reduced or added are unclear in the E-book. ⁴ How the items were formed, for example, whether or not the domains were specified at first; from where the items were taken, for example, a particular total number of items from theory/review of literature/experts, etc., is not clear from the E-book. ⁴ Moreover, no information has been provided regarding criterion or construct validity.^5,6 The discussion on reliability solely focuses on Cronbach’s Alpha, with a reported value of 0.801. However, item statistics such as mean, standard deviation, inter-item correlation and item-total statistics, including corrected item-total correlation and Cronbach’s Alpha if the item is deleted, ⁷ have not been mentioned. In the validity section, it has been indicated that the level of resemblance between the suspected disability data derived from PRASHAST and the disability certification data obtained from assessment and certification camps is 95.76%. ⁴ However, it lacks clarity regarding the method employed to determine the similarity between the Part-2 result and the disability certification data from the mentioned camps. The specific statistical test utilised for this purpose has not been specified. It is unclear whether the data considered in the PRASHAST analysis pertain solely to Part-1, Part-2 or both parts collectively. ⁴ The reliability section alludes to including 63 items in PRASHAST, along with information about the alpha score. ⁴ This suggests that the reliability assessment pertains specifically to Part-2, thus leaving questions about the reliability of items in Part-1.

It is not clear whether content validity has been assessed by following the method of Lawshe (1975), ⁸ Lynn (1986) ⁹ or any other approach. Authors did not report which diagnostic classificatory system, such as the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition ¹⁰ or International Classification of Diseases, Eleventh Revision, ¹¹ was used for the items related to ID, ASD or SLD. There are a few questions which are not very relevant to the disability being screened, for example, in Part-2, 11.8 (ID) reads as ‘Do you think this student has improper eye contact’ and 11.9 reads as ‘Do you notice repetitive gestures like spinning on a certain spot, enjoying a particular lifestyle and way of object arrangements which can otherwise cause anxiety in this student?’; 11.10 reads as ‘Does this child react to external changes in unexpected manner?’ These three questions are more relevant for children with ASD. In Part-1, item no. 12.5 ‘This student has a comparatively short attention span or is unable to concentrate on a task’ belongs to the SLD category. However, such a reduced attention span is non-specific. It can be observed in children with ASD, ID and MI, but is not included under those conditions, as if it is specific to SLD. Item no. 13 of ASD reads as ‘Does this child reverse pronouns or avoid using pronouns? for example, when the teacher asks “Have you brought your homework?” the child responds “you brought your homework/Rani brought your homework?”’. Here, better and simpler explanations could have been offered. Regarding the section on decoding steps, point five states the following: ‘Observe the child to arrive at a maximum of 2 possible disability conditions and mention these disabilities in Table 2’. Unfortunately, this statement is difficult to comprehend. The logic behind ‘maximum number of 2 disability condition’ is not clear, as some children might have more than two disability conditions.

Scope for Improvisation During Development of the Tool and While Reporting

Developers created a symptom checklist for disorders such as SLD, ASD, ID and MI and ended up including several questions as there are no specific symptoms that could help to screen any child. Instead, they should have focussed on the impact assessment like ‘Does this child has significant difficulties in reading/writing/arithmetic even though he/she is good in comprehending and he/she is as good as others for verbal questioning’, ‘Does this child has consistently low academic performance despite one to one individual attention for a considerable time’ or ‘Is this child is significantly or markedly different from others of his class in terms of his social interactions/playing with others/adjusting to changes etc.’.

Additionally, a scientific elucidation of the methodology employed by the experts in finalising the items could have been better. In the E-book, it would have been beneficial to specifically identify the designation and years of experience of the disability experts involved in the vetting and finalisation workshop. There is incongruent information regarding the place of the field test, that is, in the ‘Preface’, it has been mentioned as designated schools in Delhi. In contrast, at another location in the same E-book under the subheading ‘Validity’ of ‘Psychometric Properties of PRASHAST’, it is stated that PRASHAST was administered in 100 schools in Chandigarh. Inconsistent information about the locations of the field test could have been modified with perhaps an independent peer review before this e-book was published. A few questions were repetitive and should have been removed.

After analysing this document, we feel that it would have been better if the test development protocol had been followed, and the result of each step had been mentioned with clarity and the rationale behind the use of each step. For example, the steps of content validation development stages, that is, the theoretical definition of each disability, domain of each disability, how items were formed and how many items were formed from which sources, socio-demographic detail of each expert as per all 21 disability categories, framing of items, for example, not merging two items in one item by using ‘or’; evaluation of each elements’ representativeness and relevance; items’ structure, form, content; peer reviewing; response-scale specification; the procedure for interpretation of the total score calculated from the administration; etc., as recommended by other researchers, for example, Streiner et al. (2015). ¹² Then mentioning in detail regarding judgment quantification, for example, which method (content validity ratio or content validity index or any other) was used to evaluate the content validity of items as well as an entire tool (Scale level content validity index—S-CVI) individually for each disability was absent in the E-book. ¹³ The probability of chance agreement and modified kappa would have been mentioned with this. In the field try out, an assessment of the face validity by using both quantitative methods, for example, procedures as recommended by Juniper et al. (1997), ¹⁴ item analysis (mentioning the Cronbach alpha; item-total correlations; corrected item-total correlation; mean, skewness and kurtosis) and qualitative method, for example, cognitive interviewing. ¹⁵ Different aspects of the tool, such as cultural sensitivity, fairness, language, test duration, technical issues and overall teachers’ perceptions, would be analysed through a checklist. ¹⁶ With this, mentioning the score of the abovementioned content validation judgment quantification steps would have been done. After these, the reliability assessment in detail and other validities such as construct validity, convergent and divergent validity, and criterion validity would have been mentioned individually for each disability category.

Conclusions

The PRASHAST, a screening checklist tool for early identification of disability in school children, is made available for use by teachers and special educators by NCERT, ⁴ aligning with the provisions of the RPWD Act (2016) Section 16. ² An in-depth review of this tool and its manual for using this tool raises concerns about its sensitivity, specificity, validity and reliability and, therefore, its usefulness as a screening tool in school children. Time for using it in each child; symptom identification instead of impact assessment as if to diagnose a clinical condition; and an overly ambitious attempt to screen all 21 disabilities with a symptom checklist by non-professionals can limit its usefulness as a screening checklist in schools. More research is needed to understand its ease of administration, user-friendliness, sensitivity to pick disabilities in children and need for modifications, if any.