Neurocognitive Evaluation:
Neurocognitive testing was conducted in dedicated evaluation rooms, where the testing enviornment was free of distractions and both the examiner and examinee were comfortable. The Indian adaptation of Wechsler Intelligence Scale for Children-Fourth Edition(WISC-IVINDIA) was used for the assessment of neurocognitive deficits9. This is a valid, reliable and individually administered comprehensive clinical instrument for assessing the cognitive ability of children aged between 6 years 0 months to 16 years 11 months and is available commercially. WISC-IVINDIA assesses four primary domains of cognition, using 15 subtests, which represent a childs abilities in more discrete cognitive domains. The scores of the four domains are integrated to generate the Full Scale Intelligence Quotient(FSIQ), that represents a child’s general intellectual ability.
Before the start of study, training about the WISC-IVINDIA scale was obtained from the Child Psychologist, who supervised the assessments and was a co-investigator. During the study, one fifth of the assessments were randomly cross-checked by the Psychologist.
As per WISC-IVINDIA the first cognitive domain assessed is the Verbal Comprehension Index(VCI), using subtests on Vocabulary(VC), Comprehension(CO), Word reasoning(WR), General knowledge(Information, IN) and finding similarities(SI). The second domain assessed is Perceptual Reasoning, which is the ability of a person to draw upon visual-motor and visual-spatial skills, organize their thoughts, create solutions, and then test them. The Perceptual Reasoning Index(PRI) is assessed by Block designing(BD), Picture conceptualization(Picture concepts, PC), Matrix reasoning(MR) and Picture completion(PCm). The third domain tested is the Working Memory Index(WMI). Participants are assessed on their ability to repeat numbers in order(Digit Span, DS), read a sequence of numbers and letters(Letter-number sequencing, LN) and mentally solve arithmetic problems(Arithmetic, AR). The fourth and final domain evaluated is processing speed which tests the childs ability to focus their attention, quickly scan and discriminate between objects and put them in order sequentially. The Processing Speed Index(PSI) is calculated by drawing symbols in corresponding shapes(Coding, CD), searching for symbols(Symbol Search, SS) and marking and cancelling targeted pictures(Cancelling, CA).
There are 10 primary and 5 supplemental subtests. The primary sub tests are vocabulary, comprehension, finding similarities, block designing, picture conceptualisation, matrix reasoning, digit span, letter-number sequencing, coding, and symbol search(10 total). Two subtests were administered to obtain each of the primary index scores. FSIQ was derived from 8 of the 10 primary subtests. Acceptable substitutions as per WISC-IVINDIA manual were taken. Hence, for the purpose of this study, subtests selected for assessment of FSIQ were directed at assessing comprehension, similarities, picture completion and picture conceptualisation, digit span, arithmetic, coding and cancellation. Each subtest provided a ‘Raw Score’. These ‘Raw Scores’ were converted into a ‘Scaled Score’ using the WISC-IVINDIA manual based on the exact chronological age, using the Date of Birth or age in completed years as remembered by the guardians. Each primary index domain was derived from the sum of scaled scores.
The FSIQ was obtained from the four primary index scores. The prevalence of neurocognitive deficits was calculated based on this FSIQ, with presence of neurocognitive deficits taken at a FSIQ of <90. Spectrum of deficits was analysed by calculating the prevalence of deficits in each of the 4 primary index scores. Any score less than 90 was considered as a deficit in that particular domain. A structured record form was used to document all assessment details, scores and calculations.
Statistical Analysis: The data was analysed using with STATA 14.2 software. Categorical data was tested by the Chi-square or Fischer extract test and presented as percentage or proportion. Continuous data, which was normally distributed, was presented as Mean ± Standard Deviation(Mean ± SD) and the data which is not normally distributed is presented as Median with corresponding Inter-Quartile Range(Median + IQR). The prevalence of neurocognitive deficits in the sample population was calculated in percentage, including complete and individual demain deficits. The various factors associated with neurocognitive deficits in sample population was also assessed using multiple logistic regression analysis. Odds ratio was calculated along with 95% confidence interval. In all analysis, p value <0.05 was considered to represent a statistically significant difference.