Self-perceived health-related quality of life of Indian children with specific learning disabilityS Karande, R Venkataraman
Department of Pediatrics, Division of Pediatric Neurology, Learning Disability Clinic, Lokmanya Tilak Municipal Medical College and General Hospital, Sion, Mumbai, Maharashtra, India
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0022-3859.105442
Source of Support: None, Conflict of Interest: None
Background: Specific learning disability (SpLD) often remains undetected, resulting in the afflicted child experiencing chronic poor school performance. Aims: To measure and analyze the self-perceived health-related quality of life (HRQoL) of children with newly-diagnosed SpLD. Settings and Design: Cross-sectional questionnaire-based study in our clinic. Materials and Methods: From February to December 2008, 150 children consecutively diagnosed as having SpLD were enrolled and their HRQoL documented using the DISABKIDS chronic generic module self-report version instrument. Statistical Analysis: Multiple regression analysis was carried out for determining the 'independent' impact that each of the clinical and socio-demographic variables had on a poor facet score outcome and on a poor total score outcome. Results: Clinically significant deficits were detected in all 6 facets, namely: 'large deficits (effect size ≥−0.8)' in "social exclusion", "emotion", "limitation", "treatment", and "independence"; and 'medium deficit (effect size −0.5 to <−0.8)' in "social inclusion"; and 'large deficit' in "total score". Multivariate analysis revealed that: (i) not belonging to the upper socio-economic strata of society was an independent predictor of a poor "independence" facet outcome (P=0.010, OR=1.99, 95% CI: 1.18 to 3.37); (ii) not having experienced class detainment was an independent predictor of a poor "emotion" facet outcome (P=0.008, OR=3.04, 95% CI: 1.34 to 6.85); (iii) first-born status was an independent predictor of a poor "limitation" facet outcome (P=0.022, OR=2.60, 95% CI: 1.15 to 5.90); and (iv) female gender was an independent predictor of a poor "social exclusion" facet outcome (P=0.024, OR=0.28, 95% CI: 0.09 to 0.85) and a poor "overall health" outcome (P=0.025, OR=0.32, 95% CI: 0.12 to 0.87). Conclusions: Children with newly-diagnosed SpLD perceive their psychosocial, physical, and overall HRQoL to be significantly compromised.
Keywords: Dyslexia, quality of life, scholastic backwardness, socio-economic factors, students
Specific learning disability (SpLD) is a group of neurodevelopmental disorders, which manifest in childhood as persistent difficulties in learning to efficiently read ("dyslexia"), write ("dysgraphia"), or do simple mathematical calculations ("dyscalculia") despite normal intelligence, conventional schooling, intact hearing and vision, and adequate motivation and socio-cultural opportunity. , Up to 5-15% of "seemingly normal" school children are believed to have this disability. , Because of a general lack of awareness, children with SpLD often remain undetected leading to chronic poor school performance, class detention, and even dropping out of school. 
The health-related quality of life (HRQoL) is a multidimensional measure of the overall condition of a human life, namely: (i) physical, and (ii) psychosocial (mental and social environment).  It is known that self-reported HRQoL of children can vary from their parents-proxy ratings.  In an earlier Indian study, parents have reported that by the time this disability is diagnosed, the 'overall' HRQoL of children is already significantly compromised; with their psychosocial health being adversely affected more than their physical health.  Hence, we conducted the present study to investigate self-reported HRQoL in children with SpLD.
The aim of the present study was to: (i) measure and analyze the self-reported HRQoL of children with newly-diagnosed SpLD, and (ii) to evaluate the impact of clinical and socio-demographic characteristics on their HRQoL. Our hypothesis was that children with newly-diagnosed SpLD would also report their HRQoL to be significantly compromised in both psychosocial and physical domains of health.
Diagnosis of SpLD
Each child had been referred to our clinic for assessment of academic underachievement and was assessed by a multi-disciplinary team comprising of pediatrician, counselor, clinical psychologist, and special educator.  Only children above 8 years of age were included in the study as a conclusive diagnosis of SpLD cannot be made till then. , Audiometric and ophthalmic examinations were done to rule out non-correctable hearing and visual deficits (of ≥40 percentage disability) as such children do not qualify for a diagnosis of SpLD. , The pediatrician took a detailed clinical history and did a detailed clinical examination. The socio-demographic characteristics of each child were documented. The modified Kuppuswami's classification was used to determine the family's socio-economic status.  The counselor ruled out that emotional problems due to stress at home or at school was not primarily responsible for the child's poor school performance. The Wechsler Intelligence Scale for Children-Revised (Indian adaptation by MC Bhatt) was employed by the clinical psychologist to determine that the child's global intelligence quotient score was an average or above average (≥85). 
SpLD was diagnosed based on a curriculum-based assessment, which is a recommended method for its diagnosis. ,,,, Employing a locally-developed and validated curriculum-based test, the special educator conducted the educational assessment in specific areas of learning, namely, basic learning skills, reading comprehension, oral expression, listening comprehension, written expression, mathematical calculation, and mathematical reasoning.  Based on this test, an academic underachievement of up to 2 years below the child's actual school grade placement or chronological age led to a diagnosis of SpLD. ,,
The diagnosis of co-occurring attention-deficit/hyperactivity disorder (ADHD) or any psychiatric diagnosis, if any, was made by the pediatrician and confirmed by the psychiatrist by ascertaining that the child's specific behaviors met the diagnostic and statistical manual of mental disorders-IV-revised criteria. 
Sample size and patient enrolment
The prevalence of SpLD in Indian children is not known.  It was assumed that 10% of Indian children have SpLD. With a 95% confidence level and 5% confidence interval, the sample size (namely, the number of children that need to be interviewed) was calculated using the standard formula:
[Z=1.96 (Z value for 95% confidence level); P=0.1 (10% prevalence, expressed as decimal); C=0.05 (confidence interval, expressed as decimal)]. The sample size calculated was 138 children.
This cross-sectional questionnaire-based study was conducted in our Learning Disability clinic from February 2008 to December 2008. No child, who could understand and read English, declined to participate in the study.
Measuring the self-reported HRQoL
HRQoL was assessed using the English self-report long-version of the DISABKIDS chronic generic module instrument, namely the DCGM-37-S.  The children were asked to complete the questionnaire during their last visit in the clinic before they were informed about the SpLD diagnoses. Completion of the questionnaires lasted 15 to 25 minutes and took place in a quiet room without the presence of the children's parents.
The DCGM-37-S is an instrument, which has been developed cross-nationally across 7 European Union (EU) countries, namely, Austria, France, Greece, Germany, the Netherlands, Sweden, and the United Kingdom to measure the HRQoL of children and adolescents with any chronic medical condition or disability.  One aim of the developers of the DCGM-37-S was to enable cross-cultural pediatric HRQoL research internationally, namely, in different national and cultural contexts.  The English version of the DCGM-37-S can be used in any country for children aged between 8 to 16 years who can read and understand English.  The DCGM-37-S offers the following 6 facets or sub-scales [Table 1]: "independence" (autonomy and living without impairments), "emotion" (emotional worries and concerns), "social inclusion" (acceptance of others, positive relationships), "social exclusion" (stigma, feeling left out), "limitation" (functional limitations, perceived health), and "treatment" (impact of receiving treatment). These 6 facets are associated with 3 HRQoL domains, namely: Mental, social, and physical [Table 1]. All 37 items have a 5-point Likert-type response scale with options ranging from 1 (never), to 5 (always). The child used the recommended 4-week recall period for scoring all 37 items. Each item question is simple and easy to understand. Following the recommendations of the developers of the DISABKIDS, we offered an assistance to read out the question and explain before a child marked his / her response. 
For data analyzes, item raw scores of each facet of the DCGM-37-S were summed and transformed into a score from 0 (worst possible HRQoL) to 100 (best possible HRQoL).  Additionally, a "total score" was calculated combining the scores of the 6 facets representing a "general score" for HRQoL. 
Satisfactory internal consistency has been reported for all the 6 facets with Cronbach alpha coefficient ranging from 0.70 to 0.87 while the total score displayed a consistency coefficient of α=0.93.  Pilot testing has also revealed satisfactory results concerning validity of the DCGM-37-S instrument. 
An ethical approval for the study was granted by the scientific and ethics committees of our institution. The study was conducted in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki. Written informed consent was obtained from the parent or guardian of all participating children, and all children gave a written assent to participate in the study.
The data were analyzed using the Statistical Package for Social Sciences, Version 15.0 for Windows (Chicago, USA). First, the mean facet scores and total score of the study children were computed using SPSS files provided by the developers of the DCGM-37-S.  Since population norms for Indian children with chronic medical conditions or disabilities are not available, mean scores of the study children were compared with the published EU children norms, as recommended.  These EU norms have been computed from the results of the DISABKIDS field study, wherein the HRQoL of 1,152 children and adolescents with 7 different chronic conditions or disabilities (namely, asthma, juvenile arthritis, atopic dermatitis, diabetes, cerebral palsy, cystic fibrosis, and epilepsy) were measured.  In order to investigate the clinical importance of potential differences between the sample of our study and the EU norm data, we used effect sizes, which were calculated by dividing their mean difference in scores by the EU children's SD.  Cohen's guidelines were used for interpretation of effect size (0.2 to <0.5=small, 0.5 to <0.8=medium, and ≥0.8=large).  Second, in order to investigate the reliability of the DSGM-37-S, internal consistencies were calculated for the each facet of HRQoL and for the total HRQoL score. Third, the correlation coefficients (as measured by Pearson r) between the facet scores and total score of the study children were computed. Fourth, the independent samples t-test or the one-way analysis of variance test were used, as applicable, to evaluate the impact of each of the clinical and socio-demographic variables on the facet scores and total score of the DCGM-37-S. Furthermore, using the ordinal regression method, multivariate analysis was performed for determining the "independent" impact that each of the clinical and socio-demographic ("categorical") variables had on a poor DCGM-37-S facet score "outcome" and on a poor total score "outcome." Accordingly, each DCGM-37-S facet score and the total score were dichotomized into "poor" score (score of ≤mean −1 SD) or "good" score (score of >mean −1 SD), and used as "dependent variables" in the models.
Wherever appropriate, the odds ratio (OR) was calculated, and 95% confidence intervals (CI) were estimated around the OR. Statistical significance was considered as P<0.05 (two-tailed).
Characteristics of the study children
A number of 150 children were included in the study. Thus, the power of analysis for the present study was >95%. The mean age of study participants was 12.5 years (SD 2.2, range 8.00-16.00). The boys:girls ratio was 4.2:1 [Table 2]. Their mean global intelligence quotient score was 96.4 (SD 9.0, range 85-123). All 150 children were studying in schools situated in the city of Mumbai and in whom the medium of instruction was the English language [Table 2]. They were well-versed with the English language and did not have any language barrier. 110 (73.3%) parents reported that they had been told by a teacher or a doctor that their child had one or more associated 'non-academic' problem(s), such as, behavioral problems (aggressive behavior, temper tantrums, and stubbornness), anxiety problems, depressive thoughts or recurrent upper respiratory tract infections [Table 2]. Of the 29 (19.3%) children that had co-occurring ADHD; 14 had ADHD, combined type (ADHD-C); 8 had ADHD, primarily of the inattentive type (ADHD-I); and 7 had ADHD, primarily of the hyperactive-impulsive type (ADHD-HI). 1 child with ADHD-C had co-morbid conduct disorder. Of the 29 children with co-occurring ADHD, in 15, the co-morbidity was newly diagnosed; and in the remaining 14 (9.3%), the co-morbidity had been diagnosed before referral, and they were already on medication (either methylphenidate or atomoxetine).
Reliability of DCGM-37-S scores of study subjects
Internal consistencies of HRQoL facets due to the DCGM-37-S were ranging between "emotion," (alpha=0.77); "treatment," (alpha=0.76); "social exclusion," (alpha=0.69); "social inclusion," (alpha=0.62); "independence," (alpha=0.46) and "limitation," (alpha=0.38); while the "total score" displayed a consistency coefficient of alpha=0.83. Testing for reliability (internal consistency) involves estimating how consistently the study subjects have responded to the items within a facet.  Where items within a facet measure different elements of patient experience (as in the multi-dimensional DCGM-37-S instrument), a moderate Cronbach alpha (i.e., approximately 0.5), rather than a high alpha (i.e. ≥0.7), can be considered satisfactory. 
Comparison of HRQoL of study subjects with EU children norms
There were no missing data for the DCGM-37-S items. All the 6 DCGM-37-S mean facet scores and the total score were lower in the study subjects compared to the EU children norms [Table 3]. Calculated effect sizes indicated clinically relevant deficits of study subjects in all HRQoL facet scores as well as in the total score [Table 3].
Correlations between DCGM-37-S facet and total scores
[Table 4] shows the correlations between the DCGM-37-S facet and total scores for the whole sample. These can be used as another test of the convergent and divergent validity of the constructs.  There was a moderately strong relationship between the independence and social inclusion facets and the "total or overall" HRQoL; and a highly strong relationship between the emotion, social exclusion, limitation and treatment facets and the "total or overall" HRQoL, indicating a good convergent validity for these constructs.
Impact of clinical and socio-demographic variables on HRQOL
At the univariate level, child age being 8 to 12 years, child being in school class standard III-VII, and child not having experienced class detainment were significantly associated with a "lower" "emotion" facet score (F=7.004, df=1, 148, P=0.009; F=0.689, df=1, 148, P=0.016; and mean difference: −3.63, df=148, 95% CI: -5.49 to −1.77, P=0.0002, respectively). Child not having any associated non-academic problems(s) was significantly associated with a "lower" "social inclusion" facet score (F=1.367, df=1, 148, P=0.047). Child not having experienced class detainment was significantly associated with a "lower" "social exclusion" facet score (mean difference: −2.56, df=148, 95% CI: -4.13 to −0.98, P=0.002). Child age being 8 to 12 years, child being in school class standard III-VII, and child not having co-occurring ADHD were significantly associated with a "lower" "limitation" facet score (F=7.297, df=1, 148, P=0.008; F=7.194, df=1, 148, P=0.026; and mean difference: -6.10, df=148, 95% CI: −11.70 to −0.50, P=0.033, respectively). Child gender being female, child being in school class standard III-VII, and child not having co-occurring ADHD were significantly associated with a "lower" "total score" (mean difference: -5.96, df=148, 95% CI: −11.65 to −0.27, P=0.040; F=6.000, df=1, 148, P=0.015, and mean difference: -8.22, df=148, 95% CI: −13.84 to −2.60, P=0.004, respectively).
Multivariate analysis [Table 5]a-g revealed that one variable, viz. child not belonging to the upper socio-economic strata of society "independently" predicted a "poor" "independence" facet score of ≤41.19 (P=0.010, OR=1.99, 95% CI: 1.18 to 3.37) [Table 5]a; one variable, viz. child not having experienced class detainment "independently" predicted a "poor" "emotion" facet score of ≤3.29 (P=0.008, OR=3.04, 95% CI: 1.34 to 6.85) [Table 5]b; one variable, viz. child being female "independently" predicted a "poor" "social exclusion" facet score of ≤4.09 (P=0.024, OR=0.28, 95% CI: 0.09 to 0.85) [Table 5]d; one variable, viz. child being first born "independently" predicted a "poor" "limitation" facet score of ≤13.81 (P=0.022, OR=2.60, 95% CI: 1.15 to 5.90) [Table 5]e; and child being female "independently" predicted a "poor" "total score" of ≤28.23 (P=0.025, OR=0.32, 95% CI: 0.12 to 0.87) [Table 5]g.
In support of our hypothesis, the present study confirms that the self-reported HRQoL of children with newly-diagnosed SpLD is significantly compromised in both psychosocial and physical facets of health (social exclusion > emotion > limitation > treatment > independence > social inclusion). Our study documents that by the time their disability is diagnosed, (i) these children perceive themselves as being "socially excluded,0" viz., they feel different from their peers, are lonely, perceive being stigmatized by their teachers/peers, have problems concentrating at school and feel left out; (ii) they have already developed " emotional reactions," viz., worries, concerns, anger, and problems; (iii) they perceive themselves as being " physically limited," viz., having limitations in performing physical activities, having a poor health status, and have difficulties with sleeping; (iv) those children who are already on medication for their co-morbid ADHD perceive their " treatment" to be bothersome, hate taking medication, and worry about the medication; (v) they lack " independence", viz., they are insecure about their future and unable to live an autonomous life; and (vi) they already perceive themselves as lacking in qualities for " social inclusiveness," viz., they feel that their peers and friends do not enjoy their company, do not understand their problems, do not care about their condition, and find it difficult to develop social relationships. Thus, by the time their disability is diagnosed, the "overall health" of these children is already significantly compromised. Also, in the present study, multivariate analysis demonstrated that: (i) not belonging to the upper socio-economic strata of society was an independent predictor of a poor "independence" facet outcome; (ii) not having experienced class detainment (grade retention) was an independent predictor of a poor "emotion" facet outcome; (iii) first-born status was an independent predictor of a poor "limitation" facet outcome; and (iv) female gender was an independent predictor of a poor "social exclusion" facet outcome and a poor "overall health" outcome.
To our knowledge, the present study is the second to analyze the "self-perceived" HRQoL of children with newly-diagnosed SpLD. A recently published study from Greece has analyzed the HRQoL of 116 children with newly-diagnosed SpLD (mean age 10.7 years; SD 2.0, range 8.00-14.00) using the KINDL R , another reliable and validated generic instrument.  This Greek study has also documented that, children with newly-diagnosed SpLD report poorer emotional well-being, lower self-esteem, and deficits in their relationships with family and friends and in their physical functioning.  An Austrian study has analyzed the self-perceived HRQoL of all school children, in general, who were studying in class standards III and IV by employing the Pediatric Quality of Life Inventory™.  In this Austrian study, on multivariate regression analyzes, presence of associated dyslexia "independently" predicted low psychosocial, physical, and global HRQoL scores. 
We cannot really compare our results with the earlier Indian study wherein the parental perceptions of the HRQoL of children with newly-diagnosed SpLD was measured and that too with a different instrument, viz., the Child Health Questionnaire-Parent Form 50 (CHQ-PF50).  The earlier Indian study had detected clinically significant deficits (effect size ≥ -0.5) in 9 out of 12 domains: Limitations in family activities, emotional impact on parents, social limitations as a result of emotional-behavioral problems, time impact on parents, general behavior, physical functioning, social limitations as a result of physical health, general health perceptions and mental health; and in both summary scores (psychosocial > physical).  Their results reported that parents perceive that these children very often exhibit an aggressive, immature, delinquent behaviour; often have feelings of anxiety and depression; and have severe limitations in school work or in activities with their friends as a result of their emotional and/or behavioural problems.  These children also have poor physical health, which leads to their having lots of limitations in their physical activities, including self-care; and in doing school work or in activities with their friends.  These children's psychosocial and physical problems not only have a significant impact on their parent's emotional health and personal time but also very often limit and interrupt activities or generate tension for the entire family.  The earlier study also documented that the 'overall' health of these children is significantly compromised with their psychosocial health being adversely affected more than their physical health. 
What is the utility of the present study? First; our study has revealed that children with newly-diagnosed SpLD perceive having clinically significant deficits in their social exclusion, emotion, limitation, independence, and social inclusion facets of health. The present study highlights the importance of optimally treating these children to improve their HRQoL. Only initiating remedial education for alleviating the disability would not suffice to rehabilitate these children. Mental health professionals, educators, and school counselors should plan targeted support interventions for these children to equip them with skills to cope with their emotional experiences and deal with their negative feelings and stressful events in their daily school functioning. These interventions should also aim to improve the knowledge of classroom teachers, classmates, and family members about SpLD to help reduce the unpleasant experiences these children undergo, both at school and at home. Also, children with co-morbid ADHD who need medication should be better counseled about the rationale and benefits of their medication. Second; our study has identified the significant clinical and socio-demographic variables, which impact the HRQoL of these children, and which need to be addressed by counselors right at the time their disability is diagnosed. In our country, most parents from the middle socio-economic strata of society have huge expectations from their child to achieve good school grades. These parents consider achieving good school grades as the single most important factor that would help their child have a bright future and climb-up socially. Such parents would be unable to cope-up with their child's academic underperformance. They would react by making their child study more hours and by scolding them repeatedly for their poor school performance. This may be an explanation for "not belonging to the upper socio-economic strata of society" being an independent predictor of a poor "independence" facet outcome, viz. for these children to be significantly more insecure about their future and unable to live an autonomous life. In our society, the first-born child is often over-protected by parents. Generally, parents monitor and supervise the daily activities and health concerns of their first-born child more closely than required. Parental overindulgence along with their poor self-perceived HRQoL may be responsible for "first-born" children perceiving significantly more limitations in performing physical activities, perceiving a poor health status, and finding difficulties with sleeping. In general, girls in our society are more sensitive and reticent than boys. This would explain why "female gender" was an independent predictor of a poor "social exclusion" facet outcome, viz. for girls reporting feeling significantly more different from peers, lonelier, perceiving being stigmatized by teachers / peers and feeling left out; and a poor "overall health".
We have no proper explanation for "why not having experienced class detainment" was an independent predictor of a poor "emotion" facet outcome. It is known that class detainment may cause severe emotional stress and lead to behavioral problems in children.  It could be argued that in the present study, children who had already experienced class detainment, were protecting themselves from a stressful recognition of it, or they may have adjusted or just given up worrying about their academic problems.
Our study has its limitations. First; the mean age of diagnosis of SpLD in the study sample was 12.5 years, which is quite late. Hence, the present study results may not reflect the HRQoL of children in whom SpLD is diagnosed on time, viz., by 8 to 9 years of age. Second; as vernacular (Hindi or Marathi or Gujarati) versions of the DCGM-37-S instrument are not available, non-English-speaking children were excluded from the study. However, our numbers of these children were very small (<10). Third; children from the lower socio-economic strata of society were very few in our study population. Either their school authorities were not aware about SpLD, or their parents were not motivated enough to bring them to our clinic for an assessment. However, we do not believe that these limitations adversely affect the utility of our results. Both due to the limitations as outlined above and the general paucity of data on HRQoL of children having SpLD, future researchers should investigate whether the present study's results can be generalized to the population level.
The authors thank our Dean Dr. Sandhya Kamath for granting us permission to publish this paper; all the children and parents who participated in this study; and Professor D. P. Singh, Department of Research Methodology, Tata Institute of Social Sciences, Deonar, Mumbai, for his help in the statistical analysis of the data. We also thank Professor Dr. Monika Bullinger, DISABKIDS Project Coordinator, Department of Medical Psychology, University Hospital of Hamburg-Eppendorf, Hamburg, Germany for providing us the DISABKIDS manual free of cost and granting us permission to use the DCGM-37-S instrument. The material in this publication is the result of use of the DCGM-37-S instrument and the assistance of the EUROPEAN DISABKIDS GROUP is gratefully acknowledged.
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5]