Prevalence and risk indicators for attention deficit hyperactivity disorder (ADHD) in children at the Aga Khan University Hospital Nairobi

Date of Award


Document Type


Degree Name

Master of Medicine (MMed)

First Supervisor/Advisor

Prof. William macharia

Second Supervisor/Advisor

Dr. Roseline ochieng

Third Supervisor/Advisor

Dr. Frank njenga


Paediatrics and Child Health (East Africa)


Background: ADHD is the most common childhood neurobehavioural disorder. Affected children experience significant functional problems affecting their relationships with family and peers and academic underachievement.

Objectives: The primary objective of this study was to determine prevalence of ADHD in children aged 6-12 years attending Aga Khan University hospital Nairobi Paediatric Casualty. The secondary objective was to ascertain if medically attended physical injury and poor academic performance were risk indicators for ADHD. Additionally, this study set out to determine the diagnostic utility of Vanderbilt Assessment Scale (VAS) compared to DSM-IV in diagnosing ADHD and establish if there is an association between ADHD and co-morbid conditions.

Methods: A cross-sectional study was conducted and study participants were recruited from the paediatric casualty. The Principal Investigator used the DSM-IV diagnostic criteria to make a diagnosis of ADHD and parents thereafter filled out the VAS and risk assessment forms consisting of questions about injury and academic performance. Data collected were analysed using STATA.

Results: A total of 240 children were recruited. Prevalence of ADHD in children aged 6-12 years at the Aga Khan University Hospital Paediatric casualty was found to be at 6.3% (95%CI 3.7-10.3). The odds of repeating a class when a child had ADHD is 20.2 (95%CI4.0-100.4 p<0.001). A total of 72 children had medically attended injuries. Odds of having an injury when a child had ADHD was 2.9 (95%CI 1.0-8.4 p=0.04) and only two types of injuries were found in the children with ADHD namely; burns (67%) and wounds (37%). Co-morbidities for ADHD; Oppositional Defiant Disorder, anxiety, depression and conduct disorders were not significantly associated with ADHD. v VAS had a sensitivity of 66.7% (95%CI 39-87) and specificity of 99% (95%CI 96-99). Positive predictive value was 83% (95%CI 50-97) and negative predictive value 98% (CI 95-99). Likelihood Ratio (LR +) was 75 (95%CI 18.0-311) and Likelihood Ratio (LR-) was 0.3 (95%CI 0.2-0.7).

Conclusion and Recommendations: The ADHD prevalence in this study is similar to global prevalence of 5.29% and other African countries such as Congo and Nigeria where the prevalence was 6% and 8% respectively. Concurrence in prevalence indicates the burden of disease is similar across the three African countries and is comparable to what is happening globally. Consequently, the magnitude of the negative impact ADHD has in the local setting needs to be investigated and compared to other countries to establish its effects on children. Injury and poor academic performance were found to be weakly associated with ADHD despite the low power of this study. Subsequently, it would be recommended that doctors in outpatient settings should ask for patients’ academic history and children noted to have repeated a class should be evaluated for ADHD. Additionally, recommendation of the same would be in regards to children who come with burn injuries. VAS was demonstrated to have poor sensitivity when used in the local setting. Nevertheless, the high specificity of the tool can be used to confirm the clinically suspected cases. The VAS is written in English consequently language barrier became a limitation in this study. Therefore validated translated versions of ADHD screening tools into Swahili would be beneficial in our local setting. Finally, ADHD is a condition that that has good outcomes once diagnosed and managed hence health care providers need to be vigilant in searching for the disorder in children.

This document is available in the relevant AKU library