Attention deficit hyperactivity disorder (ADHD) is one of the most common neurodevelopmental disorders in children. Its occurrence and pattern of presentation are unknown in Botswana.
To determine the prevalence of attention-deficit hyperactivity disorder (ADHD), associated comorbid conditions and risk factors amongst school-age children in Botswana.
Primary schools in Gaborone, Botswana.
This study used a cross-sectional design. A two-stage random sampling technique was utilised to select learners from 25 out of the 29 public schools in the city. The Vanderbilt ADHD Diagnostic Rating Scale (VADRS), teacher and parent versions, was administered.
Of the 1737 children, 50.9% (
The prevalence of ADHD in Botswana is slightly higher than that reported in the literature, but the pattern of presentations and comorbidities is similar. A positive family history of mental illness and perinatal complications independently predicted ADHD. Mental health screening amongst families of the affected individuals and improved perinatal care should be considered as health care priorities in Botswana.
Attention-deficit hyperactivity disorder (ADHD) is one of the most common neurodevelopmental disorders in childhood and affects 3% – 12% of primary school children.
It is associated with an increased risk of low self-esteem, poor interpersonal relationships, poor school performance, conduct problems, criminality, substance abuse and sexual promiscuity, which may increase the risk of human immunodeficiency virus (HIV) transmission.
Attention-deficit hyperactivity disorder is characterised by a pattern of reduced sustained attention and a higher level of activity in children or adolescents than expected for someone of that age and developmental level.
Attention-deficit hyperactivity disorder is more common in males, but the rate varies amongst the presentations.
Epidemiological studies had reported varying prevalence rates across cultures and geographical regions of the world, with worldwide prevalence rates of as low as 0.5% to 20%.
This was a cross-sectional study, which involved parent and teacher survey of primary school pupils, aged 6–12 years. The study was conducted in Gaborone, the capital city of Botswana. Gaborone is the most densely populated city, accounting for approximately 10% of Botswana’s total population (Botswana Population Census Atlas 2011). It is also representative of the reference population regarding the number of primary schools and ethnic distribution.
Two hundred pupils were selected from seven regions in Gaborone to constitute 1400 participants, by using a two-stage sampling technique. Nonetheless, we increased the sample by 30%, anticipating a 30% attrition rate, which was calculated from a pilot study conducted earlier. A sample size of 1820 pupils was thus targeted.
The data collection was carried out in two phases. The teachers were instructed on how to administer the Vanderbilt ADHD Diagnostic Rating Scale-Teacher Version (VADRS-TV) on the pupils whose parents or guardians provided written consent. Only the teachers who have been with the pupils for a minimum of 6 months were interviewed. The research assistants then followed up with parent interviews only for those who met the criteria for ADHD on the teacher version. The parents were instructed on how to rate their children on the parent version of the Vanderbilt ADHD Diagnostic Rating Scale-Parent Version (VADRS-PV). The research assistants helped those unable to read or write by reading out the questions and filling their responses on the empty questionnaires. Only the agreement between the two ratings was regarded as diagnostic of ADHD in this study based on DSM-5 diagnostic criteria of symptoms in two settings.
Vanderbilt ADHD Diagnostic Rating Scale
Data were analysed by using the Statistical Package for Social Sciences (SPSS) for Windows, version 16.0. The descriptive statistical tools such as frequencies (%), mean and s.d. were used to present the prevalence of ADHD and other demographics. Continuous variables such as age were dichotomised by using the calculated mean as the cut-off. Pearson’s Chi-square tests were used to calculate differences between groups, such as the presence of ADHD, the outcome variable and gender. The relationship between ADHD and comorbid disorders, such as ODD and CD, was also tested by using Chi-square tests. A post hoc analysis was performed on those who were positive by the teachers’ rating. The relationship between identified socio-demographic risk factors and ADHD was explored using Chi-square tests and logistic regression. All tests were two-tailed, and a
Permission to embark on the study was obtained from the Ethics and Research Committee of the University of Botswana (UBR/RES/IRB/BIO/016) and the Ministry of Basic Education (DPRS 7/1/5 XXX (18) PAO-Research). Letters were sent to the schools to inform the teachers and parents or guardians about the study. Their consents to participate were also sought in writing.
Of the 1820 VADRS-TV distributed to the teachers who consented, 1799 (98.8%) were retrieved after the first phase of the study, and 420 (23.3%) children met the criteria for ADHD. The parents of these 420 participants were subsequently interviewed with VADRS-PV, and 402 parents returned the questionnaires. The questionnaires of those whose parents refused to participate and those with incomplete responses in the variables of interest were excluded. Hence, 1737 (92.4%) questionnaires were finally analysed.
There were slightly more male children than female children (50.9% vs. 49.1%), and the mean age of the participants was 9.53 (s.d. = 1.97) years, with a range from 6 to 12 years.
The prevalence of ADHD amongst 1737 school-age children according to the teachers’ rating was 23.1% (
Of the 213 patients who met the criteria for ADHD on the teacher and parent scales, 113 (53.1%) had at least one comorbid psychiatric disorder. The most common psychiatric disorder amongst those who met the criteria was ODD (
A post hoc analysis of the 402 children with parent and teacher ratings was conducted. The age of the pupils, their parents and the family size were dichotomised by using the calculated means for the bivariate analysis.
On bivariate analysis, there was a significant association between male gender and ADHD (
The association between the risk factors and attention-deficit hyperactivity disorder amongst those rated by both parent and teacher (
Variables | ADHD |
Statistics |
|||||
---|---|---|---|---|---|---|---|
Present |
Absent |
||||||
1.11 | 1 | 0.292 | |||||
9 years and older | 66 | 49.6 | 67 | 50.4 | |||
Below 9 years | 117 | 55.5 | 94 | 44.5 | |||
10.6 | 1 | 0.001 |
|||||
Male | 142 | 59.7 | 96 | 40.3 | |||
Female | 65 | 42.8 | 87 | 57.2 | |||
0.46 | 1 | 0.497 | |||||
Below 40 years | 71 | 55.0 | 58 | 45.0 | |||
40 years and older | 64 | 50.8 | 62 | 49.2 | |||
0.00 | 1 | 0.984 | |||||
34 years and younger | 102 | 53.4 | 89 | 46.6 | |||
Older than 34 years | 105 | 53.3 | 92 | 46.7 | |||
0.61 | 1 | 0.434 | |||||
Four children and less | 169 | 52.8 | 151 | 47.2 | |||
More than four children | 26 | 59.1 | 18 | 40.9 | |||
3.58 | 1 | 0.058 | |||||
Below secondary | 75 | 48.1 | 81 | 51.9 | |||
Secondary school and above |
63 | 60.0 | 42 | 40.0 | |||
2.81 | 1 | 0.094 | |||||
Below secondary | 133 | 50.0 | 133 | 50.0 | |||
Secondary school and above | 80 | 58.8 | 56 | 41.2 | |||
0.25 | 2 | 0.882 | |||||
Not employed | 9 | 50.0 | 9 | 50.0 | |||
Low skilled employment | 97 | 56.1 | 76 | 43.9 | |||
High skilled | 32 | 56.1 | 25 | 43.9 | |||
3.63 | 2 | 0.162 | |||||
Not employed | 47 | 59.5 | 32 | 40.5 | |||
Lower occupational group | 122 | 50.6 | 119 | 49.4 | |||
Upper occupational group | 32 | 62.7 | 19 | 37.3 | |||
1.05 | 1 | 0.305 | |||||
Separated or not married | 184 | 54.8 | 152 | 45.2 | |||
Married and living together | 14 | 45.2 | 17 | 54.8 | |||
9.38 | 1 | 0.002 |
|||||
Absent | 123 | 46.1 | 144 | 53.9 | |||
Present | 15 | 83.3 | 3 | 16.7 | |||
1.89 | 1 | 0.168 | |||||
Absent | 176 | 51.3 | 167 | 48.7 | |||
Present | 28 | 62.2 | 17 | 37.8 | |||
6.78 | 1 | 0.009 |
|||||
Absent | 154 | 50.0 | 154 | 50.0 | |||
Present | 59 | 53.5 | 31 | 34.4 |
Risk factors include socio-demographic and clinical variables.
, Significant
, Minimum of 12 years of formal education; ADHD, attention-deficit hyperactivity disorder.
In addition to the significant variables on bivariate analysis, fathers’ and mothers’ levels of education, which fell short of significance, were entered into the logistic regression analysis. Only a family history of mental illness (odds ratio [OR] = 6.59, 95% CI: 1.36–32.0) and history of perinatal complications (OR = 2.16, 95% CI: 1.08–4.29) emerged as the independent predictors of ADHD (
The regression model of the association between the risk factors
Risk factors | Wald | OR | 95% Confidence interval (CI) |
||
---|---|---|---|---|---|
Lower | Upper | ||||
Male | 2.61 | 0.106 | 1.67 | 0.89 | 3.11 |
Below secondary school | 2.89 | 0.089 | 0.52 | 0.24 | 1.11 |
Below secondary school | 0.08 | 0.776 | 0.89 | 0.42 | 1.93 |
Present | 5.47 | 0.019 |
6.59 | 1.36 | 32.0 |
Present | 4.79 | 0.029 |
2.16 | 1.08 | 4.29 |
, Significant
, Risk factors include socio-demographic and clinical variables.
The main finding of this study was the 12.2% prevalence of ADHD based on the agreement between teachers and parents on the Vanderbilt Rating Scale. Other key findings in this study included male predominance in those who met the criteria for ADHD and the independent association between ADHD and clinical variables such as a positive family history of mental illness as well as history of perinatal complications at birth.
The ADHD prevalence rate in this current study is comparable with 11% reported in a study in the USA,
In the United Kingdom, more stringent ICD-10 criteria are often preferred to DSM criteria, and doctors working in the United Kingdom are less likely to make a diagnosis of ADHD compared with those in the United States of America.
Even though the variations in rates across nations have mainly been attributed to methodological differences,
Notwithstanding that the rate of ADHD in Botswana is slightly higher than reports from most literature,
Evidence suggests that more than 50% of those who have ADHD have a comorbid psychiatric disorder.
A gender difference was observed in the prevalence of ADHD in the present study. Boys were 2.2 times more likely to have rated positive for ADHD than girls, and this is comparable with the pooled prevalence of 2.45 for boys from a systematic review.
The post hoc analysis revealed a significant association between gender and ADHD in this study. Whilst the initial analysis suggested that gender plays a role in the development of this disorder, the logistic regression analysis proved otherwise. The influence of other variables such as a history of mental illness in the family included in the analysis may have accounted for the significant relationship on bivariate analysis. For instance, gender-specific vulnerability to non-familial risk factors has been shown to mediate the onset of ADHD in males.
In the present study, a positive family history of mental illness was associated with ADHD on logistic regression analysis. The fact that ADHD coexists with other psychiatric disorders, as shown in this study, is not a coincidence. Attention-deficit hyperactivity disorder has been genetically linked to various psychiatric disorders. High proportions of psychiatric disorders such as ADHD and CD have been found amongst relatives of individuals with schizophrenia and schizotypal personality disorder.
Other possible reasons for this association include a lower threshold for behavioural problems in children with ill family members
The complexity observed in the inheritance of ADHD and that it does not follow the Mendelian law of segregation imply an interaction between multiple genes and environmental risk factors as alluded to previously.
In this study, the identified environmental factor associated with ADHD was a history of perinatal complications. This variable was not only found to be associated with ADHD but also suggested that those exposed to perinatal complications have two times the risk of developing ADHD. The current finding is consistent with a previous work, linking perinatal complication with later development of externalising disorders such as ADHD and CD.
The need to improve on perinatal care and discourage unplanned home delivery cannot be overstressed in this community, especially that an earlier study in Botswana had suggested a link between perinatal complications and childhood mental disorders.
The present study was conducted in one of the 17 districts in Botswana; thus, generalising the findings to other parts of the country remains hypothetical and should be interpreted with caution. However, the study location has the densest population and is representative of the reference population regarding ethnic distribution.
The present study is limited by the use of parents’ and teachers’ ratings without a clinical diagnostic interview or a direct assessment of the children, to confirm the diagnosis. Nonetheless, the tool used was based on DSM IV-TR criteria and remains valid as DSM-5 still uses the same criteria, except that the symptoms must be present before age 12 years. The use of a sub-sample to explore the risk factors of ADHD also limits our findings.
The strength of the study lies in the use of information by both the teachers and parents to control for over-reporting. Adherence to the DSM-5 diagnostic criteria, one of which emphasises the need for symptoms in two different settings, adds to the strength of this study. Furthermore, it is essential to note that this was the first study to assess the community prevalence of ADHD in Botswana with relatively modest sample size.
The prevalence rate of ADHD in Botswana is slightly higher than reports from the literature, possibly because of the nature of the instrument used in the present study. The pattern of presentation is nonetheless like the earlier reports, with the predominantly inattentive presentations as the most prevalent and ODD as the commonest comorbid condition. The role of cultural differences in the varying rates of ADHD, especially in the African setting, should be further explored.
Having a positive family history of mental illness and perinatal complications remained the independent predictors of ADHD in the subsample of children with teacher and parent surveys. Hence, mental health screening amongst families of the affected individuals and improved perinatal care should be considered as healthcare priorities in Botswana and other sub-Saharan countries.
We thank all the parents and teachers who participated in the study.
The authors declare that no competing interests exist.
A.O. conceived the idea, analysed the data and prepared the initial manuscript. A.O., S.P., R.O. and O.J. were involved in the writing and editing of the final manuscript. All authors read and approved the final manuscript.
This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.
The data sets used and analysed during the current study are available from the corresponding author on reasonable request.
The views and opinions expressed in this article are those of the authors and do not necessarily reflect the official policy or position of any affiliated agency of the authors.