Article Text

Download PDFPDF

Attention-deficit/hyperactivity disorder: variation by socioeconomic deprivation
  1. Vibhore Prasad1,
  2. Joe West2,
  3. Denise Kendrick3,
  4. Kapil Sayal4
  1. 1 School of Population Health & Environmental Sciences, King’s College London, London, UK
  2. 2 School of Medicine, Nottingham City Hospital, University of Nottingham, Nottingham, UK
  3. 3 School of Medicine, University of Nottingham, Nottingham, UK
  4. 4 School of Medicine, Developmental Psychiatry, Queen’s Medical Centre, University of Nottingham, Nottingham, UK
  1. Correspondence to Dr Vibhore Prasad, School of Population Health & Environmental Sciences, King’s College London, London, SE1 1UL, UK; vibhore.prasad{at}kcl.ac.uk

Abstract

Background In England, there is a discrepancy between the prevalence of attention-deficit/hyperactivity disorder (ADHD) ascertained from medical records and community surveys. There is also a lack of data on variation in recorded prevalence by deprivation and geographical region; information that is important for service development and commissioning.

Methods Cohort study using data from the Clinical Practice Research Datalink comprising 5196 children and young people aged 3–17 years with ADHD and 490 016 without, in 2012.

Results In 2012, the recorded prevalence of ADHD was 1.06% (95% CI 1.03 to 1.09). Prevalence in the most deprived areas was double that of the least deprived areas (prevalence rate ratio 2.58 (95% CI 2.36 to 2.83)), with a linear trend from least to most deprived areas across all regions in England.

Conclusions The low prevalence of ADHD in medical records may indicate considerable underdiagnosis. Higher rates in more disadvantaged areas indicate greater need for services in those areas.

  • child psychiatry
  • comm child health
  • epidemiology

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

What is already known?

  • Attention-deficit/hyperactivity disorder (ADHD) affects 3%–5% of children and young people in the community in the UK.

  • There is a discrepancy between the community prevalence of ADHD and the clinically recorded prevalence, which is <1%.

  • Estimates of how the clinically recorded prevalence of ADHD varies by deprivation and region are lacking.

What this study adds?

  • The clinically recorded prevalence of ADHD in children and young people was twice as high in the most compared with the least deprived areas.

  • The greatest inequality in recorded prevalence of ADHD was in the East of England and the least inequality was in London.

  • There is a greater need for health and educational services for children with ADHD in more disadvantaged areas.

Introduction

The community prevalence of attention-deficit/hyperactivity disorder (ADHD) is 3%–5% in England.1 However, the clinically recorded prevalence is much lower (<1%).2 3 Estimates of how this varies by deprivation and region are lacking. This cohort study aims to address this gap and inform development of health and education services.

Methods

Data source

We used the general practice (GP) medical records linked to hospital episodes statistics (HES) data from the Clinical Practice Research Datalink (CPRD) from 370 GP practices, which represents approximately 4% of the UK population.4 These data are broadly representative of the UK population and contain information on consultations with GPs, hospital admissions, diagnoses and prescriptions.4

Defining the population with ADHD

We extracted medical records for children and young people (CYP) aged 3–17 years registered before June 2013 with at least one diagnosis code or prescription for ADHD. We took the latest of the date of: third birthday; diagnosis; registration with the practice (in CYP diagnosed before registration) or 1 January 1998 (the first full year of the CPRD-HES link) as the date when ADHD was first known to the GP. We took the earliest of the date when the: CYP left the practice or died; practice stopped participating in the CPRD; CYP turned 18 years or 31 December 2012 (the last complete year that CPRD-HES linked data were available), as the last date of follow-up.

Estimating prevalence—numerators

To calculate the frequency of recorded ADHD in 2012, CYP were required to have received the diagnosis on or prior to 1 July and to be registered with the GP on 1 July. The number of CYP known to have ADHD by 1 July in 2012 was counted overall and by: age, sex, strategic health authority region and social deprivation quintile (English index of multiple deprivation (IMD) score 2010, at lower super output area level, based on home postcode). The IMD score comprises seven domains: income, employment, health and disability, barriers to housing and services, living environment and crime.

Denominators

We extracted medical records from the CPRD for CYP registered before 31 December 2012. CYP who were: aged from 3 to 17 years; registered with the practice and alive, between 1 January 2012 and 31 December 2012 were counted in the denominator. The number of CYP registered in the CPRD on 1July in 2012 was counted overall and by age, sex, region and deprivation.

Prevalence estimates

Prevalence rates for 2012 were estimated assuming a Poisson distribution and described overall by sex, age, region and deprivation. We assessed whether prevalence by age varied by sex and whether deprivation gradients varied by region by adding interaction terms to the model assessing significance using a likelihood ratio test (LRT).

Sensitivity analyses and subgroup analyses

To explore how altering the definition of ADHD affects estimates of recorded prevalence, we described subgroups reflecting at least one drug code or at least one diagnosis code; at least two drug codes and at least two diagnosis codes; at least one drug code; no drug codes.

Results

There were 5196 CYP with ADHD and 490 016 without ADHD (table 1). In 2012, the recorded prevalence of ADHD was 1.06% (95% CI 1.03 to 1.09). Boys had a fivefold higher prevalence than girls (1.74% vs 0.35%, prevalence rate ratio (PRR) 4.98 (95% CI 4.62 to 5.36). The prevalence was highest in those aged 15–17 years (1.91 (95% CI 1.82 to 1.99) and lowest in those aged 3–4 years (0.01 (95% CI 0.002 to 0.02). The relationship between prevalence and age did not differ significantly by sex (LRT, P=0.09). Prevalence rates increased with increasing deprivation, being twofold higher in CYP from the most compared with the least deprived areas (1.38% vs 0.73%, PRR 2.58 (95% CI 2.36 to 2.83); test for linear trend P<0.001). There was considerable geographic variation, with higher prevalence in the South East and East regions, compared with Yorkshire and Humber (1.55% and 1.34% vs 0.56%, PRR 3.13 (95% CI 2.46 to 3.99) and 2.80 (95% CI 2.19 to 2.00)), respectively. Varying definitions of ADHD did not alter prevalence patterns by sex, age, deprivation or region (see online supplementary table).

Table 1

The recorded prevalence, crude and adjusted prevalence rate ratio (PRR) of ADHD in 2012

The social gradient in ADHD prevalence was evident within all regions and also varied significantly between regions (figure 1, test for interaction P<0.001). The steepest social gradient (comparing most to least deprived areas) was in the East of England (PRR 3.43 (95% CI 2.70 to 4.37)) and the smallest gradient in London (PRR 1.38 (95% CI 1.04 to 1.83)).

Figure 1

Prevalence of attention-deficit/hyperactivity disorder by area-level deprivation for each region.

Discussion

The recorded prevalence of ADHD in CYP (1.06%) in 2012 was considerably lower than community prevalence estimates.1 Recorded prevalence estimates worldwide vary from 0.06% to 13%, with higher estimates originating from the USA.5 The prevalence was double in the most compared with the least deprived areas with a linear trend across all regions in England. The greatest inequality was within the East of England region and the least within London.

Our low prevalence estimate suggests that there are many CYP in the population with undiagnosed ADHD. This is supported by findings from the British Child and Adolescent Mental Health Survey,1 which reported ADHD prevalence among those aged 5–15 years of 2.23%. Underdiagnosis is important as it precludes receipt of appropriate child and parental support, educational support and behavioural and pharmacological treatment for ADHD. This may be particularly relevant as children transition from primary to secondary school and independent learning is increasingly required. As recorded prevalence of ADHD increases with age, and we included children aged under 5 years, this may partly explain our low prevalence estimate. It is also possible that secondary care ADHD diagnoses are not being recorded in primary care records. However, systematic reviews demonstrate accurate recording of secondary care diagnoses in primary care records and high validity across a wide range of diagnoses, so this is unlikely to explain much of the difference in prevalence rates.6 7

Worldwide studies of community prevalence of ADHD suggest CYP from disadvantaged families are 1.5 to 4 times more likely to have ADHD symptoms compared with those from more advantaged families.8 This is consistent with Millennium Cohort Study (MCS) findings of strong associations between multiple measures of socioeconomic disadvantage and parent-reported diagnosed prevalence of ADHD.9 This may represent true differences in the prevalence of ADHD, differential symptom reporting or differential access to services to diagnose and treat ADHD. The MCS found similar associations between parent-reported or teacher-reported ADHD symptoms and socioeconomic disadvantage, suggesting that clinical labelling bias does not explain the social gradient.9 Potential explanations for the social gradient include differential exposures to multiple material, psychosocial and environmental risk factors before or around the time of birth or in childhood, as well as genetic or developmental contributions.9 Genetic susceptibility may also play a role, with some individuals being more susceptible to adverse, or supportive, environmental exposures.10

Our findings suggest greater need for health and educational services for CYP with ADHD in more disadvantaged areas and can inform the development and commissioning of appropriate services, with our figures being understood as conservative estimates. Future research is needed to explore underdiagnosis or under-recording of ADHD in CYP and mechanisms by which socioeconomic disadvantage impacts on ADHD prevalence.

References

Footnotes

  • Contributors VP: conceived the idea for the study, conducted the data management, analysis and interpretation, drafted the initial manuscript and approved the final manuscript as submitted. DK: conceived the idea for the study, provided clinical input and interpretation throughout the project, critically reviewed and approved the final manuscript as submitted. KS: made contributions to the design of the study, provided clinical input and interpretation throughout the project, critically reviewed and approved the final draft of the manuscript. JW: conceived the idea for the study, guided the data management and analysis, provided clinical input and interpretation throughout the project, critically reviewed and approved the final draft of the manuscript.

  • Funding This study was supported by a National Institute for Health Research (NIHR) grant, DRF- 2011-04-116.

  • Disclaimer The views expressed are those of the author(s) and not necessarily those of the NHS, the NIHR or the Department of Health.

  • Competing interests VP reported having received research grant support administered via the University of Nottingham from the NIHR Doctoral Research Fellowship scheme. During the period of the NIHR award for VP, JW was supported by a University of Nottingham/Nottingham University Hospitals National Health Service (NHS) Senior Clinical Research Fellowship.

  • Patient consent Not required.

  • Ethics approval Clinical Practice Research Datalink (CPRD) independent scientific advisory committee (protocol reference 12_128R).

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Data sharing statement All data are available by application to the Clinical Practice Research Datalink. Further information is available on their website: https://www.cprd.com/intro.asp.