Use the following Abstract to answer Questions 27 to 36

A population survey of mental health problems in children with epilepsy

The 1999 British Child and Adolescent Mental Health Survey, a nationwide epidemiological study of rates of psychiatric disorder in children aged 5 to 15 years, provided the opportunity to investigate the mental health of children with epilepsy. These children and their families experience disability specifically because of additional emotional, behavioural, and relationship problems, and this is the first epidemiological study that directly measures these impairments. Information was obtained by interviewing a main carer and teacher for 10,316 children; 67 children with epilepsy were identified (35 males, 32 females; mean age 10 years 2 months, SD 2 years 11 months, range 5 to 15 years), and compared with the 47 children with diabetes (27 females, 20 males; mean age 10 years 4 months, SD 3 years 4 months, range 5 to 15 years) and 10,202 controls (50% male; mean age 9 years 11 months, SD 3 years 1 month, range 5 to 15 years). DSM-IV psychiatric diagnoses were derived from the Development and Well-Being Assessment in combination with the interview and a specialist clinician rating. Parental reports of emotional and behavioural problems, their impact, and associated peer problems were also obtained. Rates of psychiatric disorder were 37% (95% confidence interval [CI] 22 to 49) in epilepsy, 11% (95% CI 2 to 19%) in diabetes, and 9% (95% CI 9 to 10%) in control children. Parents of children with epilepsy consistently reported more problems, with greater impact and associated peer problems. Epilepsy, but not diabetes, was independently (adjusted for age, sex, and severe learning difficulties) associated with all behavioural variables in regression analyses. Emotional, behavioural, and relationship difficulties are common in children with epilepsy, and constitute a significant burden to the children and their families, indicating the need for effective mental health services for these children.

Reference: Davies S, Heyman I and Goodman R (2003) 'A population survey of mental health problems in children with epilepsy', Developmental Medicine & Child Neurology, 45:292-295, doi:10.1017/S0012162203000550

Question 27

The study design was:

a)	Cohort.
b)	Randomised controlled trial.
c)	Cross-sectional.
d)	Case-control.

Question 28

What is the most appropriate 'aim' or 'goal' of research for this study?

a)	Investigate causal relationships.
b)	Identify general patterns and relationships.
c)	Advance new theories.
d)	Interpret socially or historically significant phenomena.

Question 29

Epilepsy can be described as a(n) ____ variable

a)	Control.
b)	Dependent.
c)	Confounding.
d)	Independent.

Question 30

Psychiatric disorder can be described as a(n)_____ variable

a)	Control.
b)	Independent.
c)	Dependent.
d)	Confounding.

Question 31

In this study by Davies, Heyman and Goodman (2003) the measure "parental reports of emotional and behavioural problems" represents a(n):

a)	Dependent variable.
b)	Confounding variable.
c)	Independent variable.
d)	Control variable.

Question 32

Parents of the children provided information on the "presence or absence of 34 health problems or conditions". These health problems were measured as a(n):

a)	Ratio variable.
b)	Interval variable.
c)	Ordinal variable.
d)	Nominal variable.

Question 33

Epilepsy was assessed according to the following classification: 'No history of epilepsy', 'uncomplicated epilepsy', or 'complicated epilepsy'. 'Epilepsy' was measured as a(n):

a)	Ordinal variable.
b)	Nominal variable.
c)	Interval variable.
d)	Ratio variable.

Question 34

The rate of any psychiatric disorder was 37% (95% CI 22-49%) in the children with epilepsy, 11% (95% CI 2-19%) in the children with diabetes, and 9% (95% CI 9-10%) in the control children (2 = 61.2, df = 2, p < 0.001). This means that:

a)	There was no significant difference found between children with epilepsy and control children with respect to the prevalence of any psychiatric disorder.
b)	There was no significant difference found between children with epilepsy and children with diabetes with respect to the prevalence of any psychiatric disorder.
c)	There is no evidence that disease status is associated with psychiatric disorder in the population.
d)	There was a significant difference found between children with epilepsy and children with diabetes or control children with respect to prevalence of any psychiatric disorder.

Question 35

In discussing the findings (p.294), the authors noted that the rates of psychiatric disorder were consistent with a study from 30 years earlier, suggesting that the type of assessment did not appear to be a factor in the results. Which type of bias is this most relevant to?

a)	Observation bias.
b)	Attrition bias.
c)	Misclassification bias.
d)	Selection bias.

Question 36

In the methods section (p. 292) the authors also state: "The total number of children included in the sampling frame was 6,422,202, an estimated 90% coverage of the whole population. From this sampling frame, 10,438 children [aged 5 to 15 years] were selected from throughout England, Wales, and Scotland [representing 83% of those approached]." Which of the following aspects is not important when selecting this sample?

a)	The prevalence of psychiatric disorders.
b)	Narrower confidence intervals for results.
c)	Increased generalizability of findings.
d)	Overall reduction in error and bias.