 Advertisement | ||||||||
 | ||||||||
|
||||||||
 |
 |
 |
 |
 |
 |
 |
||
|
From the Department of Sociology, University of Alberta, Edmonton, Alta.
Correspondence to: Dr. Lisa A. Strohschein, Assistant Professor, Department of Sociology, University of Alberta, Rm. 5-21, Tory Building, Edmonton AB T6G 2H4; lisa.strohschein{at}ualberta.ca
 |
Abstract |
|---|
 Top Abstract MethodsResultsInterpretationREFERENCES |
|---|
Methods: I used data for children who participated in the National Longitudinal Survey of Children and Youth from 1994 to 2000. The sample was restricted to children who remained in the survey in 2000 and who, at initial interview, lived in a household with 2 biological parents (n = 4784). A generalized estimating equation model was used to compare the odds ratios of methylphenidate use among children whose parents obtained a divorce between 1994 and 2000 relative to children whose parents remained married during this period.
Results: Between 1994 and 2000, 633 children (13.2%) experienced the divorce of their parents. The proportion of children who received methylphenidate at any time between 1994 and 2000 was 3.3% among those whose parents remained married and 6.1% among those whose parents divorced during this period. After adjustment for age of the mother and sex and age of the child, I found that methylphenidate use was significantly higher among children whose parents subsequently divorced than among those whose parents remained married (odds ratio 1.82, 95% confidence interval 1.01–3.33).
Interpretation: The increased risk of children receiving a prescription for methylphenidate in the period following parental divorce raises questions about the causal links in this association. Future research is needed to replicate these findings and to investigate possible explanations.
Although researchers have reported that children living in single-parent14,15 or step-parent15,16 households are more likely than children in households with 2 biological parents to be prescribed methylphenidate, such findings do not reveal why living in a household without 2 biological parents is associated with an increased risk of methylphenidate use. The increased frequency of births outside of marriage in Western countries17,18 makes it possible for children to live in single-or step-parent households without ever having experienced parental divorce. Distinguishing between living in a particular family structure and experiencing parental divorce is important because of the potentially different causal pathways involved. Families headed by mothers who have never married are at increased risk of being poor, and higher rates of methylphenidate use may be a result of the link between poverty and child mental health problems.19,20 In contrast, losing a parent through divorce is viewed as a stressful event that may challenge child-coping processes. Experiencing parental divorce has been associated with subsequent increases in child behavioural problems,21–23 which might in turn be associated with an increased risk of receiving a prescription for methylphenidate. Furthermore, elevated rates of methylphenidate use among children in step-parent households may represent the persisting mental health consequences of an earlier parental divorce. As such, studies that can track the effects of family transitions as they unfold provide greater insight into the dynamic links between family processes and child outcomes.
To determine whether parental divorce subsequently increases the probability a child will be prescribed methylphenidate, researchers would need to design a prospective study to track children in households with 2 biological parents over time and the rates of methylphenidate use compared among those whose parents divorce during the study period with rates among those whose parents remain married. My search of MEDLINE, CINAHL, PsycINFO and Sociological Abstracts revealed no studies that specifically investigated the effect of parental divorce on methylphenidate use among children. I therefore sought to demonstrate that parental divorce is a risk factor for methylphenidate use by analyzing data from a prospective longitudinal study involving Canadian children.
|
Methods |
|---|
|
TopAbstractMethodsResultsInterpretationREFERENCES |
|---|
I restricted the sample for my study to children who participated in the first 4 cycles of the National Longitudinal Survey of Children and Youth and who, at the time of the initial interview in 1994, were living in a household with 2 biological parents (married or common-law relationship) and were between the ages of 2 and 7 (n = 5099). This age restriction ensured that the children would pass through the ages when methylphenidate is most likely to be prescribed.25,26 I excluded children whose families reported a period of separation before 1994 (n = 25) and children who were already taking methylphenidate in 1994 (n = 83). Because parental death is a fundamentally different event than divorce, I also excluded 34 children who had a parent die. Children for whom data on methylphenidate use were missing were also excluded, for a final sample of 4784 children within 3883 households.
The dependent variable (methylphenidate use) was considered to have been present if, at any time between 1994 and 2000, the parent reported that their child used the drug, and absent if, at each survey, the parent reported that their child was not taking the drug. Parental divorce was considered to have occurred if, between 1994 and 2000, it was reported to have taken place, and not to have occurred if the parents remained married during this period. The ages of the child and mother in 1994 were analyzed as continuous variables, measured in years.
A generalized estimating equation model was used to evaluate whether children whose parents divorced between 1994 and 2000 were more likely than children whose parents remained married to be prescribed methylphenidate. As in logistic regression, the outcome is binary; however, the generalized estimating equation model allows for observations within the same household to be correlated and produces the appropriate standard errors. The models controlled for age and sex of the child as well as the age of the mother in 1994.
The data were weighted up to the population according to Statistics Canada procedures to adjust for unequal probabilities of selection due to the multistage sampling strategy as well as attrition from the sample over time.
|
Results |
|---|
|
TopAbstractMethodsResultsInterpretationREFERENCES |
|---|
|
The generalized estimating equation model showed that boys were significantly more likely than girls to be prescribed methylphenidate during the study period (adjusted odds ratio 2.34, 95% confidence interval 1.33–4.09) (Table 2). It also showed that children whose parents divorced between 1994 and 2000 were significantly more likely than those whose parents remained married to be prescribed the drug (adjusted odds ratio 1.82, 95% confidence interval 1.01–3.33). Interactions between all variables in the model were tested. The interaction terms were not statistically significant and were consequently removed from the model.
|
In a subgroup analysis of data for the 633 children whose parents divorced between 1994 and 2000, the frequency of methylphenidate use did not differ significantly between the 210 children (33.2%) who experienced a subsequent transition into a step-parent household and those whose parents did not remarry.
|
Interpretation |
|---|
|
TopAbstractMethodsResultsInterpretationREFERENCES |
|---|
It is possible that the stress of divorce subsequently elevated child behavioural problems to a point that necessitated methylphenidate use. However, one cannot rule out alternative explanations. First, there is evidence that attention-deficit hyperactivity disorder has a genetic component,27,28 which suggests either direct transmission from parents to their children or an indirect tendency of parents with a history of mental health problems to divorce. Until more is known about the pathophysiology of attention-deficit hyperactivity disorder, researchers will continue to face challenges in disentangling the complex pathways through which shared genetic and environmental influences lead to this disorder and methylphenidate use.29
A second possible explanation is that, although clinicians are advised to distinguish between the symptoms of attention-deficit hyperactivity disorder and the effects of stressors such as parental divorce on child behaviour,30 a prescription for methylphenidate may result simply from increased contact with the health care system in the period following divorce or the heightened sensitivity of parents and clinicians to problematic child behaviour in the presence of stressors. This explanation suggests that methylphenidate is prescribed inappropriately to children whose parents have recently divorced. Unfortunately, there are no questions in the National Longitudinal Survey of Children and Youth that allow for testing of these causal pathways. Therefore, the findings of my study cannot reveal whether methylphenidate is inappropriately prescribed to children of divorce.
One limitation of this study is that the parents were not asked whether their child had been prescribed treatments other than methylphenidate for a mental disorder in general, or for attention-deficit hyperactivity disorder in particular. In addition, questions were restricted to a child's use of methylphenidate at the time of each survey cycle, which makes it impossible to determine whether use of the drug was started and stopped in the period between surveys. These limitations are unlikely to change the results of this study, but they may provide additional insight into how children come to be prescribed methylphenidate. Finally, to the extent that parents with a high-conflict divorce may be more likely to drop out of the survey than parents with less contentious divorces, my estimates may represent a conservative estimate of the effects of parental divorce on rates of methylphenidate use.
Methylphenidate use did not differ significantly between the children of divorced parents who remarried and those whose divorced parents did not remarry. However, given that half of all children whose parents divorce will later become part of a step-parent household,31 conducting such comparisons in this study may have been premature and should be investigated over a longer period.
Future research is needed to replicate the results of this study and to find explanations for the increased likelihood of a child being prescribed methylphenidate following parental divorce. Such findings will help shed light on the ways in which family dynamics are linked to child health and methylphenidate use.
|
Footnotes |
|---|
Acknowledgements: I thank Rachel Campbell for data analysis and Claire Pitcher for library assistance.
This research was supported by grant 512-2003-1016 from the Social Sciences and Humanities Research Council of Canada and a Support for the Advancement of Scholarship grant from the University of Alberta.
Competing interests: None declared.
|
REFERENCES |
|---|
|
TopAbstractMethodsResultsInterpretationREFERENCES |
|---|
Related Articles
This article has been cited by other articles:
|
|
 |
|
  B. K. Schultz EdS NCSP Methylphenidate use and divorce Can. Med. Assoc. J., September 25, 2007; 177(7): 751 - 751. [Full Text] [PDF]
|
|
|
|
|
|
L. A. Strohschein PhD Methylphenidate use and divorce Can. Med. Assoc. J., September 25, 2007; 177(7): 751 - 751. [Full Text] [PDF]
|
|
Read all eLetters
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | HELP | SEARCH |
|
|||||
|
Copyright 1995-2007, Canadian Medical Association. All rights reserved. ISSN 1488-2329
(e) 0820-3946 (p)
All editorial matter in CMAJ represents the opinions of the authors and not necessarily those of the Canadian Medical Association. |
|||||
