Thursday April 21 2011
The worst problems were in babies from 'multi-problem' families
“Babies who persistently cry and those who have difficulties with sleeping or feeding are more likely to develop behavioural problems in later life,” reported The Independent.
The story is based on an analysis of 22 studies looking at the possible association between difficulties such as excessive crying, problems feeding and sleeping problems in infants in the first year of life (known as regulatory problems) and the later development of childhood behavioural disorders such as ADHD and aggression. Researchers found that babies who experienced these problems were more likely to have behavioural difficulties later than those who did not. The most likely behavioural disorders for older children were “externalising” problems such as aggressive behaviour or temper tantrums.
This study collectively involved 16,848 children, of whom 1,935 had regulatory problems. The analysis is appropriate but limited by the nature of the studies included. The difficulty of defining “regulatory problems” in infants, and the problem of relying on information from parents were among the studies’ limitations. Importantly, the worst outcomes were in babies from “multi-problem families” who had poor parent-child interaction, social difficulties, depression and stress in the mother and a “negative” family environment. This indicated that these problems in both infancy and later childhood may be markers for psycho-social problems rather than directly associated with each other.
Where did the story come from?
The study was carried out by researchers from the University of Basel, Switzerland, the University of Warwick and the University of Bochum, Germany. The study was published in the peer-reviewed medical journal Archives of Disease in Childhood. Funding came from several sources including the Swiss National Science Foundation and F Hoffmann-La-Roche, a pharmaceutical company.
Generally, the study was reported accurately in the media. Although the Daily Mail reported that the findings suggest that crying babies were 40% more likely to grow up to display unruly behaviour, the 41% statistic reported by this study cannot be interpreted in this way as it represents the average change in scores across all studies using several different measures all standardised so that the study results can be pooled together. The Daily Mail did also report comments from one author who pointed out that the problems in babies that led to later issues were abnormally severe.
What kind of research was this?
This was a meta-analysis of 22 previous studies that investigated infant regulatory problems (excessive crying, sleeping difficulties and/or feeding problems in the first year of life) and their later behaviour in childhood. The results of these studies were combined and statistical tests used to look for possible associations between the two.
The researchers point out that these problems are common, with about 20% of infants affected. Although many of these difficulties are transient, persistent difficulties may predict behavioural problems later in life. The aim of this study was to test the nature and strength of any associations.
What did the research involve?
The researchers carried out a meta-analysis of 22 prospective cohort studies from 1987 to 2006 that statistically tested the association between infant regulatory problems and later childhood behavioural problems. They carried out a computer-based search of the literature on this topic, which produced an initial pool of 72 studies. To be included, studies had to meet certain inclusion criteria. Only prospective studies including at least one follow-up assessment were eligible. They had to focus on crying, sleeping and/or feeding problems in the first year of life, either occurring in isolation or in combination. They also had to include a measure of four behavioural difficulties: internalising problems (such as depression and anxiety), externalising problems (such as aggressive behaviour), ADHD symptoms (such as inattention) and general behaviour problems.
The researchers say identifying regulatory problems was a “major challenge” since consistent diagnostic criteria were lacking. For this study, excessive crying was defined as intense, unsoothable crying bouts for no apparent reason in the first three months of life. “Persistent regulatory problems” were defined as excessive crying beyond the third month of life, and sleeping and feeding problems that occurred at initial assessment and at follow-up.
The studies used a combination of parent interviews (60%), questionnaires (41%), infant diaries (32%) and observations to assess regulatory problems. Most informants were parents of the children included.
The researchers used statistical methods to assess the relationship between regulatory problems in infancy and later behavioural problems. To do this, they used a “standardised weighted mean effect size”, a statistical measure that is useful when different studies use diverse instruments with different scales to assess behavioural problems.
What were the basic results?
The researchers identified 22 eligible studies with 16,848 children, of whom 1,935 had regulatory problems.
Of the 22 studies, 10 examined the effects of excessive crying, four sleeping problems, three feeding problems and five multiple regulatory problems.
- The researchers found that children with previous regulatory problems had more behavioural problems than controls. (The standardised weighted mean effect size for this association was 0.41 [95% CI 0.28 to 0.54], which is a small-to-medium effect.)
- The strongest association was between regulatory problems and ADHD and “externalising” problems (for example, aggressive behaviour).
- Persistent crying problems had the strongest association with behavioural problems later.
- The more problems in infancy a child had, the higher the risk of behavioural problems later. Where a child had been referred to a clinician, the risk was also higher.
- Children with regulatory problems who also had family “risk factors” showed more behavioural problems than those with a small number of risk factors.
How did the researchers interpret the results?
The researchers say their analysis suggests that children with previous regulatory problems have more behavioural problems later in childhood than controls, with children from “multi-problem” families having the worst outcomes. They say their findings highlight the need for better understanding of the development of child mental disorders and for early intervention, particularly in families with other problems.
This meta-analysis had a number of limitations which the authors acknowledge:
- The studies that were included were “highly heterogenous”, meaning that they differed in populations, design, methods and outcome. Although the authors took steps to address this issue, it makes the studies difficult to compare and makes an overall analysis less reliable.
- Most studies only focused on a single regulatory problem, without controlling for any others, although crying, feeding and sleeping problems often co-exist in infancy, these disorders are difficult to define consistently in the absence of consistent diagnostic criteria.
- Different scales were used in the studies, which means that the researchers had to standardise the measurements. This means that the effect size is harder to interpret - 40% does not mean a 40% chance of developing problems as the Daily Mail has reported. Rather, it is the average increase across all studies of the difference measured using multiple different scales. These were adjusted, or standardised, so that the results could be pooled. The researchers were relying on parental reports for most measurements, which may introduce some inaccuracies as parents may have different perceptions of what constitutes a regulatory problem.
It is difficult to make any robust conclusions from these findings, but interpreting them to mean that infants with these problems are automatically at greater risk of behavioural problems later is probably unwise.
Importantly, babies with regulatory problems who went on to develop behavioural disorders often came from “multi-problem families” with poor parent-child interaction, social difficulties, depression and stress in the mother and a “negative” family environment. The researchers acknowledge that it was difficult to adjust for these factors in the analysis, and it is possible that the “regulatory problems” and later behavioural difficulties are both markers for psycho-social problems.