Poor child health linked to television

Behind the Headlines

Tuesday May 4 2010

The relationship between health and TV watching is complex

“The more TV a toddler watches, the higher the likelihood they will do badly at school and have poor health at the age of 10,” BBC News reported. This finding comes from a study on 1,300 Canadian children, which found that increased viewing time at two years was associated with lower levels of classroom engagement, poor achievement in maths, reduced physical activity and an increase in body mass index.

The study does have some strengths. For example, it collected data on TV viewing and then followed-up the children over time to see how outcomes develop. However, there were limitations, such as basing TV viewing time on parents’ estimates rather than direct monitoring. Importantly, while some outcomes were affected by early TV viewing, others, such as reading achievement and emotional distress, were not. Maternal education and family characteristics also showed links to many outcomes assessed.

This study suggests that greater TV viewing in early childhood may be associated with some poorer outcomes in later childhood. It will undoubtedly prompt further study. Such research will need to determine whether simply reducing a toddler’s TV viewing can improve outcomes, or whether more complex interventions are required.


Where did the story come from?

Dr Linda S. Pagani and colleagues from the Universities of Montreal and Michigan carried out this research. The study was supported by Canada’s Social Science and Humanities Research Council International Collaboration Fund.
The study was published in the peer-reviewed journal Archives of Pediatrics and Adolescent Medicine.

BBC News gave an accurate presentation of this study.


What kind of research was this?

This was a prospective cohort study looking at the effect of early childhood TV exposure on the academic, psychosocial and lifestyle characteristics of the children in later childhood.

This study’s strengths include its prospective design and the fact that it follows children over time. Collecting data prospectively (going forward) means that researchers can design their study to collect the exact the data that they want, and to collect this data in a standard way. This is generally considered more reliable than having to rely on people’s recall of past events or relying on records originally collected for other purposes.

The fact that an assessment of children’s TV watching was collected early in the study and compared to subsequent outcomes means that we can be sure that their TV viewing habits preceded their outcomes. Therefore the researchers can check whether these earlier behaviours potentially affect these later outcomes.

If the study had measured TV watching, academic performance and other outcomes at the same point in time, it would not be possible to say that the TV watching could have directly affected these outcomes. Equally, as children were not randomly assigned to watch differing amounts of TV to test its effect, researchers still need to consider whether it is actually children watching different amounts of TV or other differences that have affected the study’s outcomes.


What did the research involve?

The researchers analysed data on 1,314 children enrolled in the Quebec Longitudinal Study of Child Development. Their parents reported on the children’s TV watching at two points during their pre-school years (29 and 53 months of age). The children’s academic performance, psycho-social outcomes and lifestyles were then assessed at age 10. The researchers then looked at whether there were links between early TV watching and these later childhood outcomes.

The study had originally randomly selected 2,837 infants born between 1997 and 1998 in Quebec. Of this sample, 2,120 children (75%) could be contacted, were eligible, and had been granted parental consent for participation at age five months.

These children were followed-up at ages 17, 29, 41 and 53 months. Parents filled in questionnaires at age 29 and 53 months about how much TV their child watched per day. The current study looked at the 1,314 children (46% of original sample) whose parents provided this information on their early TV viewing.

The children’s academic performance and psycho-social outcomes were reported by their teachers at age 10. This included ratings of maths and reading achievement relative to other members of the class, with scores ranging from -2 (near the bottom of the class) to +2 (near the top of the class). The teachers also filled in questionnaires about classroom behaviour (particularly emotional distress, reactive aggression and victimisation), as well as classroom engagement.

Parents reported on their child’s time spent on video game use, time spent on physical activities, physical activity level relative to other children, and how often they engaged in activities requiring physical effort. They also reported on how frequently their children consumed soft drinks, sweet snacks, and fruits and vegetables: ranging from ‘never’ (score of 1) to ‘four or more times a day’ (score of 7). The children’s body mass index (BMI) was also assessed.

The researchers took into account factors which could affect results that were measured at age 17 months, including gender, temperament problems, hours of continuous sleep, family functioning, social behaviour, cognitive skills, BMI and maternal education. Analyses also took into account TV watching habits at age 10.


What were the basic results?

The children watched an average of 8.8 hours of TV a week at 29 months, increasing to 14.9 hours of TV a week at age 53 months. These averages were reported to be within current US recommendations of not more than two hours a day after two years of age, although some children watched more than the recommended levels (11% at age 29 months, 23% at age 53 months). Children who watched more TV at 29 and 53 months had mothers with lower levels of education. Children with more TV exposure at 29 months were more likely to come from single-parent families.

The researchers found a number of statistically-significant associations with TV watching at age 29 months. Each additional hour of television was associated with:

  • 6% reduction in maths achievement
  • 7% reduction in classroom engagement
  • 10% increase in classroom victimisation
  • 13% decrease in time spent doing weekend physical activity
  • 9% decrease in activities involving physical effort
  • 10% increase in time spent playing video games
  • 9% reduction in general fitness score
  • 9% increase in consumption score for soft drinks
  • 10% increase in consumption score for snacks
  • 16% decrease in consumption score for fruits and vegetables
  • 5% increase in probability of being classed as overweight

Television viewing at 29 months showed no effect on reading ability, emotional distress or reactive aggression.

Each additional hour’s increase in TV viewing between 29 and 53 months was associated with similar effects, except it showed no association with classroom engagement, maths achievement, or consumption of fruits and vegetables or soft drinks.

Maternal education and familial factors were associated with levels of TV watching, and they were related to most of the outcomes assessed.


How did the researchers interpret the results?

The researchers concluded that they had observed “modest, yet non-trivial prospective associations” between early television exposure and outcomes age 10. They also said that the long-term risks associated with early TV exposure may represent a pathway to “unhealthy dispositions” in adolescence. They say that gaining further understanding of these risks in the general population is “essential for promoting child development”.



Overall, this study suggests that there may be links between early childhood TV viewing and later health behaviours, classroom victimisation and engagement, and maths achievement. There are limitations to this study, some of which the authors note:

  • The measure of TV-watching was based on parental report, and may not have been accurate. The authors also note that they did not assess the content and quality of the TV watched.
  • Under half of the original randomly-selected population sample were analysed in the current study, which excluded a large proportion of participants. This means that the results may not be representative of the sample as a whole.
  • Measurements of maths and reading ability were based on teachers’ reports of a child’s ability relative to the rest of their class. This way of estimating ability may not have been as accurate as using performance in standardised school tests.
  • The study carried out multiple statistical tests, which can increase the likelihood of finding significant associations by chance.
  • Although the study did take into account a number of factors that could affect results, there may be other factors that are having an effect. For example, there was no specific assessment of socio-economic status, although indicators of socio-economic status (such as maternal education) did show association with the outcomes assessed.
  • Several factors had larger effects on individual outcomes than early TV viewing. For example, a child’s gender, maternal education, and family make-up had a greater effect on classroom engagement than early TV viewing. Maternal education and family make-up also had a greater effect than early TV viewing on maths achievement.

No doubt these findings will give rise to further assessment of the effects of early TV viewing on children’s later development.

Analysis by Bazian

Edited by NHS Choices

Links to the headlines

'Long-term harm' of too much TV for toddlers. BBC News, May 4 2010

Links to the science

Pagani LS, Fitzpatrick C, Barnett TA and Dubow E. Prospective Associations Between Early Childhood Television Exposure and Academic, Psychosocial, and Physical Well-being by Middle Childhood. Arch Pediatr Adolesc Med. 2010;164(5):425-431


How helpful is this page?

Average rating

Based on 2 ratings

All ratings

1  ratings
0  ratings
0  ratings
1  ratings
0  ratings

Add your rating