There is a mild to moderate negative association between screen time and language development in healthy children younger than 12 years of age. The data is mainly based on observational studies, including children without neurologic or somatic conditions. Although it is difficult to demonstrate causality in observational studies, these results indicate that long screen time may slightly or moderately increase the risk of delayed language development in children younger than 12 years of age. However, it remains unclear, whether long screen time increases the risk of developmental language delay.
| Reference | Study type | Population | Intervention and comparison | Outcomes | Risk of bias «Additional comments for included studies...»2 |
|---|---|---|---|---|---|
| «Madigan S, McArthur BA, Anhorn C, ym. Associations...»1 | Meta-analysis based on observational studies | Studies were included, if (1) mean sample age 12.0 years and younger; (2) a measure
of screen use; (3) a measure of language skill (ie, expressive, receptive, or combined);
(4) observational study; and (5) statistical data that could be transformed into an
effect size. Studies with children with autism spectrum disorder or intellectual disabilities were excluded, or the study was experimental and without baseline measures prior to experimentation |
Quantity of screen time. Quality of screen time. Onset of screen use (the age children first started screen use). |
Measure of language skill. | Moderate |
| «Mustonen R, Torppa R, Stolt S. Screen Time of Pres...»2 | cross-sectional study | 164 Finnish children, aged 2.5-4.1 years, with normal development and Finnish as the native language | Screen time alone. Screen time with parent. |
General Language Ability measured with FinCDI III and RDSL III. | Moderate |
| «Asikainen M, Kylliäinen A, Mäkelä TE, ym. Exposure...»3 | cohort | 1667 pregnant women, whose children were followed-up at 18 and 24 months of age. Children with severe neurological or somatic conditions were excluded. | Child watching TV or using electronic device <77 min/day vs. >77 min/day | Expressive vocabulary at 18 and 24 months of age | High |
| «Al Hosani SS, Darwish EA, Ayanikalath S ym. Screen...»4 | case-control | Cases: Children, aged12-48 months with language delay who came for the first time
to pediatric clinic (n=227). Controls: Age and gender matched typically developing children control subjects who were recruited from the Well Baby Clinic in Ambulatory Health Services (n=227). |
Child using an electronic device at 12-24 months of age. | NA | High |
| «Sundqvist A, Koch FS, Birberg Thornberg U, ym. Gro...»5 | cohort | All children (N = 1,324) born within a specified time period within the Linköping
municipal area were invited by letter. Ninety-two parents of 25months olds (50 boys/42 girls) participated in the study. |
Recording their home sound environment during a typical day [Language ENvironment Analysis (LENA)] and participating in an online version of the MacArthur Communicative Development Inventory, which includes a vocabulary scale as well as a grammar and pragmatics scale | High | |
| «Acebedo L, Buil-Legaz L, Adrover-Roig D, ym. Impac...»6 | The total sample included 60 preteens from different schools in Mallorca (Spain).
All the participants were in the 5th or 6th grade of Primary Education (age in years: M =
11.63, SD = 0.486). None of the participants in the sample had a diagnosis of learning difficulties nor presented educative education needs |
A standardized validated questionnaire was used to obtain data regarding participant's language level, and an ad hoc questionnaire served to quantify the frequency and type of media use within the family setting and the interaction frequency with their parents. | The study followed a transversal design, and participants were evaluated with two tests that gathered information on the frequency and types of media use, language interaction with their parents, and participants' expressive and receptive linguistic level. |
| Reference | Comments |
|---|---|
| «Madigan S, McArthur BA, Anhorn C, ym. Associations...»1 | Participants totaled 18905 from 42 studies included. Quantity of screen use included duration of screen time, defined as duration of time spent watching television, movies, or DVDs on devices (eg, tablets or televisions), as well as background television, typically reported in hours per day/week. Quality of screen use included co-viewing or how often caregivers joined their children while exposed to screens, as well as extent of exposure to educational programs (eg, Sesame Street). Onset of screen use was defined as the age children first began viewing screens. In all included studies, the method for measuring screen use (ie, duration, co-viewing, and/or educational) was parent report. Most studies in the meta-analysis are prior to the mass movement toward handheld devices. Approximately half of the studies in the meta-analysis did not present adjusted effect sizes. |
| «Mustonen R, Torppa R, Stolt S. Screen Time of Pres...»2 | The mothers completed The Screen Time Questionnaire. Screen time was defined as time
spent watching TV and watching or using mobile devices, computers, laptops, and game
consoles. The receptive part of the Reynell Developmental Language Scales III (RDLS III «Mustonen R, Torppa R, Stolt S. Screen Time of Pres...»2) and the total score of the Finnish version of the MacArthur Communicative Development Inventories III (FinCDI III Total «Mustonen R, Torppa R, Stolt S. Screen Time of Pres...»2) were used to measure the children's receptive and expressive general language ability. 95% CIs were not reported. |
| «Asikainen M, Kylliäinen A, Mäkelä TE, ym. Exposure...»3 | From the original cohort of 1667 pregnancies, data were available from 977 children
at 18 months and 822 children at 24 months. The participating families were asked to complete a set of questionnaires, prenatally, at birth and when the children were three, eight, 18 and 24 months of age. Data on children aged 18 and 24 months were collected using a web-based questionnaire that could be answered by either of the parents. Daily screen time, including television viewing and watching other electronic devices, was measured in a number of ways including the number of hours the child watched television, DVDs, videos or electronic devices. |
| «Al Hosani SS, Darwish EA, Ayanikalath S ym. Screen...»4 | Children were diagnosed with language delays based on the Receptive-Expressive Emergent
Language Test and early signs of language and speech disorders. A delay of 25% or
greater by age 16–24 months was considered important. Case and control groups were different in some characteristics, that could interfere with the results. |
| «Sundqvist A, Koch FS, Birberg Thornberg U, ym. Gro...»5 | Ninety-two parents of 25months olds (50 boys/42 girls) recorded their home sound environment during a typical day [Language ENvironment Analysis (LENA)] and participated in an online questionnaire consisting of questions pertaining to daily DM use and media mediation strategies, as well as a Swedish online version of the MacArthur Communicative Development Inventory, which includes a vocabulary scale as well as a grammar and pragmatics scale |
| «Acebedo L, Buil-Legaz L, Adrover-Roig D, ym. Impac...»6 | The language level of 60 bilingual Spanish-Catalan preteens aged 11-12 years was assessed with four subtests of the Spanish version of the standardized clinical evaluation of language fundaments (CELF-5-Spanish) as well as media use at home through a multiple-choice questionnaire. |
Results
| Reference | Number of studies and number of patients (I/C) | Follow-up time | Association (r) | Heterogeneity | Other comments |
|---|---|---|---|---|---|
| CI=confidence interval | |||||
| «Madigan S, McArthur BA, Anhorn C, ym. Associations...»1 | 38 studies with 18313 participants. | NA | -0.14 (95% CI -0.18 to -0.10) | I2 = 83.67. | The funnel plot inspection revealed asymmetry and the Egger test suggested that studies with smaller sample sizes had more extreme effect sizes. |
| «Mustonen R, Torppa R, Stolt S. Screen Time of Pres...»2 | 164 participants. | NA | RDLS III: -0.10 (ns) FINCDI III: -0.18 (p<0.05) |
||
| Level of evidence: high | |||||
| Reference | Number of studies and number of patients (I/C) | Follow-up time | risk at 18 months (OR) | Risk at 24 months (OR) | Other comments |
|---|---|---|---|---|---|
| OR=odds ratio; CI=confidence interval | |||||
| «Asikainen M, Kylliäinen A, Mäkelä TE, ym. Exposure...»3 | 977 (18 months) and 822 (24 months) participants. | NA | 0.16 (95% CI 0.07-0.24) | 1.33 (0.67-2.64) | |
| Level of evidence: low The quality of evidence is downgraded due to study limitations and imprecision. |
|||||
| Reference | Number of studies and number of patients (I/C) | Follow-up time | Association (r) | Heterogeneity | Other comments |
|---|---|---|---|---|---|
| CI=confidence interval | |||||
| «Madigan S, McArthur BA, Anhorn C, ym. Associations...»1 | 5 studies with 2792 participants. | NA | -0.19 (95% CI -0.33 to 0.05) | I2 = 76.47. | Asymmetry was revealed in the funnel plot, but the Egger test did not suggest publication bias |
| Level of evidence: moderate The quality of evidence is downgraded due to study limitations. |
|||||
| Reference | Number of studies and number of patients (I/C) | Follow-up time | Association (r) | Heterogeneity | Other comments |
|---|---|---|---|---|---|
| «Madigan S, McArthur BA, Anhorn C, ym. Associations...»1 | 4 studies with 457 participants. | NA | 0.17 (95% CI 0.07-0.27) | I2 = 10.59. | |
| Level of evidence: moderate The quality of evidence is downgraded due to study limitations. |
|||||
Results of study 4:
The following factors are significantly correlated (statistically and clinically) with language delay among children: those who use a device [OR 6.82 (4.09–11.40), P-value < 0.001], early onset of using electronic devices (12–24 months) (OR 8.22 (1.71–39.55), P-value = 0.009), and fewer TV viewing hours per day, at 3 to 4 h and 5 to 8 h (OR 2.67 (1.65–4.32), P-value < 0.001) and (OR 4.93 (1.90–12.79), P = 0.001), respectively.
Results of study 5:
There was a significant negative correlation between how much the child watched TV content (defined as any TV content viewed on any device from any source) and the vocabulary scale. There was no correlation between the child's use of smartphones and tablets, and the child's language development (p's > 0.05). Computers were used by 14 of the children, but there was a significant positive correlation between the child's use of computers and vocabulary.
Background TV was not common in our sample and did not correlate with language measures (p's > 0.05).
Results of study 6:
Results showed lower language scores in preteens who had access to more media devices, who used them more frequently, and who talked less with their parents. Language scores were also significantly lower in preteens who used media devices to communicate compared to those who used it as a school aid or to learn new things. These results are not influenced by socioeconomic level, sex, chronological age, or family language.