Matthew A. Christensen · PLoS ONE 2016 · cross-sectional study · n=653

Direct Measurements of Smartphone Screen-Time: Relationships with Demographics and Sleep

Cited 331 times in the scientific literature.

Level 4 - case-series / case-control

Cross-sectional observational study analyzing app tracking and survey data.

OpenAlex W2553875474 · doi:10.1371/journal.pone.0165331 · record verified 2026-08-31

What was done

Researchers performed a cross-sectional analysis of 653 adult participants (≥ 18 years) from the internet-based Health eHeart Study. Smartphone screen-time (active screen minutes per hour) was recorded continuously through a dedicated smartphone application over 30-day windows. Total and average screen-times across the 30 days, as well as during self-reported bedtime and sleeping periods, were computed. Demographic data, medical history, and sleep characteristics were collected via survey using the Pittsburgh Sleep Quality Index (PSQI). Linear regression was utilized to assess associations.

What was found

Median total screen-time over 30 days was 38.4 hours (IQR 21.4 to 61.3), and median average screen-time was 3.7 minutes per hour (IQR 2.2 to 5.5). Younger age and self-reported Black and "Other" race/ethnicity were associated with longer average screen-time after adjusting for confounders. Longer average screen-time was associated with shorter sleep duration and worse sleep efficiency. Longer screen-time during bedtime and sleep periods was associated with poor sleep quality, decreased sleep efficiency, and longer sleep onset latency. Numerical effect estimates and confidence intervals were not reported in the abstract.

Why it matters

This study provides direct, objective quantification of smartphone screen-time rather than relying on self-report, demonstrating clear links between higher screen usage (especially near bedtime) and worse sleep parameters.

Limits

The cross-sectional design precludes establishing causality or ruling out reverse causality (e.g., poor sleep driving higher phone use). Sleep metrics and bedtime windows were self-reported rather than objectively tracked with actigraphy or polysomnography. The sample was drawn from an internet-based volunteer cohort, potentially introducing selection bias.

Cited by