Sociodemographic patterns of nomophobia among university students: systematic review and meta-analysis with...

To estimate the pooled mean severity of nomophobia and describe associated sociodemographic patterns among university students through a systematic review with meta-analysis and meta-regression.

A systematic review of observational studies was conducted and registered in PROSPERO (CRD420261320594). Searches were performed in PubMed/MEDLINE, Scopus, Web of Science, and ERIC up to February 28, 2026. Studies including adult university students (≥ 18 years) and assessing nomophobia using the Nomophobia Questionnaire (NMP-Q) were eligible. Pooled mean nomophobia scores were estimated using random-effects models. Heterogeneity was assessed using the I² statistic and explored through subgroup analyses and meta-regression according to age, sex, and country income level.

Nineteen observational studies comprising 12,079 university students from 11 countries were included. The pooled mean nomophobia score was 75.74 (95% CI: 71.20-80.28), with extremely high heterogeneity (I² = 99.5%). Meta-regression using mean age as a continuous moderator (k = 11 studies with available age data) showed no significant linear association with nomophobia scores (β = -5.26; 95% CI: -13.81 to 3.29; p = 0.228), accounting for only 4.59% of between-study heterogeneity. Country income level was the only moderator significantly associated with nomophobia scores, with lower scores in high-income countries (β = -10.72; p = 0.043); however, it explained only 19.67% of the between-study variance and residual heterogeneity remained extreme (I2 = 99.47%). No statistically significant differences were found by sex.

Nomophobia appears to be common among university students and generally of moderate severity. However, scores varied widely across studies and this variability remains largely unexplained. Neither age nor sex was associated with severity, and the association with national income level, although statistically significant, accounted for a small share of the heterogeneity and was derived from study-level data. These findings should therefore be regarded as exploratory and hypothesis-generating, and do not support causal or individual-level interpretations.