Background <p>Depression among older adults is increasingly emerging as a global public health concern. With the rapid advancement of digital information technology, digital smart devices have profoundly reshaped the lifestyle of older adults and may influence their mental health. However, the association between digital smart device use and depression in this population remains unclear, with existing studies reporting inconsistent findings. This study employed a three-level meta-analytic approach to systematically explore the association between digital smart device use and depression among older adults.</p> Methods <p>Systematic searches were conducted on six electronic databases including PubMed, Cochrane Library, Embase, Web of Science, MEDLINE, and PsycINFO from their inception to May 6, 2025. Original observational studies (cross-sectional or cohort) were included. Exclusion criteria comprised non-English publications, review articles, conference papers or abstracts, duplicate publications, and studies without available full text or required data. Two researchers independently performed literature screening, data extraction, and quality assessment. Three-level random-effects meta-analyses were conducted to estimate the pooled standardized beta coefficients (<i>β</i>) for continuous depression outcomes and odds ratios (OR) for binary depression outcomes. Risk of bias was assessed using the Newcastle–Ottawa Scale (NOS) for cohort studies and the AHRQ criteria for cross-sectional studies. Potential moderators were explored using three-level meta-regression, and potential publication bias was evaluated using funnel plots, modified Egger’s test, and trim-and-fill analysis.</p> Results <p>A total of 28 articles (23 cross-sectional, 5 cohort) published between 2014 and 2025 were included in the systematic review, with 24 eligible for meta-analysis. All included studies were deemed high quality (mean NOS score: 7.60; mean AHRQ score: 7.57). For continuous outcomes, digital smart device use was significantly associated with lower depression scores (<i>β</i> = -0.201, 95% <i>CI</i>: -0.324, -0.079, <i>p</i> &lt; 0.01). For binary outcomes, device use was associated with a decreased risk of depression (OR = 0.676, 95% <i>CI</i>: 0.613, 0.745, <i>p</i> &lt; 0.001). Moderator analyses indicated that the association was not significantly modified by age, study design, sample size, region, measurement tools, or publication year (<i>p</i> &gt; 0.05).</p> Conclusions <p>This study has certain limitations, including publication bias (<i>p</i> &lt; 0.001). However, trim-and-fill analyses confirmed that the results remained statistically significant after imputing missing studies. This three-level meta-analysis provides robust evidence that digital device use is inversely associated with depression among older adults, supporting their potential role in mental health promotion.</p> Trial registration <p>The study protocol was registered with the International Prospective Register of Systematic Reviews (PROSPERO) in 2024 (registration number: CRD42024616432).</p>