BackgroundInternet addiction is a major global concern and health burden, compounded by the advances of Internet Plus and artificial intelligence. There is controversy on the question of recognising some types of internet addiction as mental illness, largely because there is no consensus on its link to mental health outcomes. We investigated the associations between internet addiction severity and several mental health outcomes including four psychopathologies, serious mental illness, and suicidalities. MethodsIn this cross-sectional study, all first-year undergraduate students enrolled in Sichuan University (Chengdu, China) during 2015–18 were invited to participate in the study survey. Participants were required to finish the whole questionnaire (online) by Oct 31 of their enrolment year. We excluded participants who did not respond completely, did not complete the questionnaire by the deadline, were not aged 15–23 years, or who provided obviously dishonest information (eg, selecting the same choice for all questions or several users having the same unique ID). The severities of internet addiction were categorised using Young's 20-item internet addiction test (normal 0–30 points, mild 31–49, moderate 50–79, and severe 80–100). The four psychopathologies were defined according to the recommended cutoff points of the 15-item Patient Health Questionnaire (high somatic symptom severity), the 9-item Patient Health Questionnaire (clinically significant depression), and the Symptom Checklist-90 (psychoticism and paranoia). Serious mental illness was screened using the 6-item Kessler Psychological Distress Scale (cutoff 12–13 points), and lifetime suicidal ideation, planning, and attempts were measured with the Suicidal Behaviours Questionnaire-Revised. Prevalence among student demographics or internet addiction severity groups was compared with the χ2 test, adjusted odds ratios (ORs) of mental health outcomes in addiction severity groups were compared using binary logistic regression adjusted for demographic confounders (ie, age, sex, and enrolment year), and interactions between psychopathologies were assessed using a two-tailed alpha of 0·05. Findings37 187 undergraduate students enrolled in the university during the study years. 34 140 (91·8%) of them agreed to participate and accessed the online survey system, and 32 269 (86.8%) responded fully. 12 254 (38·0%) students had mild internet addiction, 2057 (6·4%) had moderate addiction, and 67 (0·2%) had severe addiction. The prevalence and risks of the four psychopathologies and their comorbidities, serious mental illness, and suicidalities increased significantly with internet addiction severity (p values from 0·0003 to <0·0001), and most increments in risk were large (ORs increasing more than four times, except for suicidalities) for students with moderate and severe internet addiction. Prevalence of clinically significant depression, most strongly associated with internet addiction after adjustment for demographics and psychopathologies, increased from 1·0% (95% CI 0·9–1·2) for students with no addiction to 4·9% (4·5–5·3) for those with mild addiction, 24·8% (22·9–26.7) for those with moderate addiction, and to 56·7% (44·5–68·9) for those with severe addiction. Risk for the three addiction groups compared with the no-addiction group increased accordingly (OR 4·2 [95% CI 3·6–5·0] for mild, 18·9 [15·7–22·9] for moderate, and 59·9 [33·5–107·2] for severe addiction). Prevalence of the four psychopathologies increased from 4.1% (3·8–4·4) in students with no addiction to 36.8% (34·7–38·9) in those moderately addicted and to 67.2% (55·6–78·7) for those severely addicted. Similarly, prevalence increased from 25.0% (24·4–25·6) to 67·5% (65·6–69·6) and 71·6% (60·6–82·7) for suicidal ideation, from 2·6% (2·4–2·8) to 16·5% (14·9–18·1) and 32·8% (21·3–44·4) for suicidal plans, and from 0·5% (0·4–0·6) to 3·6% (1·1–1·5) and 11·9% (4·0–20·0) for suicidal attempts. InterpretationOur study format does not allow us to identify possible causal relationships between internet addiction and mental health comorbidities, but allows us to compare the prevalence and risk of these outcomes through our large population size (to our knowledge, the largest to date in such a study) and methodology. Internet addiction was strongly associated with several adverse mental health outcomes, supporting the validity of defining moderate and severe, but not mild, internet addiction as mental illness. FundingNational Scientific and Technical Fund of China (grant 81571305), Department of Science and Technology of Sichuan Provincial Government (grant 2019YFS0153), and 1.3.5 Project for Disciplines of Excellence, West China Hospital.
Read full abstract