Abstract
OBJECTIVES: The rise of smartphones and social media is widely seen as a pivotal societal shift that has fueled what the U.S. Surgeon General in 2023 described as "our epidemic of loneliness and isolation." The evidence for an increase in social isolation, however, has not accounted for age and cohort effects - i.e., variation in time spent alone over the life course and across generations. This study leverages the latest breakthrough in Age-Period-Cohort (APC) modeling to disentangle age-specific effects, societal changes, and generational shifts that contribute to social isolation in the era of smartphones. METHODS: We analyze data from the 2003-2022 waves of the American Time Use Survey (ATUS), focusing on time spent alone in nonwork activities among individuals aged 15-79 (N = 240,576). Using a recent breakthrough in APC modeling, we identify net effects of age, period, and cohort separately for men and women. RESULTS: Social isolation has increased over the past two decades for both women and men, with a notable acceleration in the mid-2010s, suggesting that societal shifts during this period may have intensified isolation. However, age and cohort effects play a much larger role in explaining the social isolation crisis of the 21st century. Gender differences are substantial, particularly in cohort trajectories and in age patterns after age 70. DISCUSSION: While smartphone-era societal shifts have contributed to a general rise in isolation, aging and generational differences explain more of the variation. Public health efforts should prioritize mitigating isolation among older adults or earlier cohorts, with attention to gender-specific patterns.