Registry-Based Dementia Incidence in Argentina: A Nationwide, Decade-Long Analysis by Age, Sex, and Etiology

BACKGROUND: National data on dementia incidence is scarce in low- and middle-income countries, limiting public health surveillance and planning. In Argentina, no nationwide estimates of dementia incidence exist despite rapid population aging. Administrative disability registries do not capture clinical disease onset; rather, they record certification events and therefore provide a registry-based proxy for population-level dementia surveillance in settings without prospective incidence cohorts. METHODS: We conducted a population-based registry study using data from Argentina's National Registry of Persons with Disabilities between 2016 and 2022, excluding pandemic-disrupted years (2020-2021). Incident registry entries with dementia recorded as the primary cause of disability were identified using ICD-10 codes for Alzheimer's disease and related disorders (ADRD), vascular dementia, and other etiologies. Crude and age-standardized incidence (ASI) of dementia certification per 100,000 inhabitants were estimated by sex, age group, dementia subtype, and early- (<65 years) versus late-onset (≥65 years) dementia. Pre-specified sensitivity analyses quantified capture fraction against regional cohorts and adjusted temporal trends for an exogenous proxy of registry maturation. RESULTS: In 2022, the registry-based crude dementia incidence was 47.7 per 100,000 inhabitants (95% CI: 46.6-48.7). Crude incidence was higher in women than men, whereas age-standardized incidence was comparable between sexes (48.1 vs. 45.5 per 100,000). Incidence increased steeply with age, from 0.9 per 100,000 at ages 40-45 to nearly 480 per 100,000 among individuals aged ≥95 years. ADRD predominated from approximately 55-60 years onward, while other etiologies were relatively more frequent at younger ages. Late-onset dementia incidence (≥65 years) was substantially higher than early-onset dementia (<65 years) (130.9 vs. 6.3 per 100,000). Provincial age-standardized rates varied from 4.7 to 137.7 per 100,000, and were strongly associated with care access metrics. Between 2016 and 2022, age-standardized incidence increased significantly (p<0.001); however, capture fractions of 6 to 12% relative to regional cohorts and a 45% attenuation of the temporal coefficient after adjustment for registry maturation. CONCLUSIONS: This study provides the first nationwide characterization of registry-based dementia certification incidence in Argentina. Distinct age-, sex-, and etiology-specific patterns highlight different epidemiological scenarios for early- and late-onset dementia. Because these estimates reflect administrative certification rather than clinical disease onset and are subject to differential ascertainment, they should be interpreted as a lower bound of true incidence; nonetheless, they constitute a feasible and informative surveillance approach for countries lacking national clinical registries and can support health system planning and prevention strategies.