Background: Benzodiazepines are widely prescribed to older adults in primary care despite concerns regarding their potential cognitive adverse outcomes. Evidence on the association between benzodiazepine use and cognitive impairment remains inconsistent and is limited in middle-income countries, particularly in routine primary care settings. Objectives: To evaluate the association between benzodiazepine use and cognitive impairment among older adults in a primary care hospital in Mexico and to explore clinical factors associated with benzodiazepine prescribing patterns. Methods: A cross-sectional analytical study was conducted using non-probabilistic convenience consecutive sampling in a sample of 228 in adults aged ≥60 years attending a primary care hospital in Irapuato, Mexico, during 2025. Cognitive status was assessed using the Mini-Mental State Examination (MMSE) and classified as normal cognition, mild/moderate cognitive impairment, or major cognitive impairment. Benzodiazepine exposure was identified through pharmacy records and categorized by duration of exposure (6 years). Sociodemographic variables were collected. Associations were evaluated using chi-square tests, and crude logistic regression models were used to estimate odds ratios (ORs) with 95% confidence intervals (CIs). An additional exploratory analysis assessed clinical factors associated with benzodiazepine type using a dichotomized outcome to ensure model stability. Results: A total of 228 older adults with documented benzodiazepine use were included. The sample represents the complete population of eligible benzodiazepine users during the study period. Cognitive impairment was identified in 61.8% of participants. Longer duration of benzodiazepine use was significantly associated with worse cognitive status (χ² = 25.81; p < 0.001), with the highest proportion of major impairment observed in individuals with more than six years of use. In exploratory analyses, benzodiazepine prescribing patterns were significantly associated with sleep disorders (χ² = 11.56; p = 0.009), polypharmacy (χ² = 10.28; p = 0.016), and MMSE category (χ² = 13.87; p = 0.031). In adjusted models, polypharmacy was associated with increased odds of receiving the most frequently prescribed benzodiazepine (OR 2.37; 95% CI 1.27–4.41; p = 0.007), while sleep disorders (OR 2.10; 95% CI 1.10–3.90; p = 0.009) and cognitive status (OR 1.85; 95% CI 1.05–3.20; p = 0.031) also showed significant associations. No comparisons with non-users were performed due to lack of comparable cognitive assessment; however, this comparison should be interpreted cautiously due to differences in assessment methods between groups. Conclusions: Among benzodiazepine users in primary care, longer duration of exposure was associated with worse cognitive status. Additionally, prescribing patterns were influenced by clinical factors such as polypharmacy, sleep disorders, and cognitive status, suggesting a predominantly clinically driven approach. These findings highlight benzodiazepine use as an important marker of cognitive vulnerability and underscore the need for careful prescribing, regular medication review, and deprescribing strategies in older adults.

End of life is a profound and culturally charged human dimension. Bioethics and palliative medicine have promoted a shift from the traditional biomedical model toward person-centered care that preserves dignity through the dying process. Physicians-in-training live in continuous contact with illness and death, yet are rarely invited to reflect on their own finitude. Reflection on one’s own mortality has been linked to greater comfort discussing prognosis, more patient-centered end-of-life decision making, and reduced avoidance behaviors in clinical encounters with dying patients; however, most training programs address death only as a technical or bioethical topic, rarely as a personal experience requiring processing. This study explored the end-of-life expectations and meaning-making of first-year medical residents at a Mexican general hospital, considering the influence of medical culture and sociocultural discourse. A qualitative phenomenological design was used. Data were collected through asynchronous, self-recorded video interviews guided by five open-ended questions addressing illness and death, family, autonomy, medical care at the end of life, advance directives, and euthanasia. Purposive sampling included twenty-eight first-year residents (2024–2025 cohort) who had previously completed coursework in medical ethics, giving the final sample of twenty-eight participants. Data were analyzed using an interpretive phenomenological approach supported by ATLAS.ti. Open coding of experiential meaning units was followed by iterative hermeneutic interpretation. Participants described the end of life through two tensions: an idealized, peaceful death surrounded by family, and a pragmatic view shaped by clinical exposure to suffering and futile treatment. Participants uniformly favored palliative care and expressed a preference for dying at home in the case of a terminal illness. Attitudes toward euthanasia ranged from support as an individual right to religious/ethical opposition, frequently anchored in personal grief experiences. Advance directives were positively valued but rarely formalized. Emotional overflow, including tears, occurred in some interviews. Despite clinical experience, medical residents are not immune to the emotional and existential tensions surrounding death. Findings suggest that integrating reflective, narrative-based pedagogical strategies into residency training may help to process death-related distress and strengthen person-centered, dignity-preserving end-of-life care.