Introduction Variable results are reported after functional treatment for Garden-1 femoral neck fracture, with no definite factors for failure, particularly in the absence of selection for risk. It seems well indicated in frail and/or dependent patients with comorbidities, but this has not been specifically assessed, and failure is frequent in this target population. We therefore performed a retrospective study to: (1) assess results of functional treatment for Garden-1 impacted femoral neck fracture in dependent patients, and (2) analyze survival in terms of complications and death, and related factors. Hypothesis Functional treatment for Garden-1 fracture in dependent patients gives poor results, with a high rate of surgical revision. Material and methods A retrospective analysis was made of Garden-1 femoral neck fracture in dependent patients (Parker score ≤ 5), with a minimum 2 years’ follow-up. One hundred and forty-six patients were included: mean age, 81.3 ± 8.7 years (range, 55.7–99.6 years). The endpoint was survival in terms of complications requiring surgery, secondary displacement requiring surgery, excessive pain requiring surgery, non-union or femoral head osteonecrosis. Secondary endpoints were overall mortality and mortality related to complications. Results Mean follow-up was 4.2 ± 2.6 years (range, 2.0–10.3 years). Ninety-one of the 146 patients (62.3%) required secondary surgery: 79 (54.1%) early (< 3 months post-fracture), with 77 (52.7%) secondary displacements and 2 cases (1.4%) of excessive pain; and 12 (8.2%) late (162.2 ± 132.3 days; range, 90–454 days), with 8 (5.4%) non-unions and 4 (2.7%) osteonecroses. Mean time to onset of secondary displacement was 13.6 ± 11.8 days (range, 0–67.0 days). Two-year survival in terms of revision surgery was 34.1% [95% CI: 26.0–42.4]. At last follow-up, 91 patients (62.3%) had died; 2-year survival in terms of death was 69.9% [95% CI: 62.4–77.3]. Survival analysis in terms of complications revealed greater mortality in absence of complications: 42 of the 55 patients (76.3%) without complications requiring surgery died, versus 49 of the 91 (53.8%) with complications requiring surgery (p = 0.012); relative risk of death in absence of complications requiring surgery was 1.42 [95% CI: 1.33–5.77]. Discussion Functional treatment for Garden-1 fracture in dependent patients gave poor short- and medium-term results. Surgery is therefore recommended in this specific population; the present findings should improve survival. Level of evidence IV, retrospective study.