Background: Frequently, a questionnaire like the EQ-5D is applied to investigate elderly health-related quality of life (HRQoL), but current literature suggests that inputs that go beyond these traditional health aspects might be of importance. The capability approach is a different method, which integrates several non-health-related factors to define the well-being of the oldest old. Objective: We propose to investigate the differences in oldest old functionings and quality of life (QoL), given different levels of dependency, using both a utility-based (EQ-5D+C) and capability-based (Currently Achieved Functioning) questionnaire. Methods: We interviewed 99 Dutch elderly, living in the Groningen, Veendam, and Hoogeveen areas. The average age of the elderly was 80 years, who were living independently, still looking after themselves; living semi-dependently with moderate care; or living in a nursing home requiring consistent care. Results: The utility score for the dependent group is the lowest of all three groups, across the diseases investigated in this study. The respective average utility scores calculated for the dependent, semi-dependent, and independent subgroups were 0.56 (SD +/- 0.10); 0.84 (SD +/- 0.11), and 0.69 (SD +/- 0.13). Mobility and pain were reported to be the major domains where problems appeared across the three groups. Additionally, dependent elderly experience deficits in the role and control functionings while the other two subgroups experience deficits in pleasure and security. Conclusion: The results suggest that it is important to take note of the achievability of functionings and HRQoL, in addition to care dependency, to obtain QoL and well-being outcomes of the oldest old.