Determination of whether the association between serum albumin and activities of daily living in frail elderly people is causal.

Objectives: Serum albumin and activities of daily living (ADL) are associated with each other, but whether the association is causal is not known. The purpose of this study was to determine whether a causal association exists between serum albumin and ADL levels.

Methods: The subjects were 116 frail elderly individuals (34 men and 82 women; mean age 83.0 years). Demographic characteristics, serum albumin, ADL, and handgrip strength were measured at a baseline examination and at a follow-up examination 2 years later. Levels of ADL were assessed with the Barthel Index. Pearson's correlation coefficients were calculated for serum albumin, ADL, and handgrip strength for baseline values and for their 2-year changes (Δ).

Results: At baseline, the mean serum albumin concentration was 4.0 g/dL and the total score of the Barthel Index (baseline Barthel Index) was 71.1. The baseline serum albumin level correlated significantly with the baseline Barthel Index (r = 0.287) and baseline handgrip strength (r = 0.315), but not with Δ Barthel Index (r = 0.096) or Δ handgrip strength (r = - 0.058). The Δ serum albumin correlated significantly with Δ Barthel Index (r = 0.296), but not with Δ handgrip strength (r = 0.182), baseline Barthel Index (r = - 0.044), or baseline handgrip strength (r = 0.047).

Conclusions: This 2-year cohort study has demonstrated that a decrease in serum albumin levels is associated with a decrease in ADL levels. A third factor may play a role in adversely affecting both serum albumin and ADL levels in frail elderly people.