Reliability of a short questionnaire for the diagnosis of severe disability in polypathological patients attended in hospital setting.

Background: A comprehensive evaluation of polypathological patients (PP) should always include a functional evaluation. For this purpose, a modified version of the Barthel Index (BI) is the most applied questionnaire, and it consists of a 10-variable scale. The aim of this study was to develop a screening and confirmation tool to diagnose high disability with the fewest number of dimensions of the BI as possible.

Method: This present cross-sectional observational multicentre study included PP attended in 36 Spanish hospitals that were divided into two geographical areas (Western and Eastern). The Western area was considered to be the derivation subgroup of PP, and the Eastern area was the validation subgroup. Complete disability for each item (value of 0) was assessed for the diagnosis of severe disability. Diagnostic validity indices (sensitivity, specificity, negative and positive predictive values [NPV and PPV, respectively], and negative and positive likelihood ratios [NLR and PLR, respectively]) were determined for the derivation subgroup. The dimensions with the best diagnostic validity indices were then used to evaluate the validation subgroup.

Results: The analysis included 1521 PP, 753 PP from the Western area and 768 PP from the Eastern area. Needing complete help for bathing showed the highest NPV and lowest NLR in the derivation/validation subgroups (NPV 96.87/95.54, NLR 0.07/0.13). Being disabled for feeding alone showed high PPV and PLR values (PPV 97.97/95.65, PLR 109.25/49.62), as did disability for transfers (PPV 98.48/97.96, PLR 143.36/107.68). In addition, complete disability for feeding and transfers had the best PPV and PLR in both subgroups (PPV 100/100, PLR X/0).

Conclusions: A two-dimension mini-Barthel Index may represent a reliable diagnostic test for severe disability in PP.