Over a quarter of older inpatients leave hospital less able than they arrived

The edition in brief

A meta-analysis of 46 studies and 439,406 older adults from 15 regions found a pooled prevalence of hospitalisation-associated disability of 27% (95% CI 24% to 31%), with individual studies ranging from 5% to 57% and a 95% prediction interval of 7% to 48%. Rates were higher in patients aged 75 to 89, in those admitted with lower respiratory tract infection, and where a pre-admission baseline was used rather than an admission-day assessment. In 200 patients aged 70 or over referred for coronary angiography, two cardiologists independently assessed frailty by the Fried test and by eyeballing: frailty prevalence was 16.5%, and eyeballing had sensitivity 93.9%, negative predictive value 98.5%, specificity 80.8%, positive predictive value 49.2% and area under the curve 0.87 - reliable for excluding frailty, poor for confirming it, with specificity lower in women and inpatients. A meta-analysis of 8 studies in Parkinson's disease found higher anticholinergic burden significantly associated with fracture (OR 1.53, 95% CI 1.28 to 1.84) and emergency department attendance (OR 1.25, 1.15 to 1.36), while the associations with dementia (OR 1.71, 0.93 to 3.17) and cognitive decline did not reach significance overall. And pooling 7 randomised trials in 472 older people with sarcopenia, strength training with whey protein or HMB improved fat-free mass, handgrip strength, chair stand and six-minute walk, with the handgrip benefit seen for whey protein but not for HMB.