acquisition and short-term retention of metered dose inhaler and Turbohaler techniques is unlikely to be successful in frail elderly people who have an abnormal Mini-Mental State Examination and/or EXIT25 test. The latter test, when abnormal, is probably the superior predictor of inability to learn inhaler techniques.

Objectives: to compare the use of two falls risk-identification tools (Downton and STRATIFY) with clinical judgment (based upon the observation of wandering behaviour) in predicting falls of medically stable patients in a rehabilitation ward for older people. Methods: in a prospective observational study, with blinded end-point evaluation, 200 patients admitted to a geriatric rehabilitation hospital had a STRATIFY and Downton Fall Risk assessment and were observed for wandering behaviour. Results:wandering had a predictive accuracy of 78%. A total of 157/200 were identified correctly compared to 100/200 using the Downton score (P<0.0001 95%, CI 0.18-0.42), or 93/200 using STRATIFY (P<0.0001; 95% CI 0.15-0.37).The Downton and STRATIFY tools demonstrated predictive accuracies of 50% and 46.5%, respectively, with no statistical significance between the two (P = 0.55; 95% CI 0.77-1.71). Sensitivity for predicting falls using wandering was 43.1% (22/51). This was significantly worse than Downton 92.2% (47/51: P<0.001) and STRATIFY 82.3% (42/51: P<0.001). Conclusions: this study showed that clinical observation had a higher accuracy than two used falls risk-assessment tools. However it was significantly less sensitive implying that fewer patients who fell were correctly identified as being at risk.

Significant differences were identified in the performance and complexity between the four risk assessment tools studied. The STRATIFY tool was the shortest and easiest to complete and had the highest predictive value but the lowest sensitivity.

No abstract

Clinical observations have shown that some older patients are unable to learn to use a metered dose inhaler (MDI) despite having a normal abbreviated mental test (AMT) score, possibly because of dyspraxia or unrecognised cognitive impairment. Thirty inhaler-naive inpatients (age 76-94) with an AMT score of 8-10 (normal) were studied. Standard MDI training was given and the level of competence reached was scored (inhalation score). A separate observer performed the minimental test (MMT), Barthel index, geriatric depression score (GDS), ideational dyspraxia test (IDT), and ideomotor dyspraxia test (IMD). No correlative or threshold relationship was found between inhalation score and Barthel index, GDS, or IDT. However, a significant correlation was found between inhalation score and IMD (r = 0.45, p = 0.039) and MMT (r = 0.48, p = 0.032) and threshold effects emerged in that no subject with a MMT score of less than 23/30 had an inhalation score of 5/10 or more (adequate technique requires 6/10 or more), and all 17/18 with an inhalation score of 6/10 or more had an IMD of 14/20 or more.The three patients with a MMT >22 and inhalation score <6 had abnormal IMD scores. Inability to learn an adequate inhaler technique in subjects with a normal AMT score appears to be due to unrecognised cognitive impairment or dyspraxia. The MMT is probably a more useful screening test than the AMT score in this context.

Background: Recurrent fallers constitute a minority of patients who fall but contribute considerably to the total number of falls recorded. Objective: To study the characteristics of recurrent fallers in a hospital setting. Methods: In a prospective observational study we investigated the characteristics of 1,025 patients admitted to a geriatric non-acute hospital. Patients were followed until discharge and were classified as non-fallers, single fallers or recurrent fallers. Results: We identified 824 non-fallers, 136 single fallers and 65 recurrent fallers contributing 175 falls. Compared to non-fallers, recurrent fallers were more likely to have pre-admission falls (p = 0.004), confusion (p < 0.0001), an unsafe gait (p = 0.0001) and be on tranquillisers (p = 0.018) and antidepressants (p = 0.006). They had longer stays in hospital (p < 0.0001) and more nursing home discharges (p = 0.0001). There was considerable overlap with risk factors for single fallers but compared to this group they were more likely to be confused (p = 0.027), and on antidepressant medication (p = 0.009). They also had a longer length of stay (p < 0.001) and more nursing home discharges (p = 0.03). Confusion (p = 0.0001), unsafe gait (p = 0.0006) and antidepressants (p = 0.018) were independently associated with recurrent falls. Conclusions: It is important to recognise the risk factors that prospectively identify a recurrent faller because of the significant contribution to total falls by a relatively small number of patients. This may be useful not only in trying to reduce total falls but also in trying to reduce injury.

No abstract

These factors should be the target of specific rehabilitation in an attempt to reduce the risk of discharge to a nursing home and improve patient outcome.