This review will describe the evidence for changing the hospital environment to improve nutrition of older people, with particular emphasis on the role of additional mealtime assistance. Poor nutrition among older people in hospital is well recognised in many countries and is associated with poor outcomes of hospital care including increased mortality and longer lengths of stay. Factors recognised to contribute to poor dietary intake include acute illness, co-morbidities, cognitive impairment, low mood and medication. The hospital environment has also been scrutinised with reports from many countries of food being placed out of reach or going cold because time-pressured ward and catering staff often struggle to help an increasingly dependent group of patients at mealtimes. Routine screening in hospital for people at risk of under nutrition is recommended. Coloured trays and protected mealtimes are widespread although there is relatively little evidence for their impact on dietary intake. Volunteers can be trained to sfely give additional mealtime assistance including feeding to older patients on acute medical wards. They can improve the quality of mealtime care for patients and nursing staff although the evidence for improved dietary intake is mixed. In conclusion, improving the nutrition of older patients in hospital is challenging. Initiatives such as routine screening, the use of coloured trays, protected mealtimes and additional mealtime assistance can work together synergistically. Volunteers are likely to be increasingly important in an era when healthcare systems are generally limited in both financial resources and the ability to recruit sufficient nursing staff.

The present study was aimed to validate two normalised short-form (SF) Mini-Nutritional Assessments (MNA) that contained either BMI (Taiwan version 1, T1) or calf circumference (CC; Taiwan version 2, T2) for rating the nutritional status of elderly Taiwanese. Both versions adopted Taiwanese anthropometric cut-offs, but T2 further had the BMI item replaced by CC. We compared the ability of the two SF in rating the nutritional status of 2674 elderly Taiwanese in the 1999 ‘Taiwan Longitudinal Survey on Aging’ with their respective full versions. We evaluated the agreement between the SF and full scales with weighted κ and performed Cox regression analysis for the follow-up 4-year mortality according to nutritional status rated at baseline. The results showed that the agreements between the respective SF and the full MNA were good (weighted κ: 0·679 for T1 and 0·667 for T2). Both SF performed well in predicting follow-up 4-year mortality relative to the full MNA. In conclusion, MNA-T1-SF and -T2-SF have good consistency with the full MNA and have quite comparable abilities in rating the nutritional status of elderly Taiwanese. Both SF versions appear appropriate for functioning as stand-alone units for rating the nutritional status of the elderly in community-living settings or the general population.

Physical functional ability and nutritional status are two major indicators for predicting the risk of mortality in older adults. The present study examined the complementarity of the Activities of Daily Living (ADL) and the Mini-Nutritional Assessment (MNA) for predicting follow-up 4-year all-cause mortality risk in elderly Taiwanese. We analysed data of the ‘Survey of Health and Living Status of the Elderly in Taiwan’, a population-based longitudinal cohort study which involved 2872 men and women of ≥ 65 years old at baseline (1999). We rated their functional dependency with the ADL scale and nutritional status with the MNA (both the long form, LF and the short form, SF) at baseline, and analysed the complementarity of the two scales in predicting follow-up 4-year all-cause mortality with Cox regression analysis and the net reclassification improvement (NRI) to quantify the improvement. The results showed that both ADL and MNA offered improvement in predicting follow-up mortality risk beyond that predicted by either one alone according to the Akaike information criterion and the NRI. The MNA-SF was nearly as effective as the MNA-LF in improving the predictive ability of the ADL. The present study suggests that the MNA (especially the SF because of its simplicity and time-saving feature) together with the ADL scale might be of value for predicting the mortality risk of frail elderly living in various settings.

Nutrition is a key element in geriatric health and is important for functional ability. The present study examined the functional status-predictive ability of the Mini-Nutritional Assessment (MNA). We analysed the dataset of the ‘Survey of Health and Living Status of the Elderly in Taiwan’, a population-based study conducted by the Bureau of Health Promotion of Taiwan. Study subjects ( ≥ 65 years old) who completed both the 1999 and 2003 surveys were rated with the long form and short form of the MNA at baseline and with the Activities of Daily Living (ADL) and the Instrument Activities of Daily Living (IADL) scales 4 years later (end-point). The ability of the MNA to predict ADL or IADL dependency was evaluated with logistic regression models. The results showed that the elderly who were rated malnourished or at risk of malnutrition at baseline generally had significantly higher ADL or IADL scores 4 years later. Lower baseline MNA scores also predicted a greater risk of ADL or IADL dependency. These associations exist even among the elderly who were free of ADL or IADL dependency at baseline. The results clearly indicate that the MNA is able to predict ADL and IADL dependency (in addition to rating current nutritional status) of the elderly. The MNA, especially the short form, should be a valuable tool for identifying elderly at risk of functional decline and/or malnutrition in clinical practice or community programmes.