The aim of the study is to estimate the prevalence of malnutrition among older adults admitted to hospital and to analyse predictive factors for malnutrition. Further, the aim was to analyse the association between nutritional status and all-cause and cause-specific mortality.

The hypothesis is that malnourished patients have a higher mortality then well-nourished patients after three years.

The baseline survey was a cross-sectional study of 1771 patients aged ≥65 years who were admitted to hospital. Nutritional status was assessed using the MNA instrument, and possible risk factors associated with malnutrition were recorded during the hospital stay. All-cause mortality was followed up after median 3.5 years s in a cohort study of 1767 participants. Cause-specific mortality was followed up after median 4.7 years in a cohort of 1767 participants. Previous dietary intake in middle-aged and older adults was recorded at basline in 725 participants, who had their nutritional status assessed with the MNA when admitted to hospital 10 years later.

Inclusion Criteria:

• Patients 65 years or older

Exclusion Criteria:

• unable to communicate and no relative could anwer the questions (n = 356)
• receiving palliative care (n = 44)
• impossible to measure height or weight (n = 22)
• miscellanious (n = 71)

The included sample comprised 1771 patients 65 years and older, consecutively admitted during fifteen months to two internal medicine wards (n = 706), two surgical wards (n = 681), and one orthopaedic ward (n = 384) at the Central hospital in Västerås, Sweden.