This longitudinal observation study investigates associations between nutritional risk in adults aged 65+ and use of health services, morbidity and mortality. The first hypothesis is that older service users at nutritional risk will use a higher number of services/consultations and have a higher morbidity and mortality, compared to older patients not at nutritional risk. Further, the study evaluates the usefulness of the tool 'nutrition plan'. The second hypothesis is that older service users at nutritional risk who have a nutrition plan have lower subsequent morbidity and mortality, compared to older service users at nutritional risk without a nutrition plan.

The study is divided into two main parts, based on two different populations.

Study A: Article 1 is based on 'The Undernutrition Registry' in Helse Bergen Local Health Authority, Norway. The Undernutrition registry is a cohort of patients (n=20 000) screened for nutritional risk at Haukeland University Hospital (HUS) during point prevalence surveys between 2008 and 2018. The point prevalence surveys included all adult patients in somatic departments, and were performed 2-4 times per year. Detailed descriptions of the data collection can be found in Tangvik et al. (2012). Previous analysis showed that 43% of the patients in the sample were above 70 years, thus we expect the sample in the study to constitute approx. 9000 patients ≥ 65 years. Nutritional risk was determined by 'Nutritional Risk Screening 2002' (NRS2002). Information on nutritional risk will be linked to data on health care use and outcomes from the hospital's patient administrative electronic database (length of stay, number of hospital stays, morbidity in terms of number and severity of diagnoses and death).

Study B: Articles 2 and 3 are based on a cohort of older service users (≥65 years) with information on nutritional variables from the 'Kommunalt pasientregister' (KPR)-registry in the period 2016-2018 (N=270,560). Municipalities report data on nutritional risk screening and nutritional plan for individual service users in KPR. These nutritional data will be linked to data from 'The Norwegian Patient Registry' (NPR). NPR includes information about diagnosis and use of specialist services for all patients who have received or are waiting for treatment/consultations in specialist health care services (hospitals and outpatient clinics).

Nutritional status and use of 'nutritional plan' will be entered as independent variables in regression and survival analysis models with health service use (type, extent) or mortality as outcome variables.

Inclusion Criteria:

• Study A. Patients in hospital aged 65+
• Study B. All service users in municipalities aged 65+

Exclusion Criteria:

• Age <65 years

Two cohorts of adults aged 65+: Cohort A includes in-patients in Haukeland University Hospital (HUS), cohort B includes services users in all Norwegian municipalities

Necessary legal and practical arrangements for data sharing have not yet been implemented in our region.