Individual and Environmental Risk Factors of Unplanned Admissions of Elderly

Background: Chronic diseases failed concerned people, notably elderly which accumulate them. Failure of failed organ in oldest old citizens (80 and over) often leads to unplanned hospitalization which may have dreadful consequences, such as worsening condition or even death. Few studies have sought to identify all individual and environmental risk factors for unplanned hospitalization of elderly citizens. The development of a nation-wide shared medical record would allow to build an alert system if such factors can be modelled.

Objective: The aim of this multicenter study is to investigate risk factors for unplanned hospitalizations, including individual and environmental factors, notably air pollution. Classifying at risk patients might help preventing unplanned hospitalization of oldest old. Correlations between medical data and the atmospheric pollution modelling will allow us to enhance the forecast system with dose-effect functions. Validation of a multi dimensional standardised instrument (InterRAI-MDS) will help in the development of a shared medical record.

Methods : Patient aged 80 and over presenting to the emergency department (ED) for medical concern eventually associated with social concern and admitted in a medical or surgical ward. Three hospitals, two located in the Paris area and one in Champagne, offering emergency care and providing geriatric services will be involved in the study. 1,200 patients will be prospectively included in the study during a one-year period.

• Individual study

Data collection involves three steps (before, during and after health event) from hospital admission (i.e. study inclusion) and relies on a multi dimensional standardized geriatric assessment using well established scales :

Retrospective data collection includes medical history, health care utilization, SOCIO-economic status and functional status the days before hospital admission ; Data collection at hospital admission includes functional, mental, nutritional and depression status assessed by geriatric physician and nurse (standardized geriatric assessment) ; social support and care plan will be recorded at discharge ; SOCIO-economic conditions will also be recorded Data collection during follow-up includes health and functional status assessed at three, six, nine months and one year after first emergency visit, with help of the family and family physician. Hospitalizations and death will be recorded.

• Environmental and contextual data Meteorological data as well as data on pollution, epidemics, strikes in health care system or public transportation and public holiday will be gathered during the study period. The use of the atmospheric transport chemistry model CHIMERE will give greater spatial (2 to 3 km) and temporal (hour) resolution than airborne measurements of pollutants.
• Analysis A bidirectional case-crossover design will be used. Cases serve as their own controls through the use of information on the study subjects both before and after the event (hospitalization). Conditional logistic regression will then be used to investigate the short-term health effects of air pollution while taking into account individual characteristics.

Survival analysis with time dependant covariates (environmental factors) will give strengthening results.

A multi dimensional analysis will be performed to determine cluster of patients and patterns of environmental parameters.

Validation of the RAI-MDS instrument will rely on internal coherency, structural validity and external validity facing usual geriatric scales (ADL, IADL, MMSE, MNA, GDS).

Perspectives Identification of individual risk factors as well as evaluation of environmental threat for older people, potentialized by the setting of medical shared across caregivers could allow to develop individualized alarm and prevention systems.

Inclusion criteria

• Patient aged 80 and over presenting to the emergency department (ED) for medical concern eventually associated with social concern and admitted in a medical or surgical ward.
• Three hospitals, two located in the Paris area and one in Champagne, offering emergency care and providing geriatric services will be involved in the study.

Exclusion criteria

• Refusal of the patient or of helping him to sign informed consent
• Refusal of the patient or of helping him to answer
• Premature exit before initial examination