Predictive Models of Readmission in Heart Failure

Objective: To create models for predicting the risk of readmission short term (30-90 days) and medium term (one year).To identify differences between patients readmitted and not readmitted, and profiles of frequently preventable readmissions in our setting.

Methodology: Observational prospective cohort of patients who are discharged with a diagnosis of heart failure in 5 participating centers. further, a nested case-control will be developed in the previous cohort, being cases those readmitted and controls will be chosen according to age, sex, etiology of heart failure, comorbidities and functional status. Sociodemographic, clinical and health related quality of life, empowerment and baseline self-efficacy as predictors of readmission were collected.

Regarding the cases and controls, the quality of life in post-discharge transitional period, empowerment and self-efficacy, caregiver burden, social support, discharge care model, use of health services and adequation, conciliation and adherence to treatment will be assessed . Cox proportional hazards models will be created, and conditional logistic regression models to identify differences between cases and controls.

Expected Results: National Health System will be provided with tools for predicting the risk of readmission useful to clinicians and managers to offer discharge individualized care. The most common characteristics of readmissions and preventable readmissions will be identified, which will be helpful to create specific actions in the future.

Inclusion criteria

• Patients admitted by acute heart failure (de novo and acute decompensated) (International Classification of Diseases, Ninth Revision, Clinical Modification: 428.x; some of the 402.x group), older tan 18 who will accept to participate.

Exclusion criteria

• patients who develop HF episodes during admission, if have been admitted for another reason,
• patients transferred from other health centers,
• myocardial infarction or stroke in the 4 weeks prior to admission,
• life expectancy less than one year, for HF at the terminal or cause different from the CI according to the assessment of the study subjects,
• inability to complete the questionnaires or with assistance external (reviewer, family, social) because of sensorineural reason, dementia or ignorance of the language.