Home Care Clients’ Risk of Unplanned Hospitalization: Predictors and Risk Classification

Jukka Rönneikkö

Research output: Book/ReportDoctoral thesisCollection of Articles

Abstract

Background. Home care clients are prone to many adverse outcomes, including unplanned hospitalization. Identifying factors predicting unplanned hospitalization and the characteristics of hospitalized home care clients would help to identify home care clients at risk and to target them with preventive interventions. The Resident Assessment Instrument-Home Care (RAI-HC) tool has been developed for the systematic comprehensive assessment of home care clients’ health state and care needs.

Objectives. By utilizing RAI-HC, the aim was to identify potentially modifiable conditions predicting unplanned hospitalization, determine discharge diagnoses and their associations with patient characteristics, and study whether the risk of unplanned hospitalization of home care clients could be rated using RAI-HC-based scales. Finally, the aim was to identify variables that could improve the accuracy of a RAI-HC-based risk prediction model.

Materials and methods. All studies were retrospective register-based studies: studies I and II were based on the first RAI-HC assessments and nationwide hospital discharge records of a total of 15,700 Finnish new home care clients, and studies III and IV were based on the RAI-HC assessments and hospital discharge records of 3,091 home care clients in the city of Tampere, Finland. Study IV focused especially on 1,972 clients aged 80 years and older.

The information of the home care clients’ RAI-HC assessments was connected to the information of their first hospitalization after the assessment with follow-up times in studies I and II of one year and in studies III and IV of 180 days. In studies I, II, and IV, regression analyses were used to evaluate the associations of RAI-HC variables with hospitalization and discharge diagnoses, and in study IV, with the different risk levels of the Detection of Indicators and Vulnerabilities for Emergency Room Trips (DIVERT) scale also. In study III, to compare the predictive accuracy of DIVERT and other RAI-HC based scales in relation to hospitalization, the Area Under the Curves (AUC) was calculated. Based on the detected associations, the aim was to modify the DIVERT scale. To study the predictive accuracy of the modified DIVERT scales for unplanned hospitalization, the AUCs were determined.

Results. Altogether 43% of the new home care clients were hospitalized at least once within one year, while 64% of those aged 80 years or older were hospitalized at least once during the study period of two years. The most significant risk factor for hospitalization was previous hospitalization during the year preceding the assessment. The most common cause for hospitalization was an infectious disease, with urinary tract infection the most common single diagnosis. The likelihood of infectious diseases was increased by skin ulcers, decreased physical and cognitive function, and daily urinary incontinence at the beginning of home care services. Indicators that seemed to protect clients from hospitalization were a body mass index ≥24 and the clients’ own belief that functional capacity could improve. Feelings of loneliness increased the probability of being hospitalized because of geriatric symptoms without a specific diagnosis.

The accuracy of the DIVERT scale in differentiating the risk of unplanned hospitalization of home care clients was relatively poor (AUC 0.62) and was weakened with age. Clients with high DIVERT scores were hospitalized earlier than others. Of the other scales studied, none of them had better predictive validity than DIVERT. Adding two variables – urinary incontinence and cognitive impairment – to the DIVERT algorithm did not improve the AUC when compared with the original scale.

Conclusions. The social and medical situation of new home care clients is often complex, and their situation should be assessed at the beginning of home care services carefully by an interdisciplinary team. RAI-HC provides an opportunity to identify the risk of unplanned hospitalization and to target necessary interventions to those at increased risk.
Original languageEnglish
Place of PublicationTampere
ISBN (Electronic) 978-952-03-2839-9
Publication statusPublished - 2023
Publication typeG5 Doctoral dissertation (articles)

Publication series

NameTampere University Dissertations - Tampereen yliopiston väitöskirjat
Volume774
ISSN (Print)2489-9860
ISSN (Electronic)2490-0028

Fingerprint

Dive into the research topics of 'Home Care Clients’ Risk of Unplanned Hospitalization: Predictors and Risk Classification'. Together they form a unique fingerprint.

Cite this