Development and validation of a life expectancy estimator for multimorbid older adults: a cohort study protocol.

Gastens, Viktoria; Del Giovane, Cinzia; Anker, Daniela; Feller, Martin; Syrogiannouli, Lamprini; Schwab, Nathalie; Bauer, Douglas C; Rodondi, Nicolas; Chiolero, Arnaud (2021). Development and validation of a life expectancy estimator for multimorbid older adults: a cohort study protocol. BMJ open, 11(8), e048168. BMJ Publishing Group 10.1136/bmjopen-2020-048168

[img]
Preview
Text
Gastens__BMJ_Open_2021.pdf - Published Version
Available under License Creative Commons: Attribution (CC-BY).

Download (325kB) | Preview

BACKGROUND

Older multimorbid adults have a high risk of mortality and a short life expectancy (LE). Providing high-value care and avoiding care overuse, including of preventive care, is a serious challenge among multimorbid patients. While guidelines recommend to tailor preventive care according to the estimated LE, there is no tool to estimate LE in this specific population. Our objective is therefore to develop an LE estimator for older multimorbid adults by transforming a mortality prognostic index, which will be developed and internally validated in a prospective cohort.

METHODS AND ANALYSIS

We will analyse data of the Optimising Therapy to Prevent Avoidable Hospital Admissions in Multimorbid Older People cohort study in Bern, Switzerland. 822 participants were included at hospitalisation with age of 70 years or older, multimorbidity (three or more chronic medical conditions) and polypharmacy (use of five drugs or more for >30 days). All-cause mortality will be assessed during 3 years of follow-up. We will apply a flexible parametric survival model with backward stepwise selection to identify the mortality risk predictors. The model will be internally validated using bootstrapping techniques. We will derive a point-based risk score from the regression coefficients. We will transform the 3-year mortality prognostic index into an LE estimator using the Gompertz survival function. We will perform a qualitative assessment of the clinical usability of the LE estimator and its application. We will conduct the development and validation of the mortality prognostic index following the Prognosis Research Strategy (PROGRESS) framework and report it following the Transparent Reporting of a Multivariable Prediction Model for Individual Prognosis or Diagnosis (TRIPOD) statement.

ETHICS AND DISSEMINATION

Written informed consent by patients themselves or, in the case of cognitive impairment, by a legal representative, was required before enrolment. The local ethics committee (Kantonale Ethikkommission Bern) has approved the study. We plan to publish the results in peer-reviewed journals and present them at national and international conferences.

Item Type:

Journal Article (Further Contribution)

Division/Institute:

04 Faculty of Medicine > Department of General Internal Medicine (DAIM) > Clinic of General Internal Medicine
04 Faculty of Medicine > Medical Education > Institute of General Practice and Primary Care (BIHAM)
04 Faculty of Medicine > Department of General Internal Medicine (DAIM) > Clinic of General Internal Medicine > Centre of Competence for General Internal Medicine
04 Faculty of Medicine > Department of General Internal Medicine (DAIM)

Graduate School:

Graduate School for Health Sciences (GHS)

UniBE Contributor:

Gastens, Viktoria; Del Giovane, Cinzia; Anker, Daniela; Feller, Martin; Syrogiannouli, Lamprini; Schwab, Nathalie Christa and Rodondi, Nicolas

Subjects:

600 Technology > 610 Medicine & health
300 Social sciences, sociology & anthropology > 360 Social problems & social services

ISSN:

2044-6055

Publisher:

BMJ Publishing Group

Funders:

[4] Swiss National Science Foundation ; [201] Staatssekretariat für Bildung, Forschung und Innovation (SBFI) = Swiss State Secretariat for Education, Research and Innovation (SERI)

Language:

English

Submitter:

Tobias Tritschler

Date Deposited:

30 Aug 2021 18:43

Last Modified:

07 Sep 2021 17:17

Publisher DOI:

10.1136/bmjopen-2020-048168

PubMed ID:

34433596

Uncontrolled Keywords:

preventive medicine public health statistics & research methods

BORIS DOI:

10.48350/159080

URI:

https://boris.unibe.ch/id/eprint/159080

Actions (login required)

Edit item Edit item
Provide Feedback