Format

Send to

Choose Destination
Intensive Care Med. 2019 Nov;45(11):1599-1607. doi: 10.1007/s00134-019-05790-z. Epub 2019 Oct 8.

ICU staffing feature phenotypes and their relationship with patients' outcomes: an unsupervised machine learning analysis.

Author information

1
Graduate Program in Translational Medicine, Department of Critical Care, D'Or Institute for Research and Education, Rua Diniz Cordeiro, 30. Botafogo, Rio De Janeiro, 22281-100, Brazil.
2
Research Institute, HCor-Hospital do Coração, São Paulo, Brazil.
3
Department of Research and Development, Epimed Solutions, Rio De Janeiro, Brazil.
4
ICU, Hospital Sírio Libanês, São Paulo, Brazil.
5
Clinical Research, Investigation, and Systems Modeling of Acute Illness (CRISMA) Center, Department of Critical Care Medicine, University of Pittsburgh School of Medicine, Pittsburgh, PA, USA.
6
Department of Health Policy & Management, University of Pittsburgh Graduate School of Public Health, Pittsburgh, PA, USA.
7
ICU, Hospital Copa D'Or, Rio De Janeiro, Brazil.
8
ICU, Hospital Quinta D'Or, Rio De Janeiro, Brazil.
9
Adult ICU, Hospital Israelita Albert Einstein, São Paulo, Brazil.
10
ICU, Hospital Santa Paula, São Paulo, Brazil.
11
ICU, Hospital Santa Luzia Rede D'Or São Luiz DF, Brasília, Brazil.
12
ICU, Hospital São Francisco, Ribeirão Preto, Brazil.
13
ICU, UDI Hospital, São Luís, Brazil.
14
ICU, Hospital São Lucas, Rio De Janeiro, Brazil.
15
ICU, Hospital Estadual Alberto Torres, São Gonçalo, Brazil.
16
ICU, Hospital Barra D'Or, Rio De Janeiro, Brazil.
17
ICU, Hospital Santa Rita, Santa Casa de Misericórdia de Porto Alegre, Porto Alegre, Brazil.
18
ICU, Hospital Unimed Vitoria, Vitoria, Brazil.
19
ICU, Hospital da Luz, Vila Mariana, São Paulo, Brazil.
20
Instituto Nacional de Infectologia Evandro Chagas, Fundação Oswaldo Cruz, Rio De Janeiro, Brazil.
21
Graduate Program in Translational Medicine, Department of Critical Care, D'Or Institute for Research and Education, Rua Diniz Cordeiro, 30. Botafogo, Rio De Janeiro, 22281-100, Brazil. marciosoaresms@gmail.com.

Abstract

PURPOSE:

To study whether ICU staffing features are associated with improved hospital mortality, ICU length of stay (LOS) and duration of mechanical ventilation (MV) using cluster analysis directed by machine learning.

METHODS:

The following variables were included in the analysis: average bed to nurse, physiotherapist and physician ratios, presence of 24/7 board-certified intensivists and dedicated pharmacists in the ICU, and nurse and physiotherapist autonomy scores. Clusters were defined using the partition around medoids method. We assessed the association between clusters and hospital mortality using logistic regression and with ICU LOS and MV duration using competing risk regression.

RESULTS:

Analysis included data from 129,680 patients admitted to 93 ICUs (2014-2015). Three clusters were identified. The features distinguishing between the clusters were: the presence of board-certified intensivists in the ICU 24/7 (present in Cluster 3), dedicated pharmacists (present in Clusters 2 and 3) and the extent of nurse autonomy (which increased from Clusters 1 to 3). The patients in Cluster 3 exhibited the best outcomes, with lower adjusted hospital mortality [odds ratio 0.92 (95% confidence interval (CI), 0.87-0.98)], shorter ICU LOS [subhazard ratio (SHR) for patients surviving to ICU discharge 1.24 (95% CI 1.22-1.26)] and shorter durations of MV [SHR for undergoing extubation 1.61(95% CI 1.54-1.69)]. Cluster 1 had the worst outcomes.

CONCLUSION:

Patients treated in ICUs combining 24/7 expert intensivist coverage, a dedicated pharmacist and nurses with greater autonomy had the best outcomes. All of these features represent achievable targets that should be considered by policy makers with an interest in promoting equal and optimal ICU care.

KEYWORDS:

Cluster analysis; ICU organization; Intensive care unit; Nurse autonomy; Outcomes; Staffing features

PMID:
31595349
DOI:
10.1007/s00134-019-05790-z

Supplemental Content

Full text links

Icon for Springer
Loading ...
Support Center