Staffing levels and hospital mortality in England: a national panel study using routinely collected data
Staffing levels and hospital mortality in England: a national panel study using routinely collected data
Objectives Examine the association between multiple clinical staff levels and case-mix adjusted patient mortality in English hospitals. Most studies investigating the association between hospital staffing levels and mortality have focused on single professional groups, in particular nursing. However, single staff group studies might overestimate effects or neglect important contributions to patient safety from other staff groups.
Design: Retrospective observational study of routinely available data.
Setting and participants: 138 National Health Service hospital trusts that provided general acute adult services in England between 2015 and 2019. Outcome measure: Standardised mortality rates were derived from the Summary Hospital level Mortality Indicator data set, with observed deaths as outcome in our models and expected deaths as offset. Staffing levels were calculated as the ratio of occupied beds per staff group. We developed negative binomial random-effects models with trust as random effects.
Results:Hospitals with lower levels of medical and allied healthcare professional (AHP) staff (e.g, occupational therapy, physiotherapy, radiography, speech and language therapy) had significantly higher mortality rates (rate ratio: 1.04, 95% CI 1.02 to 1.06, and 1.04, 95% CI 1.02 to 1.06, respectively), while those with lower support staff had lower mortality rates (0.85, 95% CI 0.79 to 0.91 for nurse support, and 1.00, 95% CI 0.99 to 1.00 for AHP support). Estimates of the association between staffing levels and mortality were stronger between-hospitals than within-hospitals, which were not statistically significant in a within-between random effects model.
Conclusions: In additional to medicine and nursing, AHP staffing levels may influence hospital mortality rates. Considering multiple staff groups simultaneously when examining the association between hospital mortality and clinical staffing levels is crucial. Trial registration number NCT04374812.
Health policy, Organisation of health services, Quality in health care
Rubbo, Bruna
938f54c6-0d65-4b5a-99be-b4dfbdeadc83
Saville, Christina
2c726abd-1604-458c-bc0b-daeef1b084bd
Dall'ora, Chiara
4501b172-005c-4fad-86da-2d63978ffdfd
Turner, Lesley
7c4a1fe5-21a1-4634-a1cc-0230322603d1
Jones, Jeremy
270b303b-6bad-4be7-8ea0-63d0e8015c91
Ball, Jane
85ac7d7a-b21e-42fd-858b-78d263c559c1
Culliford, David
25511573-74d3-422a-b0ee-dfe60f80df87
Griffiths, Peter
ac7afec1-7d72-4b83-b016-3a43e245265b
17 May 2023
Rubbo, Bruna
938f54c6-0d65-4b5a-99be-b4dfbdeadc83
Saville, Christina
2c726abd-1604-458c-bc0b-daeef1b084bd
Dall'ora, Chiara
4501b172-005c-4fad-86da-2d63978ffdfd
Turner, Lesley
7c4a1fe5-21a1-4634-a1cc-0230322603d1
Jones, Jeremy
270b303b-6bad-4be7-8ea0-63d0e8015c91
Ball, Jane
85ac7d7a-b21e-42fd-858b-78d263c559c1
Culliford, David
25511573-74d3-422a-b0ee-dfe60f80df87
Griffiths, Peter
ac7afec1-7d72-4b83-b016-3a43e245265b
Rubbo, Bruna, Saville, Christina, Dall'ora, Chiara, Turner, Lesley, Jones, Jeremy, Ball, Jane, Culliford, David and Griffiths, Peter
(2023)
Staffing levels and hospital mortality in England: a national panel study using routinely collected data.
BMJ Open, 13 (5), [e066702].
(doi:10.1136/bmjopen-2022-066702).
Abstract
Objectives Examine the association between multiple clinical staff levels and case-mix adjusted patient mortality in English hospitals. Most studies investigating the association between hospital staffing levels and mortality have focused on single professional groups, in particular nursing. However, single staff group studies might overestimate effects or neglect important contributions to patient safety from other staff groups.
Design: Retrospective observational study of routinely available data.
Setting and participants: 138 National Health Service hospital trusts that provided general acute adult services in England between 2015 and 2019. Outcome measure: Standardised mortality rates were derived from the Summary Hospital level Mortality Indicator data set, with observed deaths as outcome in our models and expected deaths as offset. Staffing levels were calculated as the ratio of occupied beds per staff group. We developed negative binomial random-effects models with trust as random effects.
Results:Hospitals with lower levels of medical and allied healthcare professional (AHP) staff (e.g, occupational therapy, physiotherapy, radiography, speech and language therapy) had significantly higher mortality rates (rate ratio: 1.04, 95% CI 1.02 to 1.06, and 1.04, 95% CI 1.02 to 1.06, respectively), while those with lower support staff had lower mortality rates (0.85, 95% CI 0.79 to 0.91 for nurse support, and 1.00, 95% CI 0.99 to 1.00 for AHP support). Estimates of the association between staffing levels and mortality were stronger between-hospitals than within-hospitals, which were not statistically significant in a within-between random effects model.
Conclusions: In additional to medicine and nursing, AHP staffing levels may influence hospital mortality rates. Considering multiple staff groups simultaneously when examining the association between hospital mortality and clinical staffing levels is crucial. Trial registration number NCT04374812.
Text
350_Staffing levels and hospital mortality in England a national panel study using routinely collected data
- Version of Record
More information
Accepted/In Press date: 27 April 2023
Published date: 17 May 2023
Additional Information:
© Author(s) (or their employer(s)) 2023. Re-use permitted under CC BY. Published by BMJ.
Keywords:
Health policy, Organisation of health services, Quality in health care
Identifiers
Local EPrints ID: 477982
URI: http://eprints.soton.ac.uk/id/eprint/477982
ISSN: 2044-6055
PURE UUID: 793f2579-24e0-4397-b722-b8a6cc58e47b
Catalogue record
Date deposited: 19 Jun 2023 16:39
Last modified: 17 Mar 2024 03:57
Export record
Altmetrics
Contributors
Author:
Bruna Rubbo
Author:
Jane Ball
Author:
David Culliford
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics