Interventions to reduce mortality from in-hospital cardiac arrest: a mixed-methods study
2019
Background Unchecked patient deterioration can lead to in-hospital cardiac arrest (IHCA) and avoidable death. The National Cardiac Arrest Audit (NCAA) has found fourfold variation in IHCA rates and survival between English hospitals. Key to reducing IHCA is both the identification of patients at risk of deterioration and prompt response. A range of targeted interventions have been introduced but implementation varies between hospitals. These differences are likely to contribute to the observed variation between and within hospitals over time. Objective To determine how interventions aimed at identification and management of deteriorating patients are associated with IHCA rates and outcomes. Design A mixed-methods study involving a systematic literature review, semistructured interviews with 60 NHS staff, an organisational survey in 171 hospitals and interrupted time series and difference-in-difference analyses (106 hospitals). Setting English hospitals participating in the NCAA audit. Participants NHS staff (approximately 300) and patients (13 million). Interventions Education, track-and-trigger systems (TTSs), standardised handover tools and outreach teams. Main outcome measures IHCA rates, survival and hospital-wide mortality. Data sources NCAA, Hospital Episode Statistics, Office for National Statistics Mortality Statistics. Methods A literature review and qualitative interviews were used to design an organisational survey that determined how interventions have been implemented in practice and across time. Associations between variations in services and IHCA rates and survival were determined using cross-sectional, interrupted time series and difference-in-difference analyses over the index study period (2009/10 to 2014/15). Results Across NCAA hospitals, IHCAs fell by 6.4% per year and survival increased by 5% per year, with hospital mortality decreasing by a similar amount. A national, standard TTS [the National Early Warning Score (NEWS)], introduced in 2012, was adopted by 70% of hospitals by 2015. By 2015, one-third of hospitals had converted from paper-based TTSs to electronic TTSs, and there had been an increase in the number of hospitals with an outreach team and an increase in the number with a team available at all times. The extent of variation in the uses of educational courses and structured handover tools was limited, with 90% of hospitals reporting use of standardised communication tools, such as situation, background, assessment and recommendation, in 2015. Introduction of the NEWS was associated with an additional 8.4% decrease in IHCA rates and, separately, a conversion from paper to electronic TTS use was associated with an additional 7.6% decrease. However, there was no associated change in IHCA survival or hospital mortality. Outreach teams were not associated with a change in IHCA rates, survival or hospital mortality. A sensitivity analysis restricted to ward-based IHCAs did not alter the findings but did identify an association between increased outreach team intensity in 2015 and IHCA survival. Limitations The organisational survey was not able to explore all aspects of the interventions and the contextual factors that influenced them. Changes over time were dependent on respondents’ recall. Conclusions Standardisation of TTSs and introduction of electronic TTSs are associated with a reduction in IHCAs. The apparent lack of impact of outreach teams may reflect their mode of introduction, that their effect is through providing support for implementation of TTS or that the organisation of the response to deterioration is not critical, as long as it is timely. Their role in end-of-life decision-making may account for the observed association with IHCA survival. Future work To assess the potential impact of outreach teams at hospital level and patient level, and to establish which component of the TTS has the greatest effect on outcomes. Funding The National Institute for Health Research Health Services and Delivery Research programme.
- Correction
- Source
- Cite
- Save
- Machine Reading By IdeaReader
142
References
10
Citations
NaN
KQI