Calculated In Death Jd Robb Epub Free 16
Berry Spitsberg
This is a mixed-methods, controlled before-and-after study with nested case studies. 31 acute hospitals in England and Wales which introduced COPD care bundles (implementation sites) or provided usual care (comparator sites) were recruited and provided monthly aggregate data. 14 sites provided additional individual patient data. Participants were adults admitted with an acute exacerbation of COPD.
COPD care bundles were not effectively implemented in this study. They were associated with a reduced number of subsequent ED attendances, but not with change in readmissions, mortality or reduced costs. This is unsurprising given the low level of bundle uptake in implementation sites, and it remains to be determined if COPD care bundles affect patient care and outcomes when they are effectively implemented.
Systematic review evidence from a small number of randomised controlled trials (RCTs) suggested that discharge bundles for patients with chronic obstructive pulmonary disease (COPD) lead to fewer readmissions but do not significantly improve mortality or quality of life. Few data are available on admission bundles for COPD.
This study found that admission and discharge care bundles for COPD had little impact on patient outcomes, including readmissions and mortality or healthcare costs, but appeared to be associated with a reduced number of subsequent attendances at the emergency department.
Chronic obstructive pulmonary disease (COPD) is one of the most common respiratory diseases in the UK, with an estimated prevalence of 1.2 million people.1 Globally, the estimated prevalence of COPD was 251 million cases in 2016, and it is estimated that 3.17 million deaths were caused by the disease in 2015, which is 5% of all deaths.2 COPD accounts for 10% of emergency hospital admissions in the UK, and the number of admissions has increased by 50% in the last decade.1 3 A third of these patients are readmitted within 28 days of discharge.3 COPD admissions are estimated to cost the National Health Service (NHS) 491 million per year.
The care provided for COPD varies across European countries, and an audit in the NHS highlighted wide variation in treatment provision and outcomes for patients admitted for COPD.4 5 This disparity was particularly marked in relation to mortality. It also showed that a significant proportion of the observed variability could be explained by access to expert care and evidence-based interventions. There is, therefore, opportunity to improve outcomes for patients with COPD by ensuring that care is consistently provided to a high standard.
This study evaluated the effectiveness of introducing admission and discharge care bundles for patients with an acute exacerbation of COPD (AECOPD) as a means of improving hospital care, and reducing readmissions and mortality, and explored the impact on cost of care and patient and staff experience.
We conducted a mixed-method evaluation using a controlled before-and-after design with nested qualitative case studies to examine the effect of implementing care bundles (online supplementary figure 1). The study was conducted between 2014 and 2017. More details about the study methodology can be found in the published protocol paper.15
A subset of these sites provided pseudoanonymised details of all individual patient-level admissions over a period of 12 months preindex and postindex date. With seven pairs of similar implementation and comparator sites, it was estimated that there would be a sample of approximately 10 000 admissions per year. Assuming 30% of patients were readmitted in comparator sites, this would allow us to detect a 9% absolute difference in the COPD readmission rate at 28 days with 90% power and 5% significance level.
Sites providing pseudoanonymised patient-level data (level 2) reported outcomes, sociodemographic, clinical and procedure codes plus the information gathered for level 1 sites. Demographic data on individual patients allowed us to study the characteristics of patients admitted with AECOPD and adjust for these in analyses. Each level 2 site was also asked to provide details from the case notes of a random sample of 140 postindex date COPD admissions to provide data on the delivery of the various components of the care bundles (n=2240). Quantitative data submitted to the research team were compiled and checked for both validity and consistency.
Qualitative data were collected from six level 3 sites about the process of bundle implementation or usual care. In addition, data on the context in which bundles were delivered, and the impact on staff, patients and carers, were collected by KM and AK. Methods used to capture data at these four implementation and two comparator sites included document analysis, non-participant observation of patient care, and indepth interviews with health professionals, patients and carers following both admission and discharge, guided by topic guides and observation schedules.
Trust-level aggregate outcomes were reported monthly for the 12 months before and after their index date. These monthly data were used to calculate the mean change of all outcome measures postindex date at each site. This mean change was then compared between implementation and comparator sites using linear regression analyses on monthly outcomes to estimate how the change postindex date differed between implementation and comparator sites after adjustment for the number of COPD admissions, overall 28-day readmission rate and in-hospital mortality rates in the preindex date period.
The characteristics of patients having at least one COPD admission in the preindex date period were compared between implementation and comparison sites. Linear, logistic and ordered logistic regression models with SEs adjusted for Trust-level clustering were used to test for differences between groups.
The extent to which sites implemented each element of the bundle was recorded using case note extraction data, and the results were summarised by group using frequencies and proportions. χ2 tests were used to compare the proportion of patients receiving bundle elements between implementation and comparator sites.
Interviews with staff, patients and carers and observational data from level 3 sites were collected, anonymised and uploaded into NVivo.16 All qualitative data were examined using cross-case thematic analysis and analysed both inductively and deductively.17 The analysis sought to identify similarities and differences between sites, highlighting aspects which might be transferable to other hospitals intending to implement care bundles. Attention was also given to overlaps or divergence between aspects of practice observed.
An economic evaluation considering the 90-day period following the index admission for COPD was undertaken in level 2 sites. We estimated per-patient secondary care NHS costs using healthcare resource group unit costing methodology,18 where patient-specific resource use was valued using nationally representative sources, for example, NHS reference costs,19 the British National Formulary,20 and Unit Costs of Health and Social Care.21 In the absence of individual-level data on self-reported quality of life or other outcome measures, 90-day mortality following the index admission served as the main outcome measure for the cost-effectiveness analysis. This permits the incremental costs associated with care bundles to be associated with incremental deaths avoided. Cost-effectiveness was calculated as a ratio of the difference in NHS secondary care costs between intervention and comparator sites to the between-site differences in 90-day mortality. Detailed methods are described in the protocol paper15 and study report.22
Throughout the 40-month study, the research team conducted a range of patient and public involvement (PPI) activities to ensure that the protocol was properly implemented and that any findings were appropriately interpreted in the light of patient and carer experience. Patients and carers were involved in suggesting the original idea for the study, then commenting on the application for funding, including aims and objectives, methods and PPI. An active PPI group was then formed which continued to provide input on recruitment and patient burden during data collection and study documentation such as consent forms and information sheets. PPI participants also provided feedback on the data available from both qualitative and quantitative sources, commenting on the extent to which it validated their own experiences of care, and provided feedback on possible approaches to the dissemination of the results of the study which would inform patient groups and the wider community.
Nineteen sites implementing COPD care bundles and 11 comparator sites provided preindex and postindex date data for analysis. One other comparator site was unable to provide data for the full study period. Pooled results are presented in table 1.
When comparing outcomes for readmissions, ED attendances, length of stay and bed days, the change observed between the preindex and postindex date periods in the implementation sites did not differ from that seen in the comparator sites (table 1). Graphs displaying the monthly 28-day readmission rates for COPD (preindex date) over time by site showed no obvious trends for most sides, although for some there was a clear, strong trend of an increasing admission rate over the year (online supplementary figure 2).
Staff perceptions of care bundles were largely positive. Bundles were described as useful for standardising working practices, supporting a clear patient care pathway, facilitating communication between different teams and identifying support required by patients following discharge from hospital. Care bundles were also perceived by staff as a means for embedding reliable, sustainable QI. Staff highlighted the need for managerial support, resourcing and regular training to facilitate QI.
with a care bundle, there is a better chance they are going to go out on the right treatment really, particularly if they have not been under the Respiratory Team, and they will have access to more services. (IMP11 ACU7, ED Consultant)
7fc3f7cf58