How quality improvement collaboratives work to improve healthcare in care homes: a realist evaluation.

BACKGROUND: Quality improvement collaboratives (QICs) bring together multidisciplinary teams in a structured process to improve care quality. How QICs can be used to support healthcare improvement in care homes is not fully understood. METHODS: A realist evaluation to develop and test a programme theory of how QICs work to improve healthcare in care homes. A multiple case study design considered implementation across 4 sites and 29 care homes. Observations, interviews and focus groups captured contexts and mechanisms operating within QICs. Data analysis classified emerging themes using context-mechanism-outcome configurations to explain how NHS and care home staff work together to design and implement improvement. RESULTS: QICs will be able to implement and iterate improvements in care homes where they have a broad and easily understandable remit; recruit staff with established partnership working between the NHS and care homes; use strategies to build relationships and minimise hierarchy; protect and pay for staff time; enable staff to implement improvements aligned with existing work; help members develop plans in manageable chunks through QI coaching; encourage QIC members to recruit multidisciplinary support through existing networks; facilitate meetings in care homes and use shared learning events to build multidisciplinary interventions stepwise. Teams did not use measurement for change, citing difficulties integrating this into pre-existing and QI-related workload. CONCLUSIONS: These findings outline what needs to be in place for health and social care staff to work together to effect change. Further research needs to consider ways to work alongside staff to incorporate measurement for change into QI.

A comparison of two national frailty scoring systems.

BACKGROUND: The electronic Frailty Index (eFI) has been developed in primary care settings. The Hospital Frailty Risk Score (HFRS) was derived using secondary care data. OBJECTIVE: Compare the two different tools for identifying frailty in older people admitted to hospital. DESIGN AND SETTING: Retrospective cohort study using the Secure Anonymised Information Linkage Databank, comprising 126,600 people aged 65+ who were admitted as an emergency to hospital in Wales from January 2013 up until December 2017. METHODS: Pearson's correlation coefficient and weighted kappa were used to assess the correlation between the tools. Cox and logistic regression were used to estimate hazard ratios (HRs) and odds ratios (ORs). The Concordance statistic and area under the receiver operating curves (AUROC) were estimated to determine discrimination. RESULTS: Pearson's correlation coefficient was 0.26 and the weighted kappa was 0.23. Comparing the highest to the least frail categories in the two scores the HRs for 90-day mortality, 90-day emergency readmission and care home admissions within 1-year using the HFRS were 1.41, 1.69 and 4.15 for the eFI 1.16, 1.63 and 1.47. Similarly, the ORs for inpatient death, length of stay greater than 10 days and readmission within 30-days were 1.44, 2.07 and 1.52 for the HFRS, and 1.21, 1.21 and 1.44 for the eFI. AUROC was determined as having no clinically relevant difference between the tools. CONCLUSIONS: The eFI and HFRS have a low correlation between their scores. The HRs and ORs were higher for the increasing frailty categories for both the HFRS and eFI.

Short and Long term predictions of Hospital emergency department attendances.

OBJECTIVE: Emergency departments in the United Kingdom (UK) experience significant difficulties in achieving the 95% NHS access standard due to unforeseen variations in patient flow. In order to maximize efficiency and minimize clinical risk, better forecasting of patient demand is necessary. The objective is therefore to create a tool that accurately predicts attendance at emergency departments to support optimal planning of human and physical resources. METHODS: Historical attendance data between Jan-2011 - December-2015 from four hospitals were used as a training set to develop and validate a forecasting model. To handle weekday variations, the data was first segmented into each weekday time series and a separate model for each weekday was performed. Seasonality testing was performed, followed by Box-Cox transformations. A modified heuristics based on a fuzzy time series model was then developed and compared with autoregressive integrated moving average and neural networks models using Harvey, Leybourne and Newbold (HLN) test. The time series models were tested in four emergency department sites to assess forecasting accuracy using the root mean square error and mean absolute percentage error. The models were tested for (i) short term prediction (four weeks ahead), using weekday time series; and (ii) long term predictions (four months ahead) using monthly time series. RESULTS: Data analysis revealed that presentations to emergency department and subsequent admissions to hospital were not a purely random process and therefore could be predicted with acceptable accuracy. Prediction accuracy improved as the forecast time intervals became wider (from daily to monthly). For each weekday time series modelling using fuzzy time series, for forecasting daily admissions, the mean absolute percentage error ranged from 2.63% to 4.72% while for monthly time series mean absolute percentage error varied from 2.01%-2.81%. For weekday time series, the mean absolute percentage error for autoregressive integrated moving average and neural network forecasting models ranged from 6.25% to 7.47% and 6.04%-7.42% respectively. The proposed fuzzy time series model proved to have statistically significant performance using Harvey, Leybourne and Newbold (HLN) test. This was explained by variations in attendances in different sites and weekdays. CONCLUSIONS: This paper described a heuristic-based fuzzy logic model for predicting emergency department attendances which could help resource allocation and reduce pressure on busy hospitals. Valid and reproducible prediction tools could be generated from these hospital data. The methodology had an acceptable accuracy over a relatively short time period, and could be used to assist better bed management, staffing and elective surgery scheduling. When compared to other prediction models usually applied for emergency department attendances prediction, the proposed heuristic model had better accuracy.

Measuring health-related quality of life of care home residents: comparison of self-report with staff proxy responses.

INTRODUCTION: care home residents are often unable to complete health-related quality of life questionnaires for themselves because of prevalent cognitive impairment. This study compared care home resident and staff proxy responses for two measures, the EQ-5D-5L and HowRU. METHODS: a prospective cohort study recruited residents ≥60 years across 24 care homes who were not receiving short stay, respite or terminal care. Resident and staff proxy EQ-5D-5L and HowRu responses were collected monthly for 3 months. Weighted kappa statistics and intra-class correlation coefficients (ICCs) adjusted for clustering at the care home level were used to measure agreement between resident and proxies for each time point. The effect of staff and resident baseline variables on agreement was considered using a multilevel mixed effect regression model. RESULTS: 117, 109 and 104 matched pairs completed the questionnaires at 1, 2 and 3 months, respectively. When clustering was controlled for, agreement between resident and staff proxy EQ-5D-5L responses was fair for mobility (ICC: 0.29) and slight for all other domains (ICC ≤ 0.20). EQ-5D Index and Quality-Adjusted Life Year scores (proxy scores higher than residents) showed better agreement than EQ-5D-VAS (residents scores higher than proxy). HowRU showed only slight agreement (ICC ≤ 0.20) between residents and proxies. Staff and resident characteristics did not influence level of agreement for either index. DISCUSSION: the levels of agreement for EQ-5D-5L and HowRU raise questions about their validity in this population.

Assessment of discharge treatment prescribed to women admitted to hospital for hyperemesis gravidarum.

AIMS: Prescribing drug treatment for the management of hyperemesis gravidarum (HG), the most severe form of nausea and vomiting in pregnancy, remains controversial. Since most manufacturers do not recommend prescribing antiemetics during pregnancy, little is known regarding which treatments are most prevalent among pregnant patients. Here, we report for the first time, evidence of actual treatments prescribed in English hospitals. METHODS: A retrospective pregnancy cohort was constructed using anonymised electronic records in the Nottingham University Hospitals Trust system for all women who delivered between January 2010 and February 2015. For women admitted to hospital for HG, medications prescribed on discharge were described and variation by maternal characteristics was assessed. Compliance with local and national HG treatment guidelines was evaluated. RESULTS: Of 33 567 pregnancies (among 30 439 women), the prevalence of HG was 1.7%. Among 530 HG admissions with records of discharge drugs, cyclizine was the most frequently prescribed (almost 73% of admissions). Prochlorperazine and metoclopramide were prescribed mainly in combination with other drugs; however, ondansetron was more common than metoclopramide at discharge from first and subsequent admissions. Steroids were only prescribed following readmissions. Thiamine was most frequently prescribed following readmission while high dose of folic acid was prescribed equally after first or subsequent admissions. Prescribing showed little variation by maternal age, ethnicity, weight, socioeconomic deprivation, or comorbidities. CONCLUSION: Evidence that management of HG in terms of discharge medications mainly followed local and national recommendations provides reassurance within the health professional community. Wider documentation of drugs prescribed to women with HG is required to enable full assessment of whether optimal drug management is being achieved.

Quality improvement collaborative aiming for Proactive HEAlthcare of Older People in Care Homes (PEACH): a realist evaluation protocol.

INTRODUCTION: This protocol describes a study of a quality improvement collaborative (QIC) to support implementation and delivery of comprehensive geriatric assessment (CGA) in UK care homes. The QIC will be formed of health and social care professionals working in and with care homes and will be supported by clinical, quality improvement and research specialists. QIC participants will receive quality improvement training using the Model for Improvement. An appreciative approach to working with care homes will be encouraged through facilitated shared learning events, quality improvement coaching and assistance with project evaluation. METHODS AND ANALYSIS: The QIC will be delivered across a range of partnering organisations which plan, deliver and evaluate health services for care home residents in four local areas of one geographical region. A realist evaluation framework will be used to develop a programme theory informing how QICs are thought to work, for whom and in what ways when used to implement and deliver CGA in care homes. Data collection will involve participant observations of the QIC over 18 months, and interviews/focus groups with QIC participants to iteratively define, refine, test or refute the programme theory. Two researchers will analyse field notes, and interview/focus group transcripts, coding data using inductive and deductive analysis. The key findings and linked programme theory will be summarised as context-mechanism-outcome configurations describing what needs to be in place to use QICs to implement service improvements in care homes. ETHICS AND DISSEMINATION: The study protocol was reviewed by the National Health Service Health Research Authority (London Bromley research ethics committee reference: 205840) and the University of Nottingham (reference: LT07092016) ethics committees. Both determined that the Proactive HEAlthcare of Older People in Care Homes study was a service and quality improvement initiative. Findings will be shared nationally and internationally through conference presentations, publication in peer-reviewed journals, a graphical illustration and a dissemination video.

Investigating the discriminative value of Early Warning Scores in patients with respiratory disease using a retrospective cohort analysis of admissions to Nottingham University Hospitals Trust over a 2-year period.

OBJECTIVE: Early Warning Scores (EWSs) are used to monitor patients for signs of imminent deterioration. Although used in respiratory disease, EWSs have not been well studied in this population, despite the underlying cardiopulmonary pathophysiology often present. We examined the performance of two scoring systems in patients with respiratory disease. DESIGN: Retrospective cohort analysis of vital signs observations of all patients admitted to a respiratory unit over a 2-year period. Scores were linked to outcome data to establish the performance of the National EWS (NEWS) compared results to a locally adapted EWS. SETTING: Nottingham University Hospitals National Health Service Trust respiratory wards. Data were collected from an integrated electronic observation and task allocation system employing a local EWS, also generating mandatory referrals to clinical staff at set scoring thresholds. OUTCOME MEASURES: Projected workload, and sensitivity and specificity of the scores in predicting mortality based on outcome within 24 hours of a score being recorded. RESULTS: 8812 individual patient episodes occurred during the study period. Overall, mortality was 5.9%. Applying NEWS retrospectively (vs local EWS) generated an eightfold increase in mandatory escalations, but had higher sensitivity in predicting mortality at the protocol cut points. CONCLUSIONS: This study highlights issues surrounding use of scoring systems in patients with respiratory disease. NEWS demonstrated higher sensitivity for predicting death within 24 hours, offset by reduced specificity. The consequent workload generated may compromise the ability of the clinical team to respond to patients needing immediate input. The locally adapted EWS has higher specificity but lower sensitivity. Statistical evaluation suggests this may lead to missed opportunities for intervention, however, this does not account for clinical concern independent of the scores, nor ability to respond to alerts based on workload. Further research into the role of warning scores and the impact of chronic pathophysiology is urgently needed.

Measuring health-related quality of life of care home residents, comparison of self-report with staff proxy responses for EQ-5D-5L and HowRu: protocol for assessing proxy reliability in care home outcome testing.

INTRODUCTION: Research into interventions to improve health and well-being for older people living in care homes is increasingly common. Health-related quality of life (HRQoL) is frequently used as an outcome measure, but collecting both self-reported and proxy HRQoL measures is challenging in this setting. This study will investigate the reliability of UK care home staff as proxy respondents for the EQ-5D-5L and HowRu measures. METHODS AND ANALYSIS: This is a prospective cohort study of a subpopulation of care home residents recruited to the larger Proactive Healthcare for Older People in Care Homes (PEACH) study. It will recruit residents ≥60 years across 24 care homes and not receiving short stay or respite care. The sample size is 160 participants. Resident and care home staff proxy EQ-5D-5L and HowRu responses will be collected monthly for 3 months. Weighted kappa statistics and intraclass correlation adjusted for clustering at the care home level will be used to measure agreement between resident and proxy responses. The extent to which staff variables (gender, age group, length of time caring, role, how well they know the resident, length of time working in care homes and in specialist gerontological practice) influence the level of agreement between self-reported and proxy responses will be considered using a multilevel mixed-effect regression model. ETHICS AND DISSEMINATION: The PEACH study protocol was reviewed by the UK Health Research Authority and University of Nottingham Research Ethics Committee and was determined to be a service development project. We will publish this study in a peer-reviewed journal with international readership and disseminate it through relevant national stakeholder networks and specialist societies.

Accurate identification of hospital admissions from care homes; development and validation of an automated algorithm.

Background: measuring the complex needs of care home residents is crucial for resource allocation. Hospital patient administration systems (PAS) may not accurately identify admissions from care homes. Objective: to develop and validate an accurate, practical method of identifying care home resident hospital admission using routinely collected PAS data. Method: admissions data between 2011 and 2012 (n = 103,105) to an acute Trust were modelled to develop an automated tool which compared the hospital PAS address details with the Care Quality Commission's (CQC) database, producing a likelihood of care home residency. This tool and the Nuffield method (CQC postcode match only) were validated against a manual check of a random sample of admissions (n = 2,000). A dataset from a separate Trust was analysed to assess generalisability. Results: the hospital PAS was inaccurate; none of the admissions from a care home identified on manual check had a care home source of admission recorded on the PAS. Both methods performed well; the automated tool had a higher positive predictive value than the Nuffield method (100% 95% confidence interval (CI) 98.23-100% versus 87.10% 95%CI 82.28-91.00%), meaning those coded as care home residents were more likely to actually be from a care home. Our automated tool had a high level of agreement 99.2% with the second Trust's data (Kappa 0.86 P < 0.001). Conclusions: care home status is not routinely or accurately captured. Automated matching offers an accurate, repeatable, scalable method to identify care home residency and could be used as a tool to benchmark how care home residents use acute hospital resources across the National Health Service.

Predicting hospital stay, mortality and readmission in people admitted for hypoglycaemia: prognostic models derivation and validation.

AIMS/HYPOTHESIS: Hospital admissions for hypoglycaemia represent a significant burden on individuals with diabetes and have a substantial economic impact on healthcare systems. To date, no prognostic models have been developed to predict outcomes following admission for hypoglycaemia. We aimed to develop and validate prediction models to estimate risk of inpatient death, 24 h discharge and one month readmission in people admitted to hospital for hypoglycaemia. METHODS: We used the Hospital Episode Statistics database, which includes data on all hospital admission to National Health Service hospital trusts in England, to extract admissions for hypoglycaemia between 2010 and 2014. We developed, internally and temporally validated, and compared two prognostic risk models for each outcome. The first model included age, sex, ethnicity, region, social deprivation and Charlson score ('base' model). In the second model, we added to the 'base' model the 20 most common medical conditions and applied a stepwise backward selection of variables ('disease' model). We used C-index and calibration plots to assess model performance and developed a calculator to estimate probabilities of outcomes according to individual characteristics. RESULTS: In derivation samples, 296 out of 11,136 admissions resulted in inpatient death, 1789/33,825 in one month readmission and 8396/33,803 in 24 h discharge. Corresponding values for validation samples were: 296/10,976, 1207/22,112 and 5363/22,107. The two models had similar discrimination. In derivation samples, C-indices for the base and disease models, respectively, were: 0.77 (95% CI 0.75, 0.80) and 0.78 (0.75, 0.80) for death, 0.57 (0.56, 0.59) and 0.57 (0.56, 0.58) for one month readmission, and 0.68 (0.67, 0.69) and 0.69 (0.68, 0.69) for 24 h discharge. Corresponding values in validation samples were: 0.74 (0.71, 0.76) and 0.74 (0.72, 0.77), 0.55 (0.54, 0.57) and 0.55 (0.53, 0.56), and 0.66 (0.65, 0.67) and 0.67 (0.66, 0.68). In both derivation and validation samples, calibration plots showed good agreement for the three outcomes. We developed a calculator of probabilities for inpatient death and 24 h discharge given the low performance of one month readmission models. CONCLUSIONS/INTERPRETATION: This simple and pragmatic tool to predict in-hospital death and 24 h discharge has the potential to reduce mortality and improve discharge in people admitted for hypoglycaemia.

Risk factors and outcome differences in hypoglycaemia-related hospital admissions: A case-control study in England.

AIMS: To evaluate risk factors for hospital admissions for hypoglycaemia and compare length of hospitalization, inpatient mortality and hospital readmission between hypoglycaemia- and non-hypoglycaemia-related admissions. MATERIALS AND METHODS: We used all admissions for hypoglycaemia in individuals with diabetes to English NHS hospital trusts between 2005 and 2014 (101 475 case admissions) and 3 random admissions per case in individuals with diabetes without hypoglycaemia (304 425 control admissions). Risk factors and differences in the 3 outcomes were estimated with logistic and negative binomial regressions. RESULTS: A U-shaped relationship between age and risk of admission for hypoglycaemia was observed until the age of 85 years; compared to the nadir at 60 years, the risk was progressively higher in younger and older patients and steadily declined after 85 years. Social deprivation (positively) and comorbidities (negatively) were associated with the risk of admission for hypoglycaemia. Compared to Caucasians, other ethnic groups had lower (Bangladeshi, Pakistani, Indians) or higher (Caribbean) risk of admission for hypoglycaemia. Length of hospitalization was 26% shorter while risk of rehospitalization was 65% higher in individuals admitted for hypoglycaemia. Compared to admissions for hypoglycaemia, risk of inpatient mortality was 50% lower for unstable angina but higher for acute myocardial infarction (3 times), acute renal failure (5 times) or pneumonia (8 times). CONCLUSIONS: Among hospital-admitted individuals with diabetes, age, social deprivation, comorbidities and ethnicity are associated with higher frequency of hospitalization for hypoglycaemia. Admission for hypoglycaemia is associated with a greater risk of readmission, a shorter length of hospitalisation and a generally lower inpatient mortality compared to admissions for other medical conditions. These results could help in identifying at-risk groups to reduce the burden of hospitalization for hypoglycaemia.

Trends in hospital admissions for hypoglycaemia in England: a retrospective, observational study.

BACKGROUND: Studies in the USA and Canada have reported increasing or stable rates of hospital admissions for hypoglycaemia. Some data from small studies are available for other countries. We aimed to gather information about long-term trends in hospital admission for hypoglycaemia and subsequent outcomes in England to help widen understanding for the global burden of hospitalisation for hypoglycaemia. METHODS: We collected data for all hospital admissions listing hypoglycaemia as primary reason of admission between Jan 1, 2005, and Dec 31, 2014, using the Hospital Episode Statistics database, which contains details of all admissions to English National Health Service (NHS) hospital trusts. We calculated trends in crude and adjusted (for age, sex, ethnic group, social deprivation, and Charlson comorbidity score) admissions for hypoglycaemia; in admissions for hypoglycaemia per total hospital admissions and per diabetes prevalence in England; and in length of stay, in-hospital mortality, and 1 month readmissions for hypoglycaemia. FINDINGS: 79 172 people had 101 475 admissions for hypoglycaemia between 2005 and 2014, of which 72 568 (72%) occurred in people aged 60 years or older. 13 924 (18%) people had more than one admission for hypoglycaemia during the study period. The number of admissions increased steadily from 7868 in 2005, to 11 756 in 2010 (49% increase) and then remained more stable until 2014 (10 977; 39% increase from baseline, range across English regions 11-89%); the trend was similar after adjustment for risk factors, with a rate ratio of 1·53 (95% CI 1·29-1·81) for 2014 versus 2005. Admissions for hypoglycaemia per 100 000 total hospital admissions increased from 63·6 to 78·9 between 2005-06 and 2010-11 (24% increase), and then fell to 72·3 per 100 000 in 2013-14 (14% overall increase). Accounting for diabetes prevalence data, rates declined from 4·64 to 3·86 admissions per 1000 person-years with diabetes between 2010-11 and 2013-14. We were unable to compare prevalence rates with data prior to 2010, as the populations were not comparable; data were available for all individuals prior to 2010 but only for those aged 17 years or older after 2010. With some differences across regions, from 2005 to 2014, the adjusted proportion of admissions to receive same-day discharge increased by 43·8% (from 18·9 to 27·1 same-day discharges per 100 admissions); in-hospital mortality decreased by 46·3% (from 4·2 to 2·3 deaths per 100 admissions); and 1 month readmissions decreased by 63·0% (from 48·1 to 17·8 per 100 readmissions). INTERPRETATION: Over 10 years, hospital admissions in England for hypoglycaemia increased by 39% in absolute terms and by 14% considering the general increase in hospitalisation; however, accounting for diabetes prevalence, there was a reduction of admission rates. Hospital length of stay, mortality, and 1 month readmissions decreased progressively and consistently during the study period. Given the continuous rise of diabetes prevalence, ageing population, and costs associated with hypoglycaemia, individual and national initiatives should be implemented to reduce the burden of hospital admissions for hypoglycaemia. FUNDING: None.

NightShift simulation to train newly qualified doctors in non-technical skills: a feasibility study.

There is growing evidence of greater rates of morbidity and mortality in hospitals during out-of-hours shifts, which appears to be exacerbated during the period in which newly qualified doctors commence work. In order to combat this issue, an online simulation of a night shift was developed and trialled in order to improve the non-technical skills of newly qualified doctors and, ultimately, improve clinical outcomes. A randomised feasibility trial of the electronic training simulation was performed with medical students (n=30) at the end of their training and in the initial weeks of working at a large teaching hospital. The study showed that participants in the intervention group completed their non-urgent tasks more rapidly than the control group: mean (SD) time to complete a non-urgent task of 85.1 (50.1) versus 157.6 (90.4) minutes, p=0.027. This difference persisted using linear regression analysis, which was undertaken using rota and task volume as independent cofactors (p=0.028). This study shows the potential for simulation technologies to improve non-technical skills.

Using venous blood gas analysis in the assessment of COPD exacerbations: a prospective cohort study.

INTRODUCTION: Identifying acute hypercapnic respiratory failure is crucial in the initial management of acute exacerbations of COPD. Guidelines recommend obtaining arterial blood samples but these are more difficult to obtain than venous. We assessed whether blood gas values derived from venous blood could replace arterial at initial assessment. METHODS: Patients requiring hospital treatment for an exacerbation of COPD had paired arterial and venous samples taken. Bland-Altman analyses were performed to assess agreement between arterial and venous pH, CO2 and HCO3-. The relationship between SpO2 and SaO2 was assessed. The number of attempts and pain scores for each sample were measured. RESULTS: 234 patients were studied. There was good agreement between arterial and venous measures of pH and HC)3- (mean difference 0.03 and -0.04, limits of agreement -0.05 to 0.11 and -2.90 to 2.82, respectively), and between SaO2 and SpO2 (in patients with an SpO2 of >80%). Arterial sampling required more attempts and was more painful than venous (mean pain score 4 (IQR 2-5) and 1 (IQR 0-2), respectively, p<0.001). CONCLUSIONS: Arterial sampling is more difficult and more painful than venous sampling. There is good agreement between pH and HCO3- values derived from venous and arterial blood, and between pulse oximetry and arterial blood gas oxygen saturations. These agreements could allow the initial assessment of COPD exacerbations to be based on venous blood gas analysis and pulse oximetry, simplifying the care pathway and improving the patient experience.