Development and validation of a Hospital Frailty Risk Score focusing on older people in acute care settings using electronic hospital records: an observational study.

BACKGROUND: Older people are increasing users of health care globally. We aimed to establish whether older people with characteristics of frailty and who are at risk of adverse health-care outcomes could be identified using routinely collected data. METHODS: A three-step approach was used to develop and validate a Hospital Frailty Risk Score from International Statistical Classification of Diseases and Related Health Problems, Tenth Revision (ICD-10) diagnostic codes. First, we carried out a cluster analysis to identify a group of older people (≥75 years) admitted to hospital who had high resource use and diagnoses associated with frailty. Second, we created a Hospital Frailty Risk Score based on ICD-10 codes that characterised this group. Third, in separate cohorts, we tested how well the score predicted adverse outcomes and whether it identified similar groups as other frailty tools. FINDINGS: In the development cohort (n=22 139), older people with frailty diagnoses formed a distinct group and had higher non-elective hospital use (33·6 bed-days over 2 years compared with 23·0 bed-days for the group with the next highest number of bed-days). In the national validation cohort (n=1 013 590), compared with the 429 762 (42·4%) patients with the lowest risk scores, the 202 718 (20·0%) patients with the highest Hospital Frailty Risk Scores had increased odds of 30-day mortality (odds ratio 1·71, 95% CI 1·68-1·75), long hospital stay (6·03, 5·92-6·10), and 30-day readmission (1·48, 1·46-1·50). The c statistics (ie, model discrimination) between individuals for these three outcomes were 0·60, 0·68, and 0·56, respectively. The Hospital Frailty Risk Score showed fair overlap with dichotomised Fried and Rockwood scales (kappa scores 0·22, 95% CI 0·15-0·30 and 0·30, 0·22-0·38, respectively) and moderate agreement with the Rockwood Frailty Index (Pearson's correlation coefficient 0·41, 95% CI 0·38-0·47). INTERPRETATION: The Hospital Frailty Risk Score provides hospitals and health systems with a low-cost, systematic way to screen for frailty and identify a group of patients who are at greater risk of adverse outcomes and for whom a frailty-attuned approach might be useful. FUNDING: National Institute for Health Research.

Frailty, hospital use and mortality in the older population: findings from the Newcastle 85+ study.

BACKGROUND: Frailty is a significant determinant of health care utilisation and associated costs, both of which also increase with proximity to death. What is not known is how the relationships between frailty, proximity to death, hospital use and costs develop in a population aged 85 years and over. METHODS: This study used data from a prospective observational cohort, the Newcastle 85+ Study, linked with hospital episode statistics and death registrations. Using the Rockwood frailty index (cut off <0.25), we analysed the relationship between frailty and mortality, proximity to death, hospital use and hospital costs over 2, 5 and 7 years using descriptive statistics, Kaplan-Meier survival curves, Cox's proportional hazards and negative binomial regression models. RESULTS: Baseline frailty was associated with a more than two-fold increased risk of mortality after 7 years, compared to people who were non-frail. Participants classified as frail spent more time in hospital over 7 years than the non-frail, but this difference declined over time. Baseline frailty was not associated with increased time spent in hospital during the last 90 days of life. CONCLUSION: Evidence continues to accrue on the impact of frailty on emergency health care use. Hospital and community services need to adapt to meet the challenge of introducing new proactive and preventative approaches, designed to achieve benefits in clinical and/or cost effectiveness of frailty management.

Safe working in a 7-day service. Experience of hip fracture care as documented by the UK National Hip Fracture Database.

Objective: to describe differences in care and 30-day mortality of patients admitted with hip fracture on weekends (Saturday-Sunday) compared to weekdays (Monday-Friday), and their relationship to the organisation of care. Methods: data came from the National Hip Fracture Database (NHFD) linked to ONS mortality data on 52,599 patients presenting to 162 units in England between 1 January and 31 December 2014. This was combined with information on geriatrician staffing and major trauma centre (MTC) status. 30-day mortality and care were compared for patients admitted at weekends and weekdays; separately for patients treated in units grouped by the mean level of input by geriatricians, weekend geriatrician clinical cover and MTC status. Differences were adjusted for variation in patients' characteristics. Results: there was no evidence of differences in 30-day mortality between patients admitted at weekends compared to weekdays (7.2 vs 7.5%, P = 0.3) before or after adjusting for patient characteristics in either MTCs or general hospitals. The proportion receiving a preoperative geriatrician assessment was lower at weekends (42.8 vs 60.7%, P < 0.001). 30-day mortality was lower in units with higher levels of geriatrician input, but there was no weekend mortality effect associated with lower levels of input or absence of weekend cover. Conclusion: there was no evidence of a weekend mortality effect among patients treated for hip fracture in the English NHS. It appears that clinical teams provide comparably safe and effective care throughout the week. However, greater geriatrician involvement in teams was associated with overall lower mortality.

Agreement between retrospectively and contemporaneously collected patient-reported outcome measures (PROMs) in hip and knee replacement patients.

PURPOSE: To investigate the relationship between retrospectively and contemporaneously collected patient-reported outcome measures (PROMs) and the influence on this relationship of patients' age and socio-economic status and the length of time. METHODS: Patients undergoing hip or knee replacement in four hospitals who had completed a pre-operative questionnaire were invited to recall their pre-operative health status shortly after surgery. The questionnaires included a disease-specific (Oxford Hip Score; Oxford Knee Score) and generic (EQ-5D-3L) PROM. Consistency and absolute agreement between contemporary and retrospective reports were investigated using intraclass correlations (ICCs). Differences were visualised using Bland-Altman plots. Linear regression analysis explored whether retrospective can predict contemporary PROMs. RESULTS: Patients' recalled health statuses were similar to their contemporaneous reports, with no significant systematic bias. Absolute agreement for disease-specific PROMs was very strong (ICC 0.82) and stronger than for the generic PROM (ICC 0.60, 0.62). Agreement was consistently strong across the range of severity of a patient's condition, age and socio-economic status. Patients' age and socio-economic status had no significant influence on size of difference and direction of recall, although reliability of recall was slightly worse among the over-75s versus under-60s for hips (Oxford Hip Score ICC 0.88 vs. 0.78). Mean retrospective PROMs for groups or populations of patients can reliably predict what mean contemporary reports of PROMs would have been. CONCLUSION: Retrospective PROMs can be used to obtain a baseline assessment of health status when contemporary collection is not feasible or cost effective. Research is needed to determine the feasibility of retrospective PROMs in emergency admissions.

Sociodemographic Characteristics, Cognitive Function, and Health-related Quality of Life of Patients Referred to Memory Assessment Services in England.

National policy in England is to encourage referral of people with suspected dementia to Memory Assessment Services (MAS). However, little is known about the characteristics of new referrals, which limits our capacity to evaluate these services. The objectives were to: describe the characteristics (age, sex, ethnicity, socioeconomic deprivation, and comorbidity) of referred patients, and examine the relationships between these characteristics and cognitive function (tertiles of Mini-Mental State Examination score) and health-related quality of life (HRQL) (DEMQOL, DEMQOL-Proxy). We used multivariable regression methods to analyze data from 1420 patients from 73 MAS, and their lay carers (n=1020). The mean age of patients was 78 years; 42% had cognitive function equivalent to Mini-Mental State Examination <24. Characteristics associated with lower function were: older age, being female, deprivation, and nonwhite ethnicity. Deprivation and nonwhite ethnicity were also associated with lower self-reported HRQL, as was having multiple comorbidities; older age was associated with better self-reported HRQL. Lower proxy-reported HRQL was associated with being female, deprivation and comorbidities, but not age and ethnicity. A large proportion of study participants had moderate or high cognitive function scores, suggesting that these patients were referred early to MAS. Research is needed to identify why apparent sociodemographic inequalities in use of MAS exist.

Change in Health-related Quality of Life After Referral to Memory Assessment Services.

Despite strong support for the provision of memory assessment services (MASs) in England and other countries, their effectiveness in improving patient outcomes is uncertain. We aimed to describe change in patients' health-related quality of life (HRQL) 6 months after referral to MASs and to examine associations with patient characteristics and use of postdiagnostic interventions. Data from 883 patients referred to 69 MASs and their informal caregivers (n=569) were collected at referral and 6 months later. Multivariable linear regression was used to examine associations of change in HRQL (DEMQOL, DEMQOL-Proxy) with patient characteristics (age, sex, ethnicity, socioeconomic deprivation, and comorbidity) and use of postdiagnostic interventions (antidementia medications and nonpharmacological therapies). Mean HRQL improved, irrespective of diagnosis: self-reported HRQL increased 3.4 points (95% CI, 2.7-4.1) and proxy-reported HRQL 1.3 points (95% CI, 0.5-2.1). HRQL change was not associated with any of the patient characteristics studied. Patients with dementia (54%) receiving antidementia drugs reported greater improvement in their HRQL but those using nonpharmacological therapies reported less improvement compared with those note receiving therapy. HRQL improved in the first 6 months after referral to MASs. Research is needed to determine longer term sustainability of the benefits and the cost-effectiveness of MASs.

Increased orthogeriatrician involvement in hip fracture care and its impact on mortality in England.

Objectives: to describe the increase in orthogeriatrician involvement in hip fracture care in England and its association with improvements in time to surgery and mortality. Study design: analysis of Hospital Episode Statistics for 196,401 patients presenting with hip fracture to 150 hospitals in England between 1 April 2010 and 28 February 2014, combined with data on orthogeriatrician hours from a national organisational survey. Methods: we examined changes in the average number of hours worked by orthogeriatricians in orthopaedic departments per patient with hip fracture, and their potential effect on mortality within 30 days of presentation. The role of prompt surgery (on day of or day after presentation) was explored as a potential confounding factor. Associations were assessed using conditional Poisson regression models with adjustment for patients' sex, age and comorbidity and year, with hospitals treated as fixed effects. Results: between 2010 and 2013, there was an increase of 2.5 hours per patient in the median number of hours worked by orthogeriatricians-from 1.5 to 4.0 hours. An increase of 2.5 hours per patient was associated with a relative reduction in mortality of 3.4% (95% confidence interval 0.9% to 5.9%, P = 0.01). This corresponds to an absolute reduction of approximately 0.3%. Higher numbers of orthogeriatrician hours were associated with higher rates of prompt surgery, but were independently associated with lower mortality. Conclusion: in the context of initiatives to improve hip fracture care, we identified statistically significant and robust associations between increased orthogeriatrician hours per patient and reduced 30-day mortality.

Diurnal and seasonal patterns in presentations with hip fracture-data from the national hip fracture database.

PURPOSE: we set out to examine diurnal and seasonal variation in hip fracture presentations to question their origin and to consider their implications for the organisation of health services for older people. METHODS: we used the National Hip Fracture Database to identify the time of presentation and surgery for 64,102 patients; all those older than 60 years who sustained this injury in England, Wales and Northern Ireland during 2014. RESULTS: we found marked diurnal variation in rates of presentation, increasing sharply after 0800 hours and decreasing only after 1800 hours. Among people who sustained their hip fracture in hospital (n = 2,761) or in a care home (n = 12,141), there were peaks in presentations around 0900 and 1800 hours. Time of presentation had a very marked effect on whether surgery was delayed by more than 24 hours but less against the national guidelines of surgery within 36 hours or by the next day. There were 15.6% more presentations during December compared to all other months (9.5% versus 8.2%, P < 0.001), a pattern also found among people living in care homes (9.1% versus 8.3%, P < 0.001). CONCLUSIONS: we have identified morning and evening peaks of presentation for inpatients and care home residents and a December increase in overall hip fracture numbers. These patterns warrant further investigation if those organising health services are to prevent this injury, and to provide appropriate beds and prompt operations for the people who sustain it.

The impact of a national clinician-led audit initiative on care and mortality after hip fracture in England: an external evaluation using time trends in non-audit data.

BACKGROUND: Hip fracture is the most common serious injury of older people. The UK National Hip Fracture Database (NHFD) was launched in 2007 as a national collaborative, clinician-led audit initiative to improve the quality of hip fracture care, but has not yet been externally evaluated. METHODS: We used routinely collected data on 471,590 older people (aged 60 years and older) admitted with a hip fracture to National Health Service (NHS) hospitals in England between 2003 and 2011. The main variables of interest were the use of early surgery (on day of admission, or day after) and mortality at 30 days from admission. We compared time trends in the periods 2003-2007 and 2007-2011 (before and after the launch of the NHFD), using Poisson regression models to adjust for demographic changes. FINDINGS: The number of hospitals participating in the NHFD increased from 11 in 2007 to 175 in 2011. From 2007 to 2011, the rate of early surgery increased from 54.5% to 71.3%, whereas the rate had remained stable over the period 2003-2007. Thirty-day mortality fell from 10.9% to 8.5%, compared with a small reduction from 11.5% to 10.9% previously. The annual relative reduction in adjusted 30-day mortality was 1.8% per year in the period 2003-2007, compared with 7.6% per year over 2007-2011 (P<0.001 for the difference). INTERPRETATION: The launch of a national clinician-led audit initiative was associated with substantial improvements in care and survival of older people with hip fracture in England.

Public reporting of surgeon outcomes: low numbers of procedures lead to false complacency.

The English National Health Service published outcome information for individual surgeons for ten specialties in June, 2013. We looked at whether individual surgeons do sufficient numbers of procedures to be able to reliably identify those with poor performance. For some specialties, the number of procedures that a surgeon does each year is low and, as a result, the chance of identifying a surgeon with increased mortality rates is also low. Therefore, public reporting of individual surgeons' outcomes could lead to false complacency. We recommend use of outcomes that are fairly frequent, considering the hospital as the unit of reporting when numbers are low, and avoiding interpretation of no evidence of poor performance as evidence of acceptable performance.

Estimating recruitment rates for routine use of patient reported outcome measures and the impact on provider comparisons.

BACKGROUND: The routine use of patient reported outcome measures (PROMs) aims to compare providers as regards the clinical need of their patients and their outcome. Simple methods of estimating recruitment rates based on aggregated data may be inaccurate. Our objectives were to: use patient-level linked data to evaluate these estimates; produce revised estimates of national and providers' recruitment rates; and explore whether or not recruitment bias exists. METHODS: Case study based on patients who were eligible to participate in the English National PROMs Programme for elective surgery (hip and knee replacement, groin hernia repair, varicose vein surgery) using data from pre-operative questionnaires and Hospital Episode Statistics. Data were linked to determine: the eligibility for including operations; eligibility of date of surgery; duplicate questionnaires; cancelled operations; correct assignment to provider. Influence of patient characteristics on recruitment rates were investigated. RESULTS: National recruitment rates based on aggregated data over-estimated the true rate because of the inclusion of ineligible operations (from 1.9% - 7.0% depending on operation) and operations being cancelled (1.9% - 3.6%). Estimates of national recruitment rates using inclusion criteria based on patient-level linked data were lower than those based on simple methods (eg hip replacement was 73% rather than 78%).Estimates of provider's recruitment rates based on aggregated data were also adversely affected by attributing patients to the wrong provider (2.4% - 4.9%). Use of linked data eliminated all estimates of over 100% recruitment, though providers still showed a wide range of rates.While the principal determinant of recruitment rates was the provider, some patients' socio-demographic characteristics had an influence on non-recruitment: non-white (Adjusted Odds Ratio 1.25-1.67, depending on operation); most deprived socio-economic group (OR 1.11-1.23); aged over 75 years (OR 1.28-1.79). However, there was no statistically significant association between providers' recruitment rates and patients' pre-operative clinical need. CONCLUSIONS: Accurate recruitment rates require the use of linked data to establish consistent inclusion criteria for numerators and denominators. Non-recruitment will bias comparisons of providers' pre-operative case-mix and may bias comparisons of outcomes if unmeasured confounders are not evenly distributed between providers. It is important, therefore, to strive for high recruitment rates.

Using patient-reported outcomes (PROs) to compare the providers of surgery: does the choice of measure matter?

BACKGROUND: Patient-reported outcomes (PROs) are being used to compare health care providers with little knowledge of how the choice of measure affects such comparisons. OBJECTIVES: To assess how much difference the choice of PRO makes to a provider's adjusted outcome and whether the choice affects a provider's rating. RESEARCH DESIGN: PROs collected in England from patients undergoing: hip replacement (243 providers; 52,692 patients); knee replacement (244; 60,118); varicose vein surgery (100; 11,163); and groin hernia repair (201; 31,714). Four case-mix-adjusted outcomes (mean postoperative disease-specific and generic PRO; proportion achieving a minimally important difference in disease-specific PRO; proportion reporting improvement on single transitional item). We calculated the associations between measures and for each measure, the proportion of providers rated as statistically above or below average and the level of agreement in ratings. RESULTS: For major surgery, disease-specific PROs were strongly correlated with the generic PRO (hip 0.90; knee 0.88), they rated high proportions of providers as above or below average (hip 25.1%; knee 19.3%) and there was agreement in ratings with the generic PRO. Even so, for a large proportion of providers (hip 30%; knee 16%) their rating depended on the choice of measure. For minor surgery, correlations between measures were mostly weak. The single transitional item identified the most outliers (varicose vein 20%, hernia 10%). CONCLUSIONS: Choice of outcome measure can determine a provider's rating. Measure selection depends on whether the priority is to avoid missing "poor" providers or avoid mislabeling average providers as "poor."

Factors associated with non-response in routine use of patient reported outcome measures after elective surgery in England.

BACKGROUND: Patient-reported outcome measures are increasingly being used to compare providers. We studied whether non-response rates to post-operative questionnaires are associated with patients' characteristics and organisational features of providers. METHODS: 131,447 patients who underwent a hip or knee replacement, hernia repair or varicose vein surgery in 2009-10 in England. Multivariable logistic regression to calculate adjusted odds ratios of non-response for characteristics of the patients and organisational characteristics of providers. Multiple imputation was used for missing patient characteristics. Providers were included as random effects. RESULTS: Response rates to the post-operative questionnaire were 85.1% for hip replacement (n = 37 961), 85.3% for knee replacements (n = 44 422), 72.9% for hernia repair (n = 34 964), and 64.8% for varicose vein surgery (n = 14 100). Across the four procedures, there were higher levels of non-response in men (odds ratios 1.03 [95% CI 0.95-1.11] - 1.35 [1.25-1.46]), younger patients (those under 55 years 3.01 [2.72-3.32] - 6.05 [5.49-6.67]), non-white patients (1.24 [1.11-1.38] - 2.08 [1.89-2.31]), patients in the most deprived quintile of socio-economic status (1.47 [1.34-1,62] - 1.86 [1.71-2.03]), those who lived alone (1.11 [0.99-1.23] - 1.27 [1.18-1.36]) and those who had been assisted when completing their pre-operative questionnaire (1.26 [1.10-1.46] -1.67 [1.56-1.79]). Non-response rates were also higher in patients who had poorer pre-operative health (three or more comorbidities: 1.14 [0.96-1.35] - 1.45 [1.30-1.63]). Providers' patient recruitment rates before surgery and the timing of pre-operative questionnaire administration did not affect the rates of response to post-operative questionnaires. CONCLUSION: If non-response can be shown to be associated with outcome, then rates of non-response to post-operative questionnaires would need to be taken into account when these measures are being used to compare the performance of providers or to evaluate surgical procedures.

Clinicians' and patients' views of metrics of change derived from patient reported outcome measures (PROMs) for comparing providers' performance of surgery.

BACKGROUND: Patient reported outcome measures (PROMs) are increasingly being used to compare the performance of health care providers. Our objectives were to determine the relative frequency of use of different metrics that can be derived from PROMs, explore clinicians' and patients' views of the options available, and make recommendations. METHODS: First a rapid review of the literature on metrics derived from two generic (EQ-5D and EQ-VAS) and three disease-specific (Oxford Hip Score; Oxford Knee Score; Aberdeen Varicose Vein Questionnaire) PROMs was conducted. Next, the findings of the literature review were mapped onto our typology of metrics to determine their relative frequency of use, Finally, seven group meetings with surgical clinicians (n = 107) and six focus groups with patients (n = 45) were held which were audio-taped, transcribed and analysed thematically. RESULTS: Only nine studies (9.3% of included papers) used metrics for comparing providers. These were derived from using either the follow-up PROM score (n = 3) or the change in score as an outcome (n = 5), both adjusted for pre-intervention score. There were no recorded uses of the proportion reaching a specified ('good') threshold and only two studies used the proportion reaching a minimally important difference (MID).Surgical clinicians wanted multiple outcomes, with most support expressed for the mean change in score, perceiving it to be more interpretable; there was also some support for the MID. For patients it was apparent that rather than the science behind these measures, the most important aspects were the use of language that would make the metrics personally meaningful and linking the metric to a familiar scale. CONCLUSIONS: For clinicians the recommended metrics are the mean change in score and the proportion achieving a MID, both adjusted for pre-intervention score. Both need to be clearly described and explained. For patients we recommend the proportion achieving a MID or proportion achieving a significant improvement in hip function, both adjusted for pre-intervention score.

Using patient-reported outcome measures for primary percutaneous coronary intervention.

Introduction: Routine measurement of the outcome of myocardial infarction is usually limited to immediate morbidity and mortality. Our aim was to determine the response to patient-reported outcome measures (PROMs) 3 months later, identify response bias and explore the feasibility of comparing outcome with their recalled view of their prior health state. Methods: Patients admitted with ST-segment-elevation myocardial infarction (STEMI) to five percutaneous coronary intervention centres were invited to complete a retrospective questionnaire containing the EQ-5D-3L and short form Seattle Angina Questionnaire (SAQ-7). Response rate for a 3-month mailed follow-up questionnaire and potential response biases were assessed. Patients' outcomes were compared with their baseline using χ2 and paired t-test to assess for differences. Results: Of 392 patients contacted, 260 (66.3%) responded. Responders were more likely to be older, female, more affluent and have a higher EQ-5D at baseline. Three months after surgery, patients' SAQ-7 and angina symptom subscale returned to their baseline score. The physical limitation subscale score was worse than at baseline (79.9 vs 73.2, p=0.002), whereas the quality-of-life subscale was better (66.6 vs 73.9; p<0.001). The EQ-5D-3L index score was similar at 3 months to baseline (0.82 vs 0.79). Evidence of bias arising from responders being in better general health at baseline needs further investigation and, if confirmed, needs to be taken into account in interpreting PROMs data. Conclusion: It is feasible to use PROMs routinely to assess the impact of emergency admissions of patients with STEMI. A larger demonstration project with more sites is needed to confirm these findings.

Feasibility of collecting and assessing patient-reported outcomes for emergency admissions: laparotomy for gastrointestinal conditions.

INTRODUCTION: Audit of emergency surgery is usually limited to immediate clinical outcomes relating to outcomes during the acute hospital episode with little attempt to capture patients' views of their longer-term outcomes. Our aim was to determine the response rate to patient-reported outcome measures (PROMs) for patients who underwent an emergency laparotomy for gastrointestinal conditions, identify response bias and explore the feasibility of comparing outcomes with their prior health based on their recalled view collected during their admission. METHODS: Patients undergoing emergency laparotomy in 11 hospitals were recruited to complete a retrospective questionnaire containing the EQ-5D-3L and Gastrointestinal Quality of Life Index (GIQLI). Response rate for 3-month mailed follow-up questionnaire and potential response biases were assessed. Patients' outcomes were compared with their baseline using χ2 and paired t-test to assess for differences. RESULTS: Of 255 patients contacted at 3 months, 190 (74.1%) responded. Responders were more likely to be older, female and more affluent. Patients' health improved significantly as regards the GIQLI (93.3 vs 97.9; p=0.048) and the subscale on symptoms (51.9 vs 59.6; p<0.001). No significant change in subscales on emotion or physical aspects or for overall health status (EQ-5D: 0.58 vs 0.64; p=0.06). According to the social subscale, patients had deteriorated (11.0 vs 9.8; p<0.0006). Differences in change scores by patient characteristics were slight, suggesting minimal response bias. CONCLUSION: This approach offers the opportunity for assessing the impact of treatment, from the patient's perspective and the potential to evaluate emergency laparotomy care using PROMs.

Comparison of control charts for monitoring clinical performance using binary data.

BACKGROUND: Time series charts are increasingly used by clinical teams to monitor their performance, but statistical control charts are not widely used, partly due to uncertainty about which chart to use. Although there is a large literature on methods, there are few systematic comparisons of charts for detecting changes in rates of binary clinical performance data. METHODS: We compared four control charts for binary data: the Shewhart p-chart; the exponentially weighted moving average (EWMA) chart; the cumulative sum (CUSUM) chart; and the g-chart. Charts were set up to have the same long-term false signal rate. Chart performance was then judged according to the expected number of patients treated until a change in rate was detected. RESULTS: For large absolute increases in rates (>10%), the Shewhart p-chart and EWMA both had good performance, although not quite as good as the CUSUM. For small absolute increases (<10%), the CUSUM detected changes more rapidly. The g-chart is designed to efficiently detect decreases in low event rates, but it again had less good performance than the CUSUM. IMPLICATIONS: The Shewhart p-chart is the simplest chart to implement and interpret, and performs well for detecting large changes, which may be useful for monitoring processes of care. The g-chart is a useful complement for determining the success of initiatives to reduce low-event rates (eg, adverse events). The CUSUM may be particularly useful for faster detection of problems with patient safety leading to increases in adverse event rates. .

How is feedback from national clinical audits used? Views from English National Health Service trust audit leads.

OBJECTIVE: To explore how the output of national clinical audits in England is used by professionals and whether and how their impact could be enhanced. METHODS: A mixed-methods study with the primary recipients of four national clinical audits of cancer care of 607 local audit leads, 274 (45%) completed a questionnaire and 32 participated in an interview. Our questions focused on how the audits were used and whether barriers existed to using the audits for local service improvement. We described variation in questionnaire responses between the audits using chi-squared tests. Results are reported as percentages with their 95% confidence intervals. Qualitative data were analysed using Framework analysis. RESULTS: More than 90% of survey respondents believed that the audit findings were relevant to their clinical work, and interviewees described how they used the audits for a range of purposes. Forty-two percent of survey respondents said they had changed their clinical practice, and 56% had implemented service improvements in response to the audits. The degree of change differed between the four audits, evident in both the questionnaire and the interview data. In the interviews, two recurring barriers emerged: (1) the importance of data quality, which, in turn, influenced the perceived relevance and validity of the audit data and therefore the ability to make changes based on it and (2) the need for clear presentation of benchmarked local performance data. The perceived authority and credibility of the professional bodies supporting the audits was a key factor underpinning the use of the audit findings. CONCLUSION: National cancer audit and feedback is used to improve services, but their impact could be enhanced by improving the data quality and relevance of feedback.

Avoidability of hospital deaths and association with hospital-wide mortality ratios: retrospective case record review and regression analysis.

OBJECTIVES: To determine the proportion of avoidable deaths (due to acts of omission and commission) in acute hospital trusts in England and to determine the association with the trust's hospital-wide standardised mortality ratio assessed using the two commonly used methods--the hospital standardised mortality ratio (HSMR) and the summary hospital level mortality indicator (SHMI). DESIGN: Retrospective case record review of deaths. SETTING: 34 English acute hospital trusts (10 in 2009 and 24 in 2012/13) randomly selected from across the spectrum of HSMR. MAIN OUTCOME MEASURES: Avoidable death, defined as those with at least a 50% probability of avoidability in view of trained medical reviewers. Association of avoidable death proportion with the HSMR and the SHMI assessed using regression coefficients, to estimate the increase in avoidable death proportion for a one standard deviation increase in standardised mortality ratio. PARTICIPANTS: 100 randomly selected hospital deaths from each trust. RESULTS: The proportion of avoidable deaths was 3.6% (95% confidence interval 3.0% to 4.3%). It was lower in 2012/13 (3.0%, 2.4% to 3.7%) than in 2009 (5.2%, 3.8% to 6.6%). This difference is subject to several factors, including reviewers' greater awareness in 2012/13 of orders not to resuscitate, patients being perceived as sicker on admission, minor differences in review form questions, and cultural changes that might have discouraged reviewers from criticising other clinicians. There was a small but statistically non-significant association between HSMR and the proportion of avoidable deaths (regression coefficient 0.3, 95% confidence interval -0.2 to 0.7). The regression coefficient was similar for both time periods (0.1 and 0.3). This implies that a difference in HSMR of between 105 and 115 would be associated with an increase of only 0.3% (95% confidence interval -0.2% to 0.7%) in the proportion of avoidable deaths. A similar weak non-significant association was observed for SHMI (regression coefficient 0.3, 95% confidence interval -0.3 to 1.0). CONCLUSIONS: The small proportion of deaths judged to be avoidable means that any metric based on mortality is unlikely to reflect the quality of a hospital. The lack of association between the proportion of avoidable deaths and hospital-wide SMRs partly reflects methodological shortcomings in both metrics. Instead, reviews of individual deaths should focus on identifying ways of improving the quality of care, whereas the use of standardised mortality ratios should be restricted to assessing the quality of care for conditions with high case fatality for which good quality clinical data exist.

Impact on hospital performance of introducing routine patient reported outcome measures in surgery.

OBJECTIVES: To determine the impact of introducing patient reported outcome measures (PROMs) on the selection of patients for surgery and on outcome for four elective operations (hip replacement, knee replacement, varicose vein surgery and groin hernia repair). METHODS: Patient-level data from the National PROMs programme for England from April 2009 to March 2012 were analysed to determine the extent of change in mean preoperative scores and mean adjusted postoperative scores using disease-specific and generic PROMs assessed using standardized effect sizes (SESs). Variation between providers was determined by intra-class correlation coefficients for each time period. Consistency in outlier ratings was tested using weighted Kappa statistics. RESULTS: There was little apparent impact. Preoperative severity increased slightly for two procedures only: varicose vein surgery (SES disease-specific PROM 0.10; generic PROM -0.07) and to a lesser extent for hip replacement (SES disease-specific PROM -0.03). There was little inter-provider variation and it did not change significantly over time. There were also slight improvements in outcomes for hip and knee replacement (SES for disease-specific and generic PROMs 0.03) though not for hernia repair and a slight worsening for varicose vein surgery. The extent of variation in performance between providers was unchanged. The proportion of providers deemed to be outliers did not change over time. There was only moderate consistency in those providers deemed to be outliers for hip and knee replacement (Kappa 0.31-0.47) and it was even weaker for the other two procedures. Although 35% of providers of hip replacement were outliers in at least one year, only 6% were consistently outliers. Such inconsistency may be partly due to regression to the mean. CONCLUSIONS: The minimal impact that the routine use and feedback of PROMs had on provider behaviour during the initial years suggests that more attention needs to be paid to how results are communicated and to the provision of advice as to what action may be taken.