Elizabeth Teale

Development and validation of an electronic frailty index using routine primary care electronic health record data

Background: frailty is an especially problematic expression of population ageing. International guidelines recommend routine identification of frailty to provide evidence-based treatment, but currently available tools require additional resource. Objectives: to develop and validate an electronic frailty index (eFI) using routinely available primary care electronic health record data. Study design and setting: retrospective cohort study. Development and internal validation cohorts were established using a randomly split sample of the ResearchOne primary care database. External validation cohort established using THIN database. Participants: patients aged 65–95, registered with a ResearchOne or THIN practice on 14 October 2008. Predictors: we constructed the eFI using the cumulative deficit frailty model as our theoretical framework. The eFI score is calculated by the presence or absence of individual deficits as a proportion of the total possible. Categories of fit, mild, moderate and severe frailty were defined using population quartiles. Outcomes: outcomes were 1-, 3- and 5-year mortality, hospitalisation and nursing home admission. Statistical analysis: hazard ratios (HRs) were estimated using bivariate and multivariate Cox regression analyses. Discrimination was assessed using receiver operating characteristic (ROC) curves. Calibration was assessed using pseudo-R2 estimates. Results: we include data from a total of 931,541 patients. The eFI incorporates 36 deficits constructed using 2,171 CTV3 codes. One-year adjusted HR for mortality was 1.92 (95% CI 1.81–2.04) for mild frailty, 3.10 (95% CI 2.91–3.31) for moderate frailty and 4.52 (95% CI 4.16–4.91) for severe frailty. Corresponding estimates for hospitalisation were 1.93 (95% CI 1.86–2.01), 3.04 (95% CI 2.90–3.19) and 4.73 (95% CI 4.43–5.06) and for nursing home admission were 1.89 (95% CI 1.63–2.15), 3.19 (95% CI 2.73–3.73) and 4.76 (95% CI 3.92–5.77), with good to moderate discrimination but low calibration estimates. Conclusions: the eFI uses routine data to identify older people with mild, moderate and severe frailty, with robust predictive validity for outcomes of mortality, hospitalisation and nursing home admission. Routine implementation of the eFI could enable delivery of evidence-based interventions to improve outcomes for this vulnerable group.

A systematic review of case-mix adjustment models for stroke

Objective: To identify any externally validated prognostic model for predicting outcome in unselected populations following acute stroke comprising variables feasible for collection in routine care. Data sources: Searches were run in MEDLINE, EMBASE, CINAHL, PsycInfo, AMED and ISI Web of Science with no limits on publication date or language. Review methods: Any study describing the development or external validation of a discernible prognostic model to predict any valid outcome following acute stroke was included. Papers were retained if they met pre-specified inclusion criteria identified from previous reviews and pertinent discussion papers. Data extraction focused on methodological quality of model development, generalizability and feasibility of variable collection. Model performance was examined through consideration of external validation studies. Results: Seventeen externally validated models were identified from 43 papers fulfilling inclusion criteria. Quality of studies describing model development was variable and model performance in external validation studies was generally poor. Models were generally constructed through secondary use of randomized trial or stroke database data. Prognostic variables broadly encompassed markers of stroke severity, pre-stroke function and comorbidities. One model that fulfilled the review criteria and had extensive external validation in a range of post-stroke populations was identified (the Six Simple Variables model). Conclusion: The Six Simple Variables model performed well in six external validation studies, although prediction of outcome in patients with milder strokes was less reliable. Other models identified in this review have been developed using robust methodology but comprise more complex clinical variables which may limit their utility in routine stroke care.

Subdural haematoma in the elderly

An 84 year old woman with a history of postural hypotension and frequent falls presented with a two week history of confusion and wandering. Her general practitioner diagnosed delirium secondary to a urinary tract infection on the basis of a positive urinary dipstick test for white blood cells and nitrites. She failed to improve with oral antibiotics and was admitted to hospital for further assessment. Clinical examination showed a lack of focal neurological signs but cranial computed tomography revealed a left sided chronic subdural haematoma. A subdural haematoma results from tearing of the veins linking the cerebral cortex and the dural sinuses, causing blood to accumulate between the dura and arachnoid maters. Brain atrophy associated with ageing or dementia stretches these fragile veins such that they are more prone to tearing after relatively trivial (and often forgotten) head injury1; in up to half of cases there is a history of fall without head trauma.1 2 3 Over time, there is fibrin deposition and organisation of the haematoma, resulting in a chronic subdural haematoma. Neovascularisation processes within the subdural haematoma increase the propensity for acute or chronic bleeding.4 Anticoagulation increases the risk of subdural haematoma and the likelihood of fresh bleeding into an established haematoma.4 #### How common is subdural haematoma?

Multicomponent delirium prevention: not as effective as NICE suggest?

Multicomponent delirium prevention strategies have been shown in intervention studies consistently to reduce the occurrence of delirium. Based on this convincing evidence base, the National Institute for Health and Care Excellence has advocated the widespread adoption of multicomponent delirium prevention interventions into the routine inpatient care of older people. However, despite successful reductions in incident delirium of about a third, anticipated reductions in mortality or admissions to long-term care--both clinically important endpoints statistically correlated with the occurrence of delirium--have not been conclusively observed. We hypothesise that the reasons for this disconnection are partly methodological, due to difficulties in delirium detection and blinding of study personnel to the intervention, but predominantly due to the underlying relationship between delirium and the abnormal health state of frailty; the interaction between these two geriatric syndromes is currently poorly understood.

A Patient Reported Experience Measure (PREM) for use by older people in community services.

BACKGROUND intermediate care (IC) services operate between health and social care and are an essential component of integrated care for older people. Patient Reported Experience Measures (PREMs) offer an objective measure of user experience and a practical way to measure person-centred, integrated care in IC settings. OBJECTIVE to describe the development of PREMs suitable for use in IC services and to examine their feasibility, acceptability and scaling properties. SETTING 131 bed-based and 143 home-based or re-ablement IC services in England. METHODS PREMs for each of home- and bed-based IC services were developed through consensus. These were incorporated into the 2013 NAIC and distributed to 50 consecutive users of each bed-based and 250 users of each home-based service. Return rates and patterns of missing data were examined. Scaling properties of the PREMs were examined with Mokken analysis. RESULTS 1,832 responses were received from users of bed-based and 4,627 from home-based services (return rates 28 and 13%, respectively). Missing data were infrequent. Mokken analysis of completed bed-based PREMs (1,398) revealed 8 items measuring the same construct and forming a medium strength (Loevinger H 0.44) scale with acceptable reliability (ρ = 0.76). Analysis of completed home-based PREMs (3,392 records) revealed a medium-strength scale of 12 items (Loevinger H 0.41) with acceptable reliability (ρ = 0.81). CONCLUSIONS the two PREMs offer a method to evaluate user experience of both bed- and home-based IC services. Each scale measures a single construct with moderate scaling properties, allowing summation of scores to give an overall measure of experience.

A review of stroke outcome measures valid and reliable for administration by postal survey

Collecting outcome measures by patient or proxy-completed postal survey in stroke research offers a pragmatic and cost-effective alternative to interview-based assessments. The psychometric properties of outcome measures cannot be assumed to be equivalent across methods of questionnaire administration. Many stroke outcome measures have variable or unproven psychometric properties when administered by post. The validity of stroke research that uses postal surveys may be improved through the adoption of questionnaires with acceptable postal psychometric properties. This review identifies 60 reports of quantitative stroke studies using one or more of 36 instruments to collect stroke outcome data by postal survey. Three of these instruments have acceptable psychometric properties for postal administration in stroke populations (the Frenchay Activities Index (FAI), Subjective Index of Physical and Social Outcome (SIPSO) and the EuroQoL (EQ5D)). Two further instruments lack evidence to support proxy reliability (Nottingham Extended Activities of Daily Living and London Handicap Score), but have otherwise acceptable properties.

A prospective observational study to investigate utility of the Delirium Observational Screening Scale (DOSS) to detect delirium in care home residents

Background care home residents are particularly at risk of delirium due to high prevalence of dementia. The Delirium Observation Screening Scale (DOSS) identifies behavioural changes associated delirium onset that nursing staff are uniquely placed to recognise. We tested the psychometric properties of the DOSS in UK care homes compared with the Confusion Assessment Method (CAM). Design prospective observational cohort study performed between 1 March 2015 and 30 June 2016. Setting nine UK residential and nursing care homes. Subjects residents over 65 years except those approaching end of life or unable to complete delirium assessments. Methods the 25-item DOSS was completed daily by care home staff and compared with the temporally closest CAM performed twice per week by trained researchers. Sensitivity, specificity, positive and negative predictive values, diagnostic odds and likelihood ratios were calculated. Results 216 residents participated; mean age 84.9 (SD 7.9); 50% had cognitive impairment (median AMTS 7 (IQR 3-9)). Half of all expected DOSS assessments occurred (30,201); of these, 11,659 (39%) were complete. 78 positive CAM measurements were made during 71 delirium episodes in 45 residents over 70 weeks. Sensitivity and specificity for delirium detection were optimised at a DOSS cut point of ≥5 (sensitivity 0.61 (95% CI: 0.39-0.80) and specificity (0.71 95% CI: 0.70-0.73)). Positive and negative predictive values were 1.6 and 99.5%, respectively. Conclusions the low sensitivity of the DOSS limits clinical utility for detection of delirium as part of routine care for care home residents, although a negative DOSS affords confidence that delirium is not present.

Delirium: a guide for the general physician

ABSTRACT Delirium describes a sudden onset change in mental status of fluctuating course. This is a state of altered consciousness characterised chiefly by inattention or lack of arousal, but can also include new impairment of language, perception and behaviour. Certain predisposing factors can make an individual more susceptible to delirium in the face of a stressor. Stressors include direct insults to the brain, insults peripheral to the brain or external changes in the environment of an individual. Delirium is varied in its presentation, and can be categorised by the psychomotor profile as: hyperactive type (overly vigilant, agitated, often wandersome), hypoactive type (sedate or withdrawn) or mixed types.

Study protocol—investigation of the Delirium Observation Screening Scale (DOSS) for the routine detection of delirium in the care home setting: a prospective cohort study

Introduction Delirium is a common and distressing condition associated with frailty, dementia and comorbidity. These are common in long-term care settings. Residents in care homes are therefore at particular risk of delirium. Despite this, methods to detect delirium in care homes are lacking, with existing diagnostic tools taking too long, or requiring specific training to deliver. This limits their feasibility for use for the routine detection of delirium by care home staff. Routine screening for delirium in care homes would allow timely attention to exacerbating factors to attenuate the episode, and facilitate future research into delirium in the care home environment. Methods Residents from 4 large care homes will be asked to consent (or their consultees asked to provide a declaration of agreement) to participate in the study. Care home staff will administer the 25-item Delirium Observation Screening Scale (DOSS)—a delirium screening tool based on observed behaviours—and this will be tested against the research standard Confusion Assessment Method (CAM) administered by trained research assistants performed two times per week for all participating residents. Analysis Sensitivity, specificity, positive and negative predictive values, likelihood ratios and a diagnostic OR will be calculated for the detection of delirium with the 25-item DOSS. The feasibility of routine delirium screening and the scaling properties of the 25-item DOSS will also be explored. Ethics and Dissemination For residents lacking capacity to participate, a consultee will be approached for a declaration of agreement for inclusion in the study. Results will be published in peer-reviewed journals and disseminated in written format to clinical commissioning groups, general practitioners and relevant third parties. Trial registration number ISRCTN14608554.

Incidence of first stroke and ethnic differences in stroke pattern in Bradford, UK: Bradford Stroke Study

Background Information on ethnic disparities in stroke between White and Pakistani population in Europe is scarce. Bradford District has the largest proportion of Pakistani people in England; this provides a unique opportunity to study the difference in stroke between the two major ethnic groups. Aim To determine the first-ever-stroke incidence and examine the disparities in stroke patterns between Whites and Pakistanis in Bradford. Methods Prospective 12 months study consisting of 273,327 adults (≥18 years) residents. Stroke cases were identified by multiple overlapping approaches. Results In the study period, 541 first-ever-strokes were recorded. The crude incidence rate was 198 per 100,000 person-years. Age adjusted-standardized rate to the World Health Organization world population of first-ever-stroke is 155 and 101 per 100,000 person-years in Pakistanis and Whites respectively. Four hundred and thirty-eight patients (81%) were Whites, 83 (15.3%) were Pakistanis, 11 (2%) were Indian and Bangladeshis, and 9 (1.7%) were of other ethnic origin. Pakistanis were significantly younger and had more obesity (p = 0.049), and diabetes mellitus (DM) (p = <0.001). They were less likely to suffer from atrial fibrillation (p = <0.001), be ex- or current smokers (p = <0.001), and drink alcohol above the recommended level (p = 0.007) compared with Whites. In comparison with Whites, higher rates of age-adjusted stroke (1.5-fold), lacunar infarction (threefold), and ischemic infarction due to large artery disease (twofold) were found in the Pakistanis. Conclusions The incidence of first-ever-stroke is higher in the Pakistanis compared with the Whites in Bradford, UK. Etiology and vascular risk factors vary between the ethnic groups. This information should be considered when investigating stroke etiology, and when planning prevention and care provision to improve outcomes after stroke.