Search

Background Simple blood tests can play an important role in identifying patients for cancer investigation. The current evidence base is limited almost entirely to tests used in isolation. However, recent evidence suggests combining multiple types of blood tests and investigating trends in blood test results over time could be more useful to select patients for further cancer investigation. Such trends could increase cancer yield and reduce unnecessary referrals. We aim to explore whether trends in blood test results are more useful than symptoms or single blood test results in selecting primary care patients for cancer investigation. We aim to develop clinical prediction models that incorporate trends in blood tests to identify the risk of cancer. Methods Primary care electronic health record data from the English Clinical Practice Research Datalink Aurum primary care database will be accessed and linked to cancer registrations and secondary care datasets. Using a cohort study design, we will describe patterns in blood testing (aim 1) and explore associations between covariates and trends in blood tests with cancer using mixed-effects, Cox, and dynamic models (aim 2). To build the predictive models for the risk of cancer, we will use dynamic risk modelling (such as multivariate joint modelling) and machine learning, incorporating simultaneous trends in multiple blood tests, together with other covariates (aim 3). Model performance will be assessed using various performance measures, including c-statistic and calibration plots. Discussion These models will form decision rules to help general practitioners find patients who need a referral for further investigation of cancer. This could increase cancer yield, reduce unnecessary referrals, and give more patients the opportunity for treatment and improved outcomes.

Colorectal cancer has low survival rates when late-stage, so earlier detection is important. The full blood count (FBC) is a common blood test performed in primary care. Relevant trends in repeated FBCs are related to colorectal cancer presence. We developed and internally validated dynamic prediction models utilising trends for early detection. We performed a cohort study. Sex-stratified multivariate joint models included age at baseline (most recent FBC) and simultaneous trends over historical haemoglobin, mean corpuscular volume (MCV), and platelet measurements up to baseline FBC for two-year risk of diagnosis. Performance measures included the c-statistic and calibration slope. We analysed 250,716 males and 246,695 females in the development cohort and 312,444 males and 462,900 females in the validation cohort, with 0.4% of males and 0.3% of females diagnosed two years after baseline FBC. Compared to average population trends, patient-level declines in haemoglobin and MCV and rise in platelets up to baseline FBC increased risk of diagnosis in two years. C-statistic: 0.751 (males) and 0.763 (females). Calibration slope: 1.06 (males) and 1.05 (females). Our models perform well, with low miscalibration. Utilising trends could bring forward diagnoses to earlier stages and improve survival rates. External validation is now required.

A Full Blood Count (FBC) is a common blood test including 20 parameters, such as haemoglobin and platelets. FBCs from Electronic Health Record (EHR) databases provide a large sample of anonymised individual patient data and are increasingly used in research. We describe the quality of the FBC data in one EHR. The Test dataset from the Clinical Research Practice Datalink (CPRD) was accessed, which contains results of tests performed in primary care, such as FBC blood tests. Medical codes and entity codes, two coding systems used within CPRD to identify FBC records, were compared, with levels of mismatched coding, and number that could be rectified reported. The reliability of units of measurement are also described and missing data discussed. There were 14 entity codes and 138 medical codes for the FBC in the data. Medical and entity codes consistently corresponded to the same FBC parameter in 95.2% (n = 217,752,448) of parameters. In the 4.8% (n = 10,955,006) mismatches, the most common parameter rectified was mean platelet volume (n = 2,041,360) and 1,191,540 could not be rectified and were removed. Units of measurement were often either missing, partially entered, or did not appear to correspond to the blood value. The final dataset contained 16,537,017 FBC tests. Applying mathematical equations to derive some missing parameters in these FBCs resulted in 15 of 20 parameters available per FBC on average, with 0.3% of FBCs having all 20 parameters. Performing data quality checks can help to understand the extent of any issues in the dataset. We emphasise balancing large sample sizes with reliability of the data.

Background: Recent studies show that adults with intellectual disabilities (ID) have high incidence of major osteoporotic fracture, especially hip fracture. In those ≥ 50 years, women and men with ID have an approximately two and four times higher rate of hip fracture than women and men without ID. Increased awareness of osteoporotic fracture risk in ID may lead to wider use of antiresorptiverugs (bisphosphonates and denosumab) in this population. We aimed to compare, between people with and without ID, the incidence of 1) major side effects, namely medication related osteoporosis of the jaw (ONJ) and oesophagitis; 2) oral pathology, which can be a risk factor for ONJ. Methods: Exploratory study investigating safety of first line osteoporosis medication within the population of a previous study comparing fracture incidence in people with and without ID in the GOLD database of the Clinical Practice Research Datalink 1998-2017. Results: The percentage of people on antiresorptive drugs was identical in the ID and non ID group (1.4%). The number of individuals who developed ONJ and oesophagitis during the study was too low to allow an accurate estimate of incidence of the events and a comparison between the two groups. The incidence of any oral pathology was 119.31 vs 64.68/10000 person year in the ID vs non ID group. Conclusions: Medication related ONJ and oesophagitis are rare in people with and without ID. There is no reason based on our findings to use antiresorptives differently in people with ID as in the rest of the population. However, the potential for side effects of antiresorptives will inherently increase with wider use of these drugs. Given the higher incidence of oral pathology in people with ID, which could put them at higher risk of ONJ, precautions should be taken to prevent this complication by attention to oral health.

Introduction A full blood count (FBC) blood test includes 20 components. We systematically reviewed studies that assessed the association of the FBC and diagnosis of colorectal cancer to identify components as risk factors. We reviewed FBC-based prediction models for colorectal cancer risk. Methods: MEDLINE, EMBASE, CINAHL, and Web of Science were searched until 3 September 2019. We meta-analysed the mean difference in FBC components between those with and without a diagnosis and critically appraised the development and validation of FBC-based prediction models. Results We included 53 eligible articles. Three of four meta-analysed components showed an association with diagnosis. In the remaining 16 with insufficient data for meta-analysis, three were associated with colorectal cancer. Thirteen FBC-based models were developed. Model performance was commonly assessed using the c-statistic (range 0.72–0.91) and calibration plots. Some models appeared to work well for early detection but good performance may be driven by early events. Conclusion Red blood cells, haemoglobin, mean corpuscular volume, red blood cell distribution width, white blood cell count, and platelets are associated with diagnosis and could be used for referral. Existing FBC-based prediction models might not perform as well as expected and need further critical testing.

BackgroundThe complexity of general practice consultations may be increasing and varies in different settings. A measure of complexity is required to test these hypotheses.AimTo develop a valid measure of general practice consultation complexity applicable to routine medical records.Design and settingDelphi study to select potential indicators of complexity followed by a cross-sectional study in English general practices to develop and validate a complexity measure.MethodThe online Delphi study over two rounds identified potential indicators of consultation complexity. The cross-sectional study used an age–sex stratified random sample of patients and general practice face-to-face consultations from 2013/2014 in the Clinical Practice Research Datalink. The authors explored independent relationships between each indicator and consultation duration using mixed-effects regression models, and revalidated findings using data from 2017/2018. The proportion of complex consultations in different age–sex groups was assessed.ResultsA total of 32 GPs participated in the Delphi study. The Delphi panel endorsed 34 of 45 possible complexity indicators after two rounds. After excluding factors because of low prevalence or confounding, 17 indicators were retained in the cross-sectional study. The study used data from 173 130 patients and 725 616 face-to-face GP consultations. On defining complexity as the presence of any of these 17 factors, 308 370 consultations (42.5%) were found to be complex. Mean duration of complex consultations was 10.49 minutes, compared to 9.64 minutes for non-complex consultations. The proportion of complex consultations was similar in males and females but increased with age.ConclusionThe present consultation complexity measure has face and construct validity. It may be useful for research, management and policy, and for informing decisions about the range of resources needed in different practices.

ObjectiveMost current cardiovascular disease (CVD) risk stratification tools are for people without CVD, but very few are for prevalent CVD. In this study, we developed and validated a CVD severity score in people with coronary heart disease (CHD) and evaluated the association between severity and adverse outcomes.MethodsPrimary and secondary care data for 213 088 people with CHD in 398 practices in England between 2007 and 2017 were used. The cohort was randomly divided into training and validation datasets (80%/20%) for the severity model. Using 20 clinical severity indicators (each assigned a weight=1), baseline and longitudinal CVD severity scores were calculated as the sum of indicators. Adjusted Cox and competing-risk regression models were used to estimate risks for all-cause and cause-specific hospitalisation and mortality.ResultsMean age was 64.5±12.7 years, 46% women, 16% from deprived areas, baseline severity score 1.5±1.2, with higher scores indicating a higher burden of disease. In the training dataset, 138 510 (81%) patients were hospitalised at least once, and 39 944 (23%) patients died. Each 1-unit increase in baseline severity was associated with 41% (95% CI 37% to 45%, area under the receiver operating characteristics (AUROC) curve=0.79) risk for 1 year for all-cause mortality; 59% (95% CI 52% to 67%, AUROC=0.80) for cardiovascular (CV)/diabetes mortality; 27% (95% CI 26% to 28%) for any-cause hospitalisation and 37% (95% CI 36% to 38%) for CV/diabetes hospitalisation. Findings were consistent in the validation dataset.ConclusionsHigher CVD severity score is associated with higher risks for any-cause and cause-specific hospital admissions and mortality in people with CHD. Our reproducible score based on routinely collected data can help practitioners better prioritise management of people with CHD in primary care.

High Mountain Disease (HMD) in cattle is a consequence of altitude-induced (> 1500 m) pulmonary hypertension (PH). Pulmonary arterial pressure (PAP; mmHg) is an indicator of PH-HMD and moderately heritable. This study analyzed repeated measures of PAP in growing bulls and estimated correlations across elevations and ages. Data included breed, PAP, elevation, weights and ages in spring-born bulls from the Agriculture Research, Development, and Education Center (ARDEC) in Colorado. Five PAP measurements were collected from each bull: 1) Weaning PAP at ARDEC (1,525 m); 2) Yearling PAP at ARDEC (i.e., gain-test); 3) PAP after acclimating for 28 days at Fort Lewis College (FLC; 2,470 m); 4) before returning to ARDEC from FLC after 110 days of grazing; and 5) after re-acclimating for 57 days to the moderate elevation and feeding systems at ARDEC when bulls were 18 months of age. Across these 5 times in Angus (n = 18) and Hereford (n = 12) bulls, PAP measures were: 1) 37.97 ± 0.37; 2) 40.93 ± 0.77; 3) 46.10 ± 1.48; 4) 45.00 ± 1.11; and 5) 42.23 ± 0.67, respectively. Breed and elevation-age were important (P < 0.05) sources of variation. Also, PAP changed (P < 0.05) across the 5 time periods (i.e., increased with altitude and then declined as bulls returned to moderate altitude). Yearling PAP measurements collected at 1,525 m were highly correlated with PAP measures at 2,470 m, as well as a PAP collected at 18.3 ± 0.1 mo of age at 1,525 m (r = 0.93 to 0.99); therefore, suggesting that the yearling PAP measurement collected at 1,525 m in gain-tested bulls was an adequate predictor of PAP at elevations of 1,525 and 2,470 m as bulls aged to 18 months; however, additional study of bovine PAP across elevations, ages, and rates of gain is warranted.

The original article [1] contains an omitted grant acknowledgement and affiliation as relates to the contribution of co-author, Rafael Perera-Salazar. As such, the following two amendments should apply to the original article

© 2017 The Authors. Cancer Medicine published by John Wiley & Sons Ltd. Earlier detection of colorectal cancer greatly improves prognosis, largely through surgical excision of neoplastic polyps. These include benign adenomas which can transform over time to malignant adenocarcinomas. This progression may be associated with changes in full blood count indices. An existing risk algorithm derived in Israel stratifies individuals according to colorectal cancer risk using full blood count data, but has not been validated in the UK. We undertook a retrospective analysis using the Clinical Practice Research Datalink. Patients aged over 40 with full blood count data were risk-stratified and followed up for a diagnosis of colorectal cancer over a range of time intervals. The primary outcome was the area under the receiver operating characteristic curve for the 18–24-month interval. We also undertook a case–control analysis (matching for age, sex, and year of risk score), and a cohort study of patients undergoing full blood count testing during 2012, to estimate predictive values. We included 2,550,119 patients. The area under the curve for the 18–24-month interval was 0.776 [95% confidence interval (CI): 0.771, 0.781]. Performance improves as the time interval reduces. The area under the curve for the age-matched case–control analysis was 0.583 [0.574, 0.591]. For the population risk-scored in 2012, the positive predictive value at 99.5% specificity was 8.8% with negative predictive value 99.6%. The algorithm offers an additional means of identifying risk of colorectal cancer, and could support other approaches to early detection, including screening and active case finding.

Background and Purpose— Oral anticoagulants (OAC) substantially reduce risk of stroke in atrial fibrillation, but uptake is suboptimal. Electronic health records enable automated identification of people at risk but not receiving treatment. We investigated the effectiveness of a software tool (AURAS-AF [Automated Risk Assessment for Stroke in Atrial Fibrillation]) designed to identify such individuals during routine care through a cluster-randomized trial. Methods— Screen reminders appeared each time the electronic health records of an eligible patient was accessed until a decision had been taken over OAC treatment. Where OAC was not started, clinicians were prompted to indicate a reason. Control practices continued usual care. The primary outcome was the proportion of eligible individuals receiving OAC at 6 months. Secondary outcomes included rates of cardiovascular events and reports of adverse effects of the software on clinical decision-making. Results— Forty-seven practices were randomized. The mean proportion–prescribed OAC at 6 months was 66.3% (SD=9.3) in the intervention arm and 63.9% (9.5) in the control arm (adjusted difference 1.21% [95% confidence interval −0.72 to 3.13]). Incidence of recorded transient ischemic attack was higher in the intervention practices (median 10.0 versus 2.3 per 1000 patients with atrial fibrillation; P =0.027), but at 12 months, we found a lower incidence of both all cause stroke ( P =0.06) and hemorrhage ( P =0.054). No adverse effects of the software were reported. Conclusions— No significant change in OAC prescribing occurred. A greater rate of diagnosis of transient ischemic attack (possibly because of improved detection or overdiagnosis) was associated with a reduction (of borderline significance) in stroke and hemorrhage over 12 months. Clinical Trial Registration— URL: http://www.isrctn.com . Unique Identifier: ISRCTN55722437.

Background Most people with type 2 diabetes are overweight, so initial treatment is aimed at reducing weight and increasing physical activity. Even modest weight loss can improve control of blood glucose. If drug treatment is necessary, the drug of first choice is metformin. However, some people cannot tolerate metformin, which causes diarrhoea in about 10%, and it cannot be used in people with renal impairment. This review appraises three of the newest class of drugs for monotherapy when metformin cannot be used, the sodium–glucose co-transporter 2 (SGLT2) inhibitors. Objective To review the clinical effectiveness and cost-effectiveness of dapagliflozin (Farxiga, Bristol-Myers Squibb, Luton, UK), canagliflozin (Invokana, Janssen, High Wycombe, UK) and empagliflozin (Jardiance, Boehringer Ingelheim, Ingelheim, Germany/Eli Lilly and Company, Indianapolis, IN, USA), in monotherapy in people who cannot take metformin. Sources MEDLINE (1946 to February 2015) and EMBASE (1974 to February 2015) for randomised controlled trials lasting 24 weeks or more. For adverse events, a wider range of studies was used. Three manufacturers provided submissions. Methods Systematic review and economic evaluation. A network meta-analysis was carried out involving the three SGLT2 inhibitors and key comparators. Critical appraisal of submissions from three manufacturers. Results We included three trials of dapagliflozin and two each for canagliflozin and empagliflozin. The trials were of good quality. The canagliflozin and dapagliflozin trials compared them with placebo, but the two empagliflozin trials included active comparators. All three drugs were shown to be effective in improving glycaemic control, promoting weight loss and lowering blood pressure (BP). Limitations There were no head-to-head trials of the different flozins, and no long-term data on cardiovascular outcomes in this group of patients. Most trials were against placebo. The trials were done in patient groups that were not always comparable, for example in baseline glycated haemoglobin or body mass index. Data on elderly patients were lacking. Conclusions Dapagliflozin, canagliflozin and empagliflozin are effective in improving glycaemic control, with added benefits of some reductions in BP and weight. Adverse effects are urinary and genital tract infections in a small proportion of users. In monotherapy, the three drugs do not appear cost-effective compared with gliclazide or pioglitazone, but may be competitive against sitagliptin (Januvia, Merck Sharp & Dohme Limited, Kenilworth, NJ, USA). Funding The National Institute for Health Research Health Technology Assessment programme.

IntroductionColorectal cancer is the fourth most common type of cancer and the second most common cause of cancer-related deaths in the UK. The full blood count (FBC) is a blood test that may play a role in early detection of the disease. Previous studies have aimed to identify how levels of individual components, such as haemoglobin, can be used to assist the diagnosis. We aim to systematically review studies to identify whether components of the FBC are risk factors for diagnosis of colorectal cancer, critically appraise the methods used to assess the association and assess performance of the components.Methods and analysisThe MEDLINE (via OVID), EMBASE (via OVID), CINAHL (via EBSCOhost) and Web of Science databases will be searched to identify studies reporting the association between the levels of at least one FBC component and the risk of a future diagnosis of colorectal cancer in undiagnosed individuals. Clincialtrials.gov and the WHO registry will be searched to identify relevant ongoing research. Search terms will include relevant Medical Subject Headings and Emtree headings, and free-text terms relating to FBC, colorectal cancer and diagnosis. No date or language restrictions will be applied. Two reviewers will independently identify the studies for inclusion and perform data extraction. Time intervals between the blood tests and diagnosis will form the subgroups for analysis.Ethics and disseminationThere is no direct patient involvement and only published articles will be reviewed; no ethical approval is required. Results from this review will set a foundation for intended future work on developing a new risk score for early detection of colorectal cancer, derived using FBC data. This systematic review will also provide guidance on the analysis of time to diagnosis. The model will be freely available to UK primary care practices.PROSPERO registration numberCRD42019134400.

BackgroundOral anticoagulants reduce the risk of stroke in patients with atrial fibrillation (AF), but are underused. AURAS-AF (AUtomated Risk Assessment for Stroke in AF) is a software tool designed to identify eligible patients and promote discussions within consultations about initiating anticoagulants.AimTo investigate the implementation of the software in UK general practice.Design and settingProcess evaluation involving 23 practices randomly allocated to use AURAS-AF during a cluster randomised trial.MethodAn initial invitation to discuss anticoagulation was followed by screen reminders appearing during consultations until a decision had been made. The reminders required responses, giving reasons for cases where an anticoagulant was not initiated. Qualitative interviews with clinicians and patients explored acceptability and usability.ResultsIn a sample of 476 patients eligible for the invitation letter, only 159 (33.4%) were considered suitable for invitation by their GPs. Reasons given were frequently based on frailty, and risk of falls or haemorrhage. Of those invited, 35 (22%) started an anticoagulant (7.4% of those originally identified). A total of 1695 main-screen reminders occurred in 940 patients. In 883 instances, the decision was taken not to initiate and a range of reasons offered. Interviews with 15 patients and seven clinicians indicated that the intervention was acceptable, though the issue of disruptive screen reminders was raised.ConclusionAutomated risk assessment for stroke in atrial fibrillation and prompting during consultations are feasible and generally acceptable, but did not overcome concerns about frailty and risk of haemorrhage as barriers to anticoagulant uptake.

Background Shoulder dislocations are the most common joint dislocations seen in emergency departments. Most traumatic cases are anterior and cause recurrent dislocations. Management options include surgical and conservative treatments. There is a lack of evidence about which method is most effective after the first traumatic anterior shoulder dislocation (TASD). Objectives To produce UK age- and sex-specific incidence rates for TASD. To assess whether or not surgery within 6 months of a first-time TASD decreases re-dislocation rates compared with no surgery. To identify clinical predictors of recurrent dislocation. Design A population-based cohort study of first-time TASD patients in the UK. An initial validation study and subsequent propensity-score-matched analysis to compare re-dislocation rates between surgery and no surgery after a first-time TASD. Prediction modelling was used to identify potential predictors of recurrent dislocation. Setting UK primary and secondary care data. Participants Patients with a first-time TASD between 1997 and 2015. Interventions Stabilisation surgery within 6 months of a first-time TASD (compared with no surgery). Stabilisation surgery within 12 months of a first-time TASD was also carried out as a sensitivity analysis. Main outcome measure Re-dislocation rate up to 2 years after the first TASD. Methods Eligible patients were identified from the Clinical Practice Research Datalink (CPRD) (1997–2015). Accuracy of shoulder dislocation coding was internally validated using the CPRD General Practitioner questionnaire service. UK age- and sex-specific incidence rates for TASD were externally validated against rates from the USA and Canada. A propensity-score-matched analysis using linked CPRD and Hospital Episode Statistics (HES) data compared re-dislocation rates for patients aged 16–35 years, comparing surgery with no surgery. Multivariable Cox regression models for predicting re-dislocation were developed for the surgical and non-surgical cohorts. Results Shoulder dislocation was coded correctly for 89% of cases in the CPRD [95% confidence interval (CI) 83% to 95%], with a ‘primary’ dislocation confirmed for 76% of cases (95% CI 67% to 85%). Far fewer patients than expected received stabilisation surgery within 6 months of a first TASD, leading to an underpowered study. Around 20% of re-dislocation rates were observed for both surgical and non-surgical patients. The sensitivity analysis at 12 months also showed little difference in re-dislocation rates. Missing data on risk factors limited the value of the prediction modelling; however, younger age, epilepsy and sex (male) were identified as statistically significant predictors of re-dislocation. Limitations Far fewer than the expected number of patients had surgery after a first-time TASD, resulting in an underpowered study. This and residual confounding from missing risk factors mean that it is not possible to draw valid conclusions. Conclusions This study provides, for the first time, UK data on the age- and sex-specific incidence rates for TASD. Most TASD occurs in men, but an unexpected increased incidence was observed in women aged > 50 years. Surgery after a first-time TASD is uncommon in the NHS. Re-dislocation rates for patients receiving surgery after their first TASD are higher than previously expected; however, important residual confounding risk factors were not recorded in NHS primary and secondary care databases, thus preventing useful recommendations. Future work The high incidence of TASD justifies investigation into preventative measures for young men participating in contact sports, as well as investigating the risk factors in women aged > 50 years. A randomised controlled trial would account for key confounders missing from CPRD and HES data. A national TASD registry would allow for a more relevant data capture for this patient group. Study registration Independent Scientific Advisory Committee (ISAC) for the Medicines and Healthcare Products Regulatory Agency (ISAC protocol 15_0260). Funding The National Institute for Health Research Health Technology Assessment programme.

Background The presence of additional chronic conditions has a significant impact on the treatment and management of type 2 diabetes (T2DM). Little is known about the patterns of comorbidities in this population. The aims of this study are to quantify comorbidity patterns in people with T2DM, to estimate the prevalence of six chronic conditions in 2027 and to identify clusters of similar conditions. Methods We used the Clinical Practice Research Datalink (CPRD) linked with the Index of Multiple Deprivation (IMD) data to identify patients diagnosed with T2DM between 2007 and 2017. 102,394 people met the study inclusion criteria. We calculated the crude and age-standardised prevalence of 18 chronic conditions present at and after the T2DM diagnosis. We analysed longitudinally the 6 most common conditions and forecasted their prevalence in 2027 using linear regression. We used agglomerative hierarchical clustering to identify comorbidity clusters. These analyses were repeated on subgroups stratified by gender and deprivation. Results More people living in the most deprived areas had ≥ 1 comorbidities present at the time of diagnosis (72% of females; 64% of males) compared to the most affluent areas (67% of females; 59% of males). Depression prevalence increased in all strata and was more common in the most deprived areas. Depression was predicted to affect 33% of females and 15% of males diagnosed with T2DM in 2027. Moderate clustering tendencies were observed, with concordant conditions grouped together and some variations between groups of different demographics. Conclusions Comorbidities are common in this population, and high between-patient variability in comorbidity patterns emphasises the need for patient-centred healthcare. Mental health is a growing concern, and there is a need for interventions that target both physical and mental health in this population. Electronic supplementary material The online version of this article (10.1186/s12916-019-1373-y) contains supplementary material, which is available to authorized users.

In this issue, Fernando et al present a systematic review and meta-analysis assessing the prognostic accuracy of the HEART score for prediction of major adverse cardiac events (MACE) in adult patients presenting with chest pain at the emergency department (ED).1 The authors conclude that the HEART score has excellent performance for prediction of MACE (particularly mortality and myocardial infarction) in chest pain patients and should be the primary clinical decision instrument used for risk-stratification of this patient population. This article is protected by copyright. All rights reserved.

HighlightsData-driven model is presented for the response of glucose to food intake.Model describes blood glucose dynamics in people with and without diabetes.First study introducing a continuous data-driven nonlinear stochastic model.Model's parameters belong to different ranges for diabetes and controls.Variational Bayesian learning approach was employed for model identification. The development of adequate mathematical models for blood glucose dynamics may improve early diagnosis and control of diabetes mellitus (DM). We have developed a stochastic nonlinear second order differential equation to describe the response of blood glucose concentration to food intake using continuous glucose monitoring (CGM) data. A variational Bayesian learning scheme was applied to define the number and values of the system's parameters by iterative optimisation of free energy. The model has the minimal order and number of parameters to successfully describe blood glucose dynamics in people with and without DM. The model accounts for the nonlinearity and stochasticity of the underlying glucose-insulin dynamic process. Being data-driven, it takes full advantage of available CGM data and, at the same time, reflects the intrinsic characteristics of the glucose-insulin system without detailed knowledge of the physiological mechanisms. We have shown that the dynamics of some postprandial blood glucose excursions can be described by a reduced (linear) model, previously seen in the literature. A comprehensive analysis demonstrates that deterministic system parameters belong to different ranges for diabetes and controls. Implications for clinical practice are discussed. This is the first study introducing a continuous data-driven nonlinear stochastic model capable of describing both DM and non-DM profiles.

Background Acute coronary syndrome (ACS) is a common, sometimes difficult to diagnose spectrum of diseases occurring after abrupt reduction in blood flow through a coronary artery. Given the diagnostic challenge, it is sensible for emergency physicians to have an approach to prognosticate patients with possible ACS. Multiple prediction models have been developed to help identify patients at increased risk of adverse outcomes. The HEART score is the first model to be derived, validated, and undergo clinical impact studies in emergency department (ED) patients with possible ACS. Objective To develop a protocol for a prognostic systematic review of the literature evaluating the HEART score as a predictor of major adverse cardiac events (MACE) in patients presenting to the ED with possible ACS. Methods/design This protocol is reported according to the PRISMA-P statement and is registered on PROSPERO. All methodological tools to be used are endorsed by the Cochrane Prognosis Methods Group. Pre-defined eligibility criteria are provided. Multiple strategies will be used to identify potentially relevant studies. Studies will be selected and data extracted using standardised forms based on the CHARMS checklist. The QUIPS tool will be used to assess the risk of bias within individual studies. Outcome measures will include prevalence, risk ratio, and absolute risk reduction for MACE within 6 weeks of ED evaluation, comparing HEART scores 0–3 versus 4–10. HEART score prognostic performance will be evaluated with the concordance (C) statistic (model discrimination), observed to expected events ratio (model calibration), and a decision curve analysis. Reporting biases and methodological, clinical, and statistical heterogeneity will be scrutinised. Unless deemed inappropriate, a meta-analysis and pre-defined subgroup and sensitivity analyses will be performed. Overall judgements about evidence quality and strength of recommendations will be summarised using the GRADE approach. Discussion This review will identify, select, and appraise studies evaluating the prognostic performance of the HEART score, producing results of interest to emergency physicians. These results may encourage shared clinical decision-making in the ED by facilitating risk communication with patients and health care providers. Systematic review registration PROSPERO 2017 CRD42017084400. Electronic supplementary material The online version of this article (10.1186/s13643-018-0816-4) contains supplementary material, which is available to authorized users.

Background The ESTEEM trial was a cluster randomised controlled trial that compared two telephone triage management systems (general practitioner (GP) or a nurse supported by computer decision support software) with usual care, in response to a request for same-day consultation in general practice. Aim To investigate associations between trial patients’ demographic, health, and lifestyle characteristics, and their reported experiences of care. Setting Recruitment of 20 990 patients occurred between May 2011 and December 2012 in 42 GP practices in England (13 GP triage, 15 nurse triage, 14 usual care). Method Patients reported their experiences via a postal questionnaire issued 4 weeks after their initial request for a same-day consultation. Overall satisfaction, ease of accessing medical help/advice, and convenience of care were analysed using linear hierarchical modelling. Results Questionnaires were returned by 12 132 patients (58%). Older patients reported increased overall satisfaction compared with patients aged 25–59 years, but patients aged 16–24 years reported lower satisfaction. Compared with white patients, patients from ethnic minorities reported lower satisfaction in all three arms, although to a lesser degree in the GP triage arm. Patients from ethnic minorities reported higher satisfaction in the GP triage than in usual care, whereas white patients reported higher satisfaction with usual care. Patients unable to take time away from work or who could only do so with difficulty reported lower satisfaction across all three trial arms. Conclusions Patient characteristics, such as age, ethnicity and ability to attend their practice during work hours, were associated with their experiences of care following a same-day consultation request in general practice. Telephone triage did not increase satisfaction among patients who were unable to attend their practice during working hours. Trial registration number ISCRTN20687662.

Purpose To investigate whether information from the literature could be used to identify periods of practice data in an electronic healthcare database during which rheumatoid factor (RF) test results are likely to be missing-not-at-random (MNAR). Methods RF tests recorded in the Clinical Practice Research Datalink (CPRD) were identified and defined as having a positive, negative or missing result. The proportion of positive test results was then calculated based on (i) complete-case analysis (ii) after restriction to tests from practice years with no missing test results and (iii) following multiple imputation of missing test results. The same three analyses were then carried out after excluding practice years with a proportion of positive tests incompatible with the missing completely at random (MCAR) assumption. Results We identified 127 969 RF test records, 30.4% of which did not have an associated test result. Among tests with results available, 19% were positive. Both multiple imputation of the 38 867 missing test results and restriction of the study population to the 491 practice years with complete data had little impact on the percentage of positive tests. Following exclusion of the 544 practice years in which data were likely to be MNAR the percentage of positive tests in all analyses decreased to ~7%. Conclusions Recording of RF tests and RF test results in the CPRD is incomplete, with data likely to be MNAR in many practices. Exclusion of practice years with a high proportion of positive tests brought the distribution of positive tests in the study in line with the literature. Copyright © 2015 John Wiley & Sons, Ltd.

ObjectivesConsultation duration has previously been shown to be associated with patient, practitioner and practice characteristics. However, previous studies were conducted outside the UK, considered only small numbers of general practitioner (GP) consultations or focused primarily on practitioner-level characteristics. We aimed to determine the patient-level and practice-level factors associated with duration of GP and nurse consultations in UK primary care.Design and settingCross-sectional data were obtained from English general practices contributing to the Clinical Practice Research Datalink (CPRD) linked to data on patient deprivation and practice staffing, rurality and Quality and Outcomes Framework (QOF) achievement.Participants218 304 patients, from 316 English general practices, consulting from 1 April 2013 to 31 March 2014.AnalysisMultilevel mixed-effects models described the association between consultation duration and patient-level and practice-level factors (patient age, gender, smoking status, ethnic group, deprivation and practice rurality, number of full-time equivalent GPs/nurses, list size, consultation rate, quintile of overall QOF achievement and training status).ResultsMean duration of face-to-face GP consultations was 9.24 min and 5.32 min for telephone consultations. Nurse face-to-face and telephone consultations lasted 9.70 and 5.73 min on average, respectively. Longer GP consultation duration was associated with female patient gender, practice training status and older patient age. Shorter duration was associated with higher deprivation and consultation rate. Longer nurse consultation duration was associated with male patient gender, older patient age and ever smoking; and shorter duration with higher consultation rate. Observed differences in duration were small (eg, GP consultations with female patients compared with male patients were 8 s longer on average).ConclusionsSmall observed differences in consultation duration indicate that patients are treated similarly regardless of background. Increased consultation duration may be beneficial for older or comorbid patients, but the benefits and costs of increased consultation duration require further study.

BackgroundWorkload in general practice has risen during the last decade, but the factors associated with this increase are unclear.AimTo examine factors associated with consultation rates in general practice.Design and settingA cross-sectional study examining a sample of 304 937 patients registered at 316 English practices between 2013 and 2014, drawn from the Clinical Practice Research Datalink.MethodAge, sex, ethnicity, smoking status, and deprivation measures were linked with practice-level data on staffing, rurality, training practice status, and Quality and Outcomes Framework performance. Multilevel analyses of patient consultation rates were conducted.ResultsConsultations were grouped into three types: all (GP or nurse), GP, and nurse. Non-smokers consulted less than current smokers (all: rate ratio [RR] = 0.88, 95% CI = 0.87 to 0.89; GP: RR = 0.88, 95% CI = 0.87 to 0.89; nurse: RR = 0.91, 95% CI = 0.90 to 0.92). Consultation rates were higher for those in the most deprived quintile compared with the least deprived quintile (all: RR = 1.18, 95% CI = 1.16 to 1.19; GP: RR = 1.17, 95% CI = 1.15 to 1.19; nurse: RR = 1.13, 95% CI = 1.11 to 1.15). For all three consultation types, consultation rates increased with age and female sex, and varied by ethnicity. Rates in practices with >8 and ≤19 full-time equivalent (FTE) GPs were higher compared with those with ≤2 FTE GPs (all: RR = 1.26, 95% CI = 1.06 to 1.49; GP: RR = 1.36, 95% CI = 1.19 to 1.56).ConclusionThe analyses show consistent trends in factors related to consultation rates in general practice across three types of consultation. These data can be used to inform the development of more sophisticated staffing models, and resource allocation formulae.

Objective To ascertain long term cardiovascular outcomes in patients whose chest pain remained undiagnosed six months after first presentation. Design Cohort study. Setting UK electronic health record database (CALIBER) linking primary care, secondary care, coronary registry, and death registry information. Participants 172 180 adults aged ≥18 from 223 general practices presenting with a first episode of recorded chest pain, classified from medical records as diagnosed (non-coronary condition or angina) or undiagnosed (cause unattributed) at first consultation between 2002 and 2009 and with no previous record of cardiovascular disease. Main outcome measures Fatal or non-fatal cardiovascular events over 5.5 years’ follow-up. Adjustments were made for age, sex, deprivation, body mass index, smoking status, year of index presentation, and previous records of diabetes or hypertension or previous prescriptions for lipid lowering drugs. Results At the index presentation, 72.4% of patients (124 688) did not have a cause attributed for their chest pain; 118 687 (95.2%) of these did not receive any type of cardiovascular diagnosis over the next six months. Only a minority of patients in all three groups (non-coronary 2.0% (769 of 39 232); unattributed 11.7% (14 582 of 124 688); angina 31.5% (2606 of 8260)) had a recorded cardiac diagnostic investigation in the first six months after presentation. The long term incidence of cardiovascular events was higher in those whose chest pain remained unattributed after six months (5126 of 109 628; 4.7%) compared with patients with an initial diagnosis of non-coronary pain (1073 of 36 097; 3.0%) (adjusted hazard ratios for 0.5-1 year after presentation: 1.95, 95% confidence interval 1.66 to 2.31; for 1-3 years: 1.35, 1.23 to 1.48); for 3-5.5 years: 1.21, 1.08 to 1.37). Owing to the larger number of patients in the unattributed group, there were more excess myocardial infarctions in the long term in this group (214 more than expected based on the rate in the non-coronary group) than in the angina group (132 more than expected). Patients who had cardiac diagnostic investigations in the first six months had a higher long term risk of cardiovascular events, regardless of the initial chest pain label. Incidence of unattributed chest pain and angina decreased between 2002 (124 per 10 000 person years and 13 per 10 000 person years, respectively) and 2009 (107 per 10 000 person years and 5 per 10 000 person years, respectively), but the incidence of chest pain attributed to a non-coronary cause remained stable (37-40 per 10 000 person years). Risk of cardiovascular events did not change over time. Conclusions Most patients with first onset chest pain do not have a diagnosis recorded at presentation or in the subsequent six months, including those who undergo cardiac investigations. These patients have an increased risk of cardiovascular events for at least five years. Efforts to better assess and reduce the cardiovascular risk of such patients are warranted.

BACKGROUND: Telephone triage is an increasingly common means of handling requests for same-day appointments in general practice.AIM: To determine whether telephone triage (GP-led or nurse-led) reduces clinician-patient contact time on the day of the request (the index day), compared with usual care.DESIGN AND SETTING: A total of 42 practices in England recruited to the ESTEEM trial.METHOD: Duration of initial contact (following the appointment request) was measured for all ESTEEM trial patients consenting to case notes review, and that of a sample of subsequent face-to-face consultations, to produce composite estimates of overall clinician time during the index day.RESULTS: Data were available from 16 711 initial clinician-patient contacts, plus 1290 GP, and 176 nurse face-to-face consultations. The mean (standard deviation) duration of initial contacts in each arm was: GP triage 4.0 (2.8) minutes; nurse triage 6.6 (3.8) minutes; and usual care 9.5 (5.0) minutes. Estimated overall contact duration (including subsequent contacts on the same day) was 10.3 minutes for GP triage, 14.8 minutes for nurse triage, and 9.6 minutes for usual care. In nurse triage, more than half the duration of clinician contact (7.7 minutes) was with a GP. This was less than the 9.0 minutes of GP time used in GP triage.CONCLUSION: Telephone triage is not associated with a reduction in overall clinician contact time during the index day. Nurse-led triage is associated with a reduction in GP contact time but with an overall increase in clinician contact time. Individual practices may wish to interpret the findings in the context of the available skill mix of clinicians.

© 2017 Herrington et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Background It is uncertain whether being overweight, but not obese, is associated with advanced chronic kidney disease (CKD) and how the size and shape of associations between body-mass index (BMI) and advanced CKD differs among different types of people. Methods We used Clinical Practice Research Datalink records (2000-2014) with linkage to English secondary care and mortality data to identify a prospective cohort with at least one BMI measure. Cox models adjusted for age, sex, smoking and social deprivation and subgroup analyses by diabetes, hypertension and prior cardiovascular disease assessed relationships between BMI and CKD stages 4-5 and end-stage renal disease (ESRD). Findings 1,405,016 adults aged 20-79 with mean BMI 27.4kg/m2 (SD 5.6) were followed for 7.5 years. Compared to a BMI of 20 to

Stacey J Ackerman,1 David W Polly Jr,2 Tyler Knight,3 Karen Schneider,4 Tim Holt,5 John Cummings Jr6 1Covance Market Access Services Inc., San Diego, CA, USA; 2University of Minnesota, Orthopaedic Surgery, Minneapolis, MN, USA; 3Covance Market Access Services Inc., Gaithersburg, MD, USA; 4Covance Market Access Services Inc., Sydney, Australia; 5Montgomery Spine Center, Orthopedic Surgery, Montgomery, AL, USA; 6Community Health Network, Neurosurgery, Indianapolis, IN, USA Introduction: Low back pain is common and treatment costly with substantial lost productivity and lost wages in the working-age population. Chronic low back pain originating in the sacroiliac (SI) joint (15%–30% of cases) is commonly treated with nonoperative care, but new minimally invasive surgery (MIS) options are also effective in treating SI joint disruption. We assessed whether the higher initial MIS SI joint fusion procedure costs were offset by decreased nonoperative care costs from a US commercial payer perspective. Methods: An economic model compared the costs of treating SI joint disruption with either MIS SI joint fusion or continued nonoperative care. Nonoperative care costs (diagnostic testing, treatment, follow-up, and retail pharmacy pain medication) were from a retrospective study of Truven Health MarketScan® data. MIS fusion costs were based on the Premier's Perspective™ Comparative Database and professional fees on 2012 Medicare payment for Current Procedural Terminology code 27280. Results: The cumulative 3-year (base-case analysis) and 5-year (sensitivity analysis) differentials in commercial insurance payments (cost of nonoperative care minus cost of MIS) were $14,545 and $6,137 per patient, respectively (2012 US dollars). Cost neutrality was achieved at 6 years; MIS costs accrued largely in year 1 whereas nonoperative care costs accrued over time with 92% of up front MIS procedure costs offset by year 5. For patients with lumbar spinal fusion, cost neutrality was achieved in year 1. Conclusion: Cost offsets from new interventions for chronic conditions such as MIS SI joint fusion accrue over time. Higher initial procedure costs for MIS were largely offset by decreased nonoperative care costs over a 5-year time horizon. Optimizing effective resource use in both nonoperative and operative patients will facilitate cost-effective health care delivery. The impact of SI joint disruption on direct and indirect costs to commercial insurers, health plan beneficiaries, and employers warrants further consideration. Keywords: epidural injection, iFuse, economic model, sacroiliac joint fusion, sacroiliac joint pain, insurance

Introduction: Acute coronary syndrome (ACS) is a common, sometimes difficult to diagnose spectrum of diseases. Given the diagnostic challenge, it is sensible for emergency physicians to have an approach to prognosticate patients with possible ACS. The objective of this review was to investigate the ability of the HEART score to predict major adverse cardiac events (MACE) in patients presenting to the ED with possible ACS. Methods: Eleven databases and other sources identified 468 potentially relevant studies. Sixty-seven studies underwent full text review with 25 studies meeting eligibility criteria. Main outcome measures were pooled prevalence, risk ratio (RR), and absolute risk reduction (ARR) for MACE within six weeks of ED evaluation, comparing HEART score 0–3 versus 4–10. Model discrimination (sensitivity, specificity, concordance statistic) and calibration (observed to expected events ratio) were also evaluated. Results: Data from 25 studies including 41,397 patients were combined in the meta-analysis. In total, 4815 patients (11.6%) developed MACE. Among 18,866 patients with HEART score 0–3, 396 (2.1%) developed MACE (RR 0.08; ARR 0.20). Outcome measures were consistent across planned subgroup and sensitivity analyses. Among studies with secondary outcome data for patients with HEART score 0–3, 5 of 6461 (0.1%) died and 75 of 7556 (1.0%) had a myocardial infarction. Conclusion: The HEART score provides a reliable quantitative risk assessment of MACE in ED patients with possible ACS. Emergency clinicians should consider using the HEART score to facilitate risk communication and shared decision making with patients and other care providers.

BACKGROUND: Electronic diabetes registers promote structured care and enable identification of undiagnosed diabetes, but they require consistent coding of the diagnosis in electronic medical records. We investigated the potential of electronic medical records to identify undiagnosed diabetes and to support diabetes management in a large primary care population in the United States. METHODS: We conducted a cross-sectional study and retrospective observational cohort analysis of primary care electronic medical records from a nationally representative US database (GE Centricity). We tested the feasibility of identifying patients with undiagnosed diabetes by applying simple algorithms to the electronic medical record data. We compared the quality of care provided to patients in the United States who had diabetes (coded and uncoded) for at least 15 months with the quality of care provided in England using a set of 16 indicators. RESULTS: We included 11 540 454 electronic medical records from more than 9000 primary care clinics across the United States. Of the 1 110 398 records indicating diagnosed diabetes, only 61.9% contained a diagnostic code. Of the 10 430 056 records for nondiabetic patients, 0.4% (n = 40 359) had at least 2 abnormal fasting or random blood glucose values, and 0.2% (n = 23 261) of the remaining records had at least 1 documented glycated hemoglobin (HbA1c) value of 6.5% or higher. Among the 622 260 patients for whom information on quality-of-care indicators was available, those with a coded diagnosis of diabetes had a significantly higher level of quality of care than those with uncoded diabetes (p < 0.01); however, the quality of care was generally lower than that indicated in England. INTERPRETATION: We were able to identify a substantial number of patients with uncoded diabetes and probable undiagnosed diabetes using simple algorithms applied to the primary care electronic records. Electronic coding of the diagnosis was associated with improved quality of care. Electronic diabetes registers are underused in US primary care and provide opportunities to facilitate the systematic, structured approach that is established in England.

BACKGROUND: Oral anticoagulants substantially reduce the risk of stroke in atrial fibrillation but are underutilised in current practice. AIM: To measure the distribution of stroke risk in patients with atrial fibrillation (using the CHADS(2) and CHA(2)DS(2)-VASc scores) and changes in oral anticoagulant use during 2007-2010. DESIGN AND SETTING: Longitudinal series of cross-sectional survey in 583 UK practices linked to the QResearch(®) database providing 99 351 anonymised electronic records from people with atrial fibrillation. METHOD: The proportion of patients in each CHADS(2) and CHA(2)DS(2)-VASc risk band in 2010 was calculated; for each of the years 2007-2010, the proportions with risk scores ≥2 that were using anticoagulants or antiplatelet agents were estimated. The proportions identified at high risk were re-estimated using alternative definitions of hypertension based on coded data. Finally, the prevalence of comorbid conditions in treated and untreated high-risk (CHADS(2) ≥2) groups was derived. RESULTS: The proportion at high risk of stroke in 2010 was 56.9% according to the CHADS(2) ≥2 threshold, and 84.5% according to CHA(2)DS(2)-VASc ≥2 threshold. The proportions of these groups receiving anticoagulants were 53.0% and 50.7% respectively and increased during 2007-2010. The means of identifying the population of individuals with hypertension significantly influenced the estimated proportion at high risk. Comorbid conditions associated with avoidance of anticoagulants included history of falls, use of nonsteroidal anti-inflammatory drugs, and dementia. CONCLUSION: Oral anticoagulant use in atrial fibrillation has increased in UK practice since 2007, but remains suboptimal. Improved coding of hypertension is required to support systematic identification of individuals at high risk of stroke and could be assisted by practice-based software.

John Paley’s article is good teaching material because it illustrates some classic rhetorical moves used in the ancient sport of argumentation. These include the ‘post hoc ergo propter hoc’ fallacy (assuming that if B occurred after A then A must have caused B); the ‘slippery slope’ fallacy (suggesting that the opponent is guilty of A, then taking the audience through a series of small steps in which A becomes B, B becomes C etc, and concluding that the opponent is guilty of Z); the ‘ad hominem’ fallacy (depicting the opponent as ignorant or foolish, and then concluding that everything s/he says is weak and superficial); and the ‘ad populum’ fallacy (appealing to some positive but ill-defined quality to which the audience is assumed to aspire – such as ‘rigour’ or ‘depth of understanding’). It is ironic, for example, that in the very first sentence of a paper which purports to explain complexity theory, Paley draws a direct and linear link between a series of introductory articles which we published back in 2001 and alleged misunderstandings and misapplications of complexity theory by others in the health care field. Paley claims that in our 2001 series, we ‘partially understood’ what he chose to define as the second principle of complexity – that ‘successive states of the system, globally defined, are determined by previous states, locally defined’. We apparently ‘failed to recognize’ what Paley decided was complexity theory’s first principle – that complexity is an explanatory concept. It is presumably coincidental, then, that Paley chose to use the same example (a termite colony) to illustrate this principle as we ourselves used to illustrate it in the first article in our series. Paley’s third principle of complexity is expressed thus: ‘Complexity explanations account for global order by specifying the local behaviour of units which have no awareness of the order thereby being produced, and which have no intention to produce it’. If, as he claims, we ‘failed to recognize’ this, why did we say in our first article ‘Order, innovation, and progress can emerge naturally from the interactions within a complex system; they do not need to be imposed centrally or from outside. For example, termite colonies construct the highest structures on the planet relative to the size of the builders. Yet there is no chief executive termite, no architect termite, and no blueprint. Each individual termite acts locally, seemingly following only a few simple shared rules of behaviour, within a context of other termites also acting locally. The termite mound emerges from a process of self organization’? Paley suggests that in our article on complexity, leadership and management, we misinterpreted the notion of the self-organizing system to mean that bottom-up approaches to organizational strategy and development should replace top-down ones. Had this been true, it would have been a grievous fault. What Plsek and Wilson actually said was ‘Complexity based organizational thinking suggests that goals and resources are established with a view towards the whole system, rather than artificially allocating them to parts of the system’. The article, written at a time when National Service Frameworks and other rigid, nationally imposed performance targets were stifling local initiative and flexibility, was arguing that tight central control can be counterproductive, not that organizations work better with nobody in charge. The analysis of an argument is incomplete without a consideration of the audience for which it was Trisha Greenhalgh MD, Professor of Primary Health Care, University College London, 206 Holborn Union Building, Highgate Hill, London N19 5LW, UK; Paul Plsek MSc, Independent consultant, Paul E. Plsek & Associates, Inc., Atlanta, USA; Tim Wilson FRCGP, Partner, PricewaterhouseCoopers, London, UK; Sarah Fraser DProf, Director, Sarah Fraser & Associates Ltd, Aylesbury, UK; Tim Holt FRCGP, Clinical Lecturer, University of Warwick, Coventry, UK.