Comparison of diagnosis-related groups (DRG)-based hospital payment system design and implementation strategies in different countries: The case of ischemic stroke

Diagnosis-related groups (DRG) based hospital payment systems are gradually becoming the main mechanism for reimbursement of acute inpatient care. We reviewed the existing literature to ascertain the global use of DRG-based hospital payment systems, compared the similarities and differences of original DRG versions in ten countries, and used ischemic stroke as an example to ascertain the design and implementation strategies for various DRG systems. The current challenges with and direction for the development of DRG-based hospital payment systems are also analyzed. We found that the DRG systems vary greatly in countries in terms of their purpose, grouping, coding, and payment mechanisms although based on the same classification concept and that they have tended to develop differently in countries with different income classifications. In high-income countries, DRG-based hospital payment systems have gradually begun to weaken as a mainstream payment method, while in middle-income countries DRG-based hospital payment systems have attracted increasing attention and increased use. The example of ischemic stroke provides suggestions for mutual promotion of DRG-based hospital payment systems and disease management. How to determine the level of DRG payment incentives and improve system flexibility, balance payment goals and disease management goals, and integrate development with other payment methods are areas for future research on DRG-based hospital payment systems.

Predictive model for early functional outcomes following acute care after traumatic brain injuries: A machine learning-based development and validation study

INTRODUCTION

Few studies on early functional outcomes following acute care after traumatic brain injury (TBI) are available. The aim of this study was to develop and validate a predictive model for functional outcomes at discharge for TBI patients using machine learning methods.

PATIENTS AND METHODS

In this retrospective study, data from 5281 TBI patients admitted for acute care who were identified in the Beijing hospital discharge abstract database were analysed. Data from 4181 patients in 52 tertiary hospitals were used for model derivation and internal validation. Data from 1100 patients in 21 secondary hospitals were used for external validation. A poor outcome was defined as a Barthel Index (BI) score ≤ 60 at discharge. Logistic regression, XGBoost, random forest, decision tree, and back propagation neural network models were used to fit classification models. Performance was evaluated by the area under the receiver operating characteristic curve (AUC), the area under the precision-recall curve (AP), calibration plots, sensitivity/recall, specificity, positive predictive value (PPV)/precision, negative predictive value (NPV) and F1-score.

RESULTS

Compared to the other models, the random forest model demonstrated superior performance in internal validation (AUC of 0.856, AP of 0.786, and F1-score of 0.724) and external validation (AUC of 0.779, AP of 0.630, and F1-score of 0.604). The sensitivity/recall, specificity, PPV/precision, and NPV of the model were 71.8%, 69.2%, 52.2%, and 84.0%, respectively, in external validation. The BI score at admission, age, use of nonsurgical treatment, neurosurgery status, and modified Charlson Comorbidity Index were identified as the top 5 predictors for functional outcome at discharge.

CONCLUSIONS

We established a random forest model that performed well in predicting early functional outcomes following acute care after TBI. The model has utility for informing decision-making regarding patient management and discharge planning and for facilitating health care quality assessment and resource allocation for TBI treatment.

Effect of early mobilization in patients with stroke and severe disturbance of consciousness: Retrospective study

OBJECTIVES

This study aimed to investigate the effectiveness and safety of early mobilization with a physiatrist and registered therapist Operating rehabilitation (PROr) for patients with stroke and severe disturbance of consciousness (DoC).

MATERIALS AND METHODS

We retrospectively screened records from patients with stroke admitted to our hospital from January 2015 to June 2021. Eligible patients with severe DoC were classified into two groups: patients who received standard rehabilitation (control group) and patients who received PROr (PROr group). We studied longitudinal change in the level of consciousness using the Japan Coma Scale (JCS) during hospital stay and compared in-hospital mortality, the incidence of respiratory complication, and modified Rankin Scale of discharge between the two groups.

RESULTS

Among the 2191 patients screened for inclusion, 16 patients were included in the PROr group, and 12 patients were included in the control group. Early mobilization was more promoted in the PROr group compared to the control group, but there were no significant differences in in-hospital mortality, the incidence of respiratory complication, or modified Rankin Scale at discharge between the two groups. In patients who survived during their hospital stay, JCS scores 2 weeks after the onset of stroke and JCS scores at discharge significantly improved from the start of rehabilitation in the PROr group, but not in the control group.

CONCLUSIONS

Early mobilization provided with the PROr program appears to be a safe treatment and may contribute to the improvement of consciousness level for patients with acute stroke and severe DoC.

Combined Functional Assessment for Predicting Clinical Outcomes in Stroke Patients After Post-acute Care: A Retrospective Multi-Center Cohort in Central Taiwan

Background and Objective

In 2014, Taiwan’s National Health Insurance administration launched a post-acute care (PAC) program for patients to improve their functions after acute stroke. The present study was aimed to determine PAC assessment parameters, either alone or in combination, for predicting clinical outcomes.

Methods

We retrospectively enrolled stroke adult patients through one PAC network in central Taiwan between January 2014 and December 2020. We collected data on post-stroke patients’ functional ability at baseline and after PAC stay. The comprehensive assessment included the following: Modified Rankin Scale (MRS), Functional Oral Intake Scale (FOIS), Mini-Nutritional Assessment (MNA), Berg Balance Scale (BBS), Fugl-Meyer Assessment (FMA), Mini-Mental State Examination (MMSE), aphasia test, and quality of life. The above items were assessed first at baseline and again at discharge from PAC. Logistic regression was used to determine factors that were associated with PAC length of stay (LOS), 14-day hospital readmission, and 1-year mortality.

Results

A total of 267 adults (mean age 67.2 ± 14.7 years) with completed data were analyzed. MRS, activities of daily living (ADLs), instrumental activities of daily living (IADLs), BBS, and MMSE all had improved between disease onset and PAC discharge. Higher baseline and greater improvement of physical and cognitive functions between initial and final PAC assessments were significantly associated with less readmission, and lower mortality. Furthermore, the improved ADLs, FOIS, MNA, FMA-motor, and MMSE scores were related to LOS during PAC. Using logistic regression, we found that functional improvements ≥5 items [adjusted odds ratio (aOR) = 0.16; 95% confidence interval (CI) = 0.05–0.45] and improved MMSE (aOR = 0.19; 95% CI = 0.05–0.68) were significantly associated with reduced post-PAC mortality or readmission. Whereas, functional improvements ≥7 items, improved FOIS, and MNA significantly prolonged LOS during PAC.

Conclusion

Physical performance parameters of patients with acute stroke improved after PAC. PAC assessment with multiple parameters better predicted clinical outcomes. These parameters could provide information on rehabilitation therapy for acute stroke patients receiving PAC.

Predictive Model and Mortality Risk Score during Admission for Ischaemic Stroke with Conservative Treatment

Background: Stroke is the second cause of mortality worldwide and the first in women. The aim of this study is to develop a predictive model to estimate the risk of mortality in the admission of patients who have not received reperfusion treatment. Methods: A retrospective cohort study was conducted of a clinical–administrative database, reflecting all cases of non-reperfused ischaemic stroke admitted to Spanish hospitals during the period 2008–2012. A predictive model based on logistic regression was developed on a training cohort and later validated by the “hold-out” method. Complementary machine learning techniques were also explored. Results: The resulting model had the following nine variables, all readily obtainable during initial care. Age (OR 1.069), female sex (OR 1.202), readmission (OR 2.008), hypertension (OR 0.726), diabetes (OR 1.105), atrial fibrillation (OR 1.537), dyslipidaemia (0.638), heart failure (OR 1.518) and neurological symptoms suggestive of posterior fossa involvement (OR 2.639). The predictability was moderate (AUC 0.742, 95% CI: 0.737–0.747), with good visual calibration; Pearson’s chi-square test revealed non-significant calibration. An easily consulted risk score was prepared. Conclusions: It is possible to create a predictive model of mortality for patients with ischaemic stroke from which important advances can be made towards optimising the quality and efficiency of care. The model results are available within a few minutes of admission and would provide a valuable complementary resource for the neurologist.

Predictors of death within 6 months of stroke onset: A model with Barthel index, platelet/lymphocyte ratio and serum albumin

AIMS

To develop and internally validate a nomogram to predict the risk of death within 6 months of onset of stroke in Chinese. Identifying risk factors with potentially direct effects on the nomogram will improve the quality of risk assessment and help nurses implement preventive measures based on patient-specific risk factors.

DESIGN

A retrospective study.

METHODS

We performed a least absolute shrinkage and selection operator (LASSO) regression modelling and multivariate logistic regression analysis to establish a prediction model of death risk in stroke patients within 6 months of onset. LASSO and time-dependent Cox regression models were further used to analyse the 6-month survival of stroke patients. Data were collected from 21 October 2013-6 May 2019.

RESULTS

The independent predictors of the nomogram were Barthel index (odds ratio (OR) = 0.980, 95% confidence interval (CI) = 0.961-0.998, p = .03), platelet/lymphocyte ratio (OR = 1.005, 95% CI = 1.000-1.010, p = .04) and serum albumin (OR = 0.854, 95% CI = 0.774-0.931, p < .01). This model showed good discrimination and consistency, and its discrimination evaluation C-statistic was 0.879 in the training set and 0.891 in the internal validation set. The DCA indicated that the nomogram had a higher overall net benefit over most of the threshold probability range. The time-dependent Cox regression model established the impact of the time effect of the age variable on survival time.

CONCLUSIONS

Our results identified three predictors of death within 6 months of stroke in Chinese. These predictors can be used as risk assessment indicators to help caregivers performing clinical nursing work, and in clinical practice, it is suggested that nurses should evaluate the self-care ability of stroke patients in detail. The constructed nomogram can help identify patients at high risk of death within 6 months, so that intervention can be performed as early as possible.

Use of Electronic Health Data for Disease Prediction: A Comprehensive Literature Review

Disease prediction has the potential to benefit stakeholders such as the government and health insurance companies. It can identify patients at risk of disease or health conditions. Clinicians can then take appropriate measures to avoid or minimize the risk and in turn, improve quality of care and avoid potential hospital admissions. Due to the recent advancement of tools and techniques for data analytics, disease risk prediction can leverage large amounts of semantic information, such as demographics, clinical diagnosis and measurements, health behaviours, laboratory results, prescriptions and care utilisation. In this regard, electronic health data can be a potential choice for developing disease prediction models. A significant number of such disease prediction models have been proposed in the literature over time utilizing large-scale electronic health databases, different methods, and healthcare variables. The goal of this comprehensive literature review was to discuss different risk prediction models that have been proposed based on electronic health data. Search terms were designed to find relevant research articles that utilized electronic health data to predict disease risks. Online scholarly databases were searched to retrieve results, which were then reviewed and compared in terms of the method used, disease type, and prediction accuracy. This paper provides a comprehensive review of the use of electronic health data for risk prediction models. A comparison of the results from different techniques for three frequently modelled diseases using electronic health data was also discussed in this study. In addition, the advantages and disadvantages of different risk prediction models, as well as their performance, were presented. Electronic health data have been widely used for disease prediction. A few modelling approaches show very high accuracy in predicting different diseases using such data. These modelling approaches have been used to inform the clinical decision process to achieve better outcomes.

Long-Term Survival after Stroke in 1.4 Million Japanese Population: Shiga Stroke and Heart Attack Registry

Background and Purpose

Although numerous measures for stroke exist, stroke remains one of the leading causes of death in Japan. In this study, we aimed to determine the long-term survival rate after first-ever stroke using data from a large-scale population-based stroke registry study in Japan.

Methods

Part of the Shiga Stroke and Heart Attack Registry, the Shiga Stroke Registry is an ongoing population-based registry study of stroke, which covers approximately 1.4 million residents of Shiga Prefecture in Japan. A total 1,880 patients with non-fatal first-ever stroke (among 29-day survivors after stroke onset) registered in 2011 were followed up until December 2016. Five-year cumulative survival rates were estimated using the Kaplan-Meier method, according to subtype of the index stroke. Cox proportional hazards models were used to assess predictors of subsequent all-cause death.

Results

During an average 4.3-year follow-up period, 677 patients died. The 5-year cumulative survival rate after non-fatal first-ever stroke was 65.9%. Heterogeneity was present in 5-year cumulative survival according to stroke subtype: lacunar infarction, 75.1%; large-artery infarction, 61.5%; cardioembolic infarction, 44.9%; intracerebral hemorrhage, 69.1%; and subarachnoid hemorrhage, 77.9%. Age, male sex, Japan Coma Scale score on admission, and modified Rankin Scale score before stroke onset were associated with increased mortality during the chronic phase of ischemic and hemorrhagic stroke.

Conclusions

In this study conducted in a real-world setting of Japan, the 5-year survival rate after non-fatal first-ever stroke remained low, particularly among patients with cardioembolic infarction and large-artery infarction in the present population-based stroke registry.

A Dynamic Model for Predicting Survival up to 1 Year After Ischemic Stroke

BACKGROUND

This study developed and validated a dynamic prediction model for survival after ischaemic stroke up to 1 year.

METHODS

Patients with stroke (n = 425) who participated in a sub-study (2002-2004) from the South London Stroke Register (SLSR) were selected for model derivation. The model was developed using the extended Cox model with time-dependent covariates. The two temporal validation cohorts from SLSR included 1735 (1995-2002) and 2155 patients (2004-2016). The discrimination, calibration and clinical utility of the model were assessed.

RESULTS

Six strong predictors were used in the model, namely, age, sex, stroke subtype, stroke severity and pre-stroke and post-stroke disabilities. The c-statistics was 0.822 at 1 year in the derivation cohort. The model had a fair performance with prognostic accuracies of 77%-83% in the validation 1 cohort and 70%-75% in the validation 2 cohort. A good calibration was observed in the derivation cohort.

CONCLUSION

The proposed model can accurately predict survival up to 1 year after ischaemic stroke.

Trends in hospital standardized mortality ratios for stroke in Japan between 2012 and 2016: a retrospective observational study

OBJECTIVE

Stroke is one of the leading causes of death and disability, and imposes a major healthcare burden. The aim of this study was to determine the characteristics of hospital standardized mortality ratios (HSMRs) for stroke in Japan for the year 2012-16 to describe the trend.

DESIGN

Retrospective observational study.

SETTING

Data from the Japanese administrative database.

PARTICIPANTS

All hospital admissions for stroke were identified from diagnostic procedures combination (DPC) database from 2012 to 2016.

MAIN OUTCOME MEASURES

HSMR was calculated using the actual number of in-hospital deaths and expected deaths. To obtain the expected death number, a logistic regression model was developed to get the coefficient with a number of explanatory variables. Predictive accuracy of the logistic models was assessed using c-index and calibration was evaluated using the Hosmer-Lemeshow test.

RESULTS

A total of 63 084 patients admitted for stroke from January 2012 to December 2016 were analyzed. HSMRs showed declining tendency over these 5 years, suggesting stroke-related mortality has been improving. While the HSMRs varied from year to year, a wide variation was also seen among the different hospitals in Japan. The proportion of hospitals with HSMR less than 100 increased from 41.0% in 2012 to 59.0% in 2016.

CONCLUSION

This study demonstrated that HSMR can be calculated using DPC data and found wide variation in HSMR of stroke among hospitals in Japan and enabled us to image the trend. By examining these trends, facilities, authorities and provinces can initiate designs that will ultimately lead to an upgraded healthcare delivery system.

Stroke ICU Patient Mortality Day Prediction

This article presents a study on development of methods for analysis of data reflecting the process of treatment of stroke inpatients to predict clinical outcomes at the emergency care unit. The aim of this work is to develop models for the creation of validated risk scales for early intravenous stroke with minimum number of parameters with maximum prognostic accuracy and possibility to calculate the time of “expected intravenous stroke mortality”. The study of experience in the development and use of medical information systems allows us to state the insufficient ability of existing models for adequate data analysis, weak formalization and lack of system approach in the collection of diagnostic data, insufficient personalization of diagnostic data on the factors determining early intravenous stroke mortality.

In our study we divided patients into 3 subgroups according to the time of death - up to 1 day, 1 to 3 days, and 4 to 10 days. Early mortality in each subgroup was associated with a number of demographic, clinical, and instrumental-laboratory characteristics based on the interpretation of the results of calculating the significance of predictors of binary classification models by machine learning methods from the Scikit-Learn library. The target classes in training were “mortality rate of 1 day”, “mortality rate of 1–3 days”, “mortality rate from 4 days”. AUC ROC of trained models reached 91% for the method of random forest. The results of interpretation of decision trees and calculation of significance of predictors of built-in methods of random forest coincide that can prove to correctness of calculations.

Predicting need for advanced illness or palliative care in a primary care population using electronic health record data

Timely outreach to individuals in an advanced stage of illness offers opportunities to exercise decision control over health care. Predictive models built using Electronic health record (EHR) data are being explored as a way to anticipate such need with enough lead time for patient engagement. Prior studies have focused on hospitalized patients, who typically have more data available for predicting care needs. It is unclear if prediction driven outreach is feasible in the primary care setting. In this study, we apply predictive modeling to the primary care population of a large, regional health system and systematically examine the impact of technical choices, such as requiring a minimum number of health care encounters (data density requirements) and aggregating diagnosis codes using Clinical Classifications Software (CCS) groupings to reduce dimensionality, on model performance in terms of discrimination and positive predictive value. We assembled a cohort of 349,667 primary care patients between 65 and 90 years of age who sought care from Sutter Health between July 1, 2011 and June 30, 2014, of whom 2.1% died during the study period. EHR data comprising demographics, encounters, orders, and diagnoses for each patient from a 12 month observation window prior to the point when a prediction is made were extracted. L1 regularized logistic regression and gradient boosted tree models were fit to training data and tuned by cross validation. Model performance in predicting one year mortality was assessed using held-out test patients. Our experiments systematically varied three factors: model type, diagnosis coding, and data density requirements. We found substantial, consistent benefit from using gradient boosting vs logistic regression (mean AUROC over all other technical choices of 84.8% vs 80.7% respectively). There was no benefit from aggregation of ICD codes into CCS code groups (mean AUROC over all other technical choices of 82.9% vs 82.6% respectively). Likewise increasing data density requirements did not affect discrimination (mean AUROC over other technical choices ranged from 82.5% to 83%). We also examine model performance as a function of lead time, which is the interval between death and when a prediction was made. In subgroup analysis by lead time, mean AUROC over all other choices ranged from 87.9% for patients who died within 0 to 3 months to 83.6% for those who died 9 to 12 months after prediction time.

Two-Year Survival After First-Ever Stroke in a General Population of 1.4 Million Japanese - Shiga Stroke Registry

BACKGROUND

Stroke is one of the leading causes of disability and mortality in Japan. The aim of the present analysis was to determine the non-acute survival rate after first-ever stroke using data from a large-scale population-based stroke registry in Japan. Methods and Results: Shiga Stroke Registry is an ongoing population-based registry of stroke, which covers approximately 1.4 million residents of Shiga Prefecture in central Japan. A total of 2,176 first-ever stroke patients, who were registered in 2011, were followed up until December 2013. The 2-year cumulative survival rates were estimated using Kaplan-Meier method according to index stroke subtype. Cox proportional hazards models were used to assess predictors of all-cause death. During a 2-year follow-up period, 663 patients (30.5%) died. The 2-year cumulative survival rate after first-ever stroke was 69.5%. There was heterogeneity in 2-year cumulative survival according to stroke subtype: lacunar infarction, 87.2%; large artery infarction, 76.1%; cardioembolic infarction, 55.4%; intracerebral hemorrhage, 65.9%; and subarachnoid hemorrhage, 56.7%. Older age, male sex, medical history, higher Japan coma scale score on admission, and stroke subtype were associated with risk of all-cause death in ≤2 years.

CONCLUSIONS

In the present population-based stroke registry with a real-world setting in Japan, 2-year cumulative mortality after first-ever stroke is still high (>30%), particularly for cardioembolic infarction, subarachnoid hemorrhage and intracerebral hemorrhage.

Risk prediction models for mortality in patients with cardiovascular disease: The BioBank Japan project

Background

Cardiovascular disease (CVD) is a leading cause of death in Japan. The present study aimed to develop new risk prediction models for long-term risks of all-cause and cardiovascular death in patients with chronic phase CVD.

Methods

Among the subjects registered in the BioBank Japan database, 15,058 patients aged ≥40 years with chronic ischemic CVD (ischemic stroke or myocardial infarction) were divided randomly into a derivation cohort (n = 10,039) and validation cohort (n = 5019). These subjects were followed up for 8.55 years in median. Risk prediction models for all-cause and cardiovascular death were developed using the derivation cohort by Cox proportional hazards regression. Their prediction performances for 5-year risk of mortality were evaluated in the validation cohort.

Results

During the follow-up, all-cause and cardiovascular death events were observed in 2962 and 962 patients from the derivation cohort and 1536 and 481 from the validation cohort, respectively. Risk prediction models for all-cause and cardiovascular death were developed from the derivation cohort using ten traditional cardiovascular risk factors, namely, age, sex, CVD subtype, hypertension, diabetes, total cholesterol, body mass index, current smoking, current drinking, and physical activity. These models demonstrated modest discrimination (c-statistics, 0.703 for all-cause death; 0.685 for cardiovascular death) and good calibration (Hosmer-Lemeshow χ²-test, P = 0.17 and 0.15, respectively) in the validation cohort.

Conclusions

We developed and validated risk prediction models of all-cause and cardiovascular death for patients with chronic ischemic CVD. These models would be useful for estimating the long-term risk of mortality in chronic phase CVD.

•

We developed risk prediction models for death after cardiovascular disease (CVD).

•

Performances of these models were validated in an independent cohort.

•

Our models may be used to estimate mortality risk in chronic CVD patients.

Stroke severity may predict causes of readmission within one year in patients with first ischemic stroke event

INTRODUCTION

Readmissions after stroke are costly. Risk assessment using information available upon admission could identify high-risk patients for potential interventions to reduce readmissions. Baseline stroke severity has been suspected to be a factor in readmission; however, the exact nature of the impact has not been adequately understood.

METHODS

Hospitalized adult patients with first-ever ischemic stroke were identified from a nationwide administrative database. Stroke severity was assessed using a validated claims-based stroke severity index. Cox proportional hazards models were used to investigate the relationship between stroke severity and first readmission within one year.

RESULTS

Of the 10,877 patients, 4295 (39.5%) were readmitted in one year. The cumulative risk of readmission was 34.1%, 44.7%, and 62.9% in patients with mild, moderate, and severe stroke, respectively. Patients with greater stroke severity had a significantly higher adjusted risk of first readmission for infection, metabolic disorders, neurological sequelae, and pulmonary diseases, whereas those with lesser stroke severity were prone to first readmission due to accidents. Stroke severity did not affect the risk of first readmission for recurrent stroke/transient ischemic attack, other cardiovascular events, malignancy, ulcers/upper gastrointestinal bleeding, kidney diseases, and others.

CONCLUSIONS

Stroke severity in patients with first-ever ischemic stroke not only predicts readmission but also relates to the cause of readmission. Our results might provide important information for tailoring discharge planning to prevent readmissions.

Validation of a novel claims-based stroke severity index in patients with intracerebral hemorrhage

BACKGROUND

Stroke severity is an important outcome predictor for intracerebral hemorrhage (ICH) but is typically unavailable in administrative claims data. We validated a claims-based stroke severity index (SSI) in patients with ICH in Taiwan.

METHODS

Consecutive ICH patients from hospital-based stroke registries were linked with a nationwide claims database. Stroke severity, assessed using the National Institutes of Health Stroke Scale (NIHSS), and functional outcomes, assessed using the modified Rankin Scale (mRS), were obtained from the registries. The SSI was calculated based on billing codes in each patient's claims. We assessed two types of criterion-related validity (concurrent validity and predictive validity) by correlating the SSI with the NIHSS and the mRS. Logistic regression models with or without stroke severity as a continuous covariate were fitted to predict mortality at 3, 6, and 12 months.

RESULTS

The concurrent validity of the SSI was established by its significant correlation with the admission NIHSS (r = 0.731; 95% confidence interval [CI], 0.705-0.755), and the predictive validity was verified by its significant correlations with the 3-month (r = 0.696; 95% CI, 0.665-0.724), 6-month (r = 0.685; 95% CI, 0.653-0.715) and 1-year (r = 0.664; 95% CI, 0.622-0.702) mRS. Mortality models with NIHSS had the highest area under the receiver operating characteristic curve, followed by models with SSI and models without any marker of stroke severity.

CONCLUSIONS

The SSI appears to be a valid proxy for the NIHSS and an effective adjustment for stroke severity in studies of ICH outcome with administrative claims data.

Ischemic stroke outcome: A review of the influence of post-stroke complications within the different scenarios of stroke care

Stroke remains one of the main causes of death and disability worldwide. The challenge of predicting stroke outcome has been traditionally assessed from a general point of view, where baseline non-modifiable factors such as age or stroke severity are considered the most relevant factors. However, after stroke occurrence, some specific complications such as hemorrhagic transformations or post stroke infections, which lead to a poor outcome, could be developed. An early prediction or identification of these circumstances, based on predictive models including clinical information, could be useful for physicians to individualize and improve stroke care. Furthermore, the addition of biological information such as blood biomarkers or genetic polymorphisms over these predictive models could improve their prognostic value. In this review, we focus on describing the different post-stroke complications that have an impact in short and long-term outcome across different time points in its natural history and on the clinical-biological information that might be useful in their prediction.

The use of national administrative data to describe the spatial distribution of in-hospital mortality following stroke in France, 2008-2011

Background

In the context of implementing the National Stroke Plan in France, a spatial approach was used to measure inequalities in this disease. Using the national PMSI-MCO databases, we analyzed the in-hospital prevalence of stroke and established a map of in-hospital mortality rates with regard to the socio-demographic structure of the country.

Methods

The principal characteristics of patients identified according to ICD10 codes relative to stroke (in accordance with earlier validation work) were studied. A map of standardized mortality rates at the level of PMSI geographic codes was established. An exploratory analysis (principal component analysis followed by ascending hierarchical classification) using INSEE socio-economic data and mortality rates was also carried out to identify different area profiles.

Results

Between 2008 and 2011, the number of stroke patients increased by 3.85 %, notably for ischemic stroke in the 36–55 years age group (60 % of men). Over the same period, in-hospital mortality fell, and the map of standardized rates illustrated the diagonal of high mortality extending from the north-east to the south-west of the country. The most severely affected areas were also those with the least favorable socio-professional indicators.

Conclusions

The PMSI-MCO database is a major source of data on the health status of the population. It can be used for the area-by-area observation of the performance of certain healthcare indicators, such as in-hospital mortality, or to follow the implementation of the National Stroke Plan. Our study showed the interplay between social and demographic factors and stroke-related in-hospital mortality. The map derived from the results of the exploratory analysis illustrated a variety of areas where social difficulties, aging and high mortality seemed to meet. The study raises questions about access to neuro-vascular care in isolated areas and in those in demographic decline. Telemedicine appears to be the solution favored by decision makers. The aging of the population managed for stroke must not mask the growing incidence in younger people, which raises questions about the development of classical (smoking, hypertension) or new (drug abuse) risk factors.

Procedure-based severity index for inpatients: development and validation using administrative database

Background

Risk adjustment is important in studies using administrative databases. Although utilization of diagnostic and therapeutic procedures can represent patient severity, the usability of procedure records in risk adjustment is not well-documented. Therefore, we aimed to develop and validate a severity index calculable from procedure records.

Methods

Using the Japanese nationwide Diagnosis Procedure Combination database of acute-care hospitals, we identified patients discharged between 1 April 2012 and 31 March 2013 with an admission-precipitating diagnosis of acute myocardial infarction, congestive heart failure, acute cerebrovascular disease, gastrointestinal hemorrhage, pneumonia, or septicemia. Subjects were randomly assigned to the derivation cohort or the validation cohort. In the derivation cohort, we used multivariable logistic regression analysis to identify procedures performed on admission day which were significantly associated with in-hospital death, and a point corresponding to regression coefficient was assigned to each procedure. An index was then calculated in the validation cohort as sum of points for performed procedures, and performance of mortality-predicting model using the index and other patient characteristics was evaluated.

Results

Of the 539 385 hospitalizations included, 270 054 and 269 331 were assigned to the derivation and validation cohorts, respectively. Nineteen significant procedures were identified from the derivation cohort with points ranging from −3 to 23, producing a severity index with possible range of −13 to 69. In the validation cohort, c-statistic of mortality-predicting model was 0.767 (95 % confidence interval: 0.764–0.770). The ω-statistic representing contribution of the index relative to other variables was 1.09 (95 % confidence interval: 1.03–1.17).

Conclusions

Procedure-based severity index predicted mortality well, suggesting that procedure records in administrative database are useful for risk adjustment.

Electronic supplementary material

The online version of this article (doi:10.1186/s12913-015-0889-x) contains supplementary material, which is available to authorized users.

The SOAR stroke score predicts hospital length of stay in acute stroke: an external validation study

AIMS

The objective of this study is to externally validate the SOAR stroke score (Stroke subtype, Oxfordshire Community Stroke Project Classification, Age and prestroke modified Rankin score) in predicting hospital length of stay (LOS) following an admission for acute stroke.

METHODS

We conducted a multi-centre observational study in eight National Health Service hospital trusts in the Anglia Stroke & Heart Clinical Network between September 2008 and April 2011. The usefulness of the SOAR stroke score in predicting hospital LOS in the acute settings was examined for all stroke and then stratified by discharge status (discharged alive or died during the admission).

RESULTS

A total of 3596 patients (mean age 77 years) with first-ever or recurrent stroke (92% ischaemic) were included. Increasing LOS was observed with increasing SOAR stroke score (p < 0.001 for both mean and median) and the SOAR stroke score of 0 had the shortest mean LOS (12 ± 20 days) while the SOAR stroke score of 6 had the longest mean LOS (26 ± 28 days). Among patients who were discharged alive, increasing SOAR stroke score had a significantly higher mean and median LOS (p < 0.001 for both mean and median) and the LOS peaked among patients with score value of 6 [mean (SD) 35 ± 31 days, median (IQR) 23 (14-48) days]. For patients who died as in-patient, there was no significant difference in mean or median LOS with increasing SOAR stroke score (p = 0.68 and p = 0.79, respectively).

CONCLUSION

This external validation study confirms the usefulness of the SOAR stroke score in predicting LOS in patients with acute stroke especially in those who are likely to survive to discharge. This provides a simple prognostic score useful for clinicians, patients and service providers.

Survival analyses of postoperative lung cancer patients: an investigation using Japanese administrative data

Long-term survival rates of cancer patients represent important information for policymakers and providers, but analyses from voluntary cancer registries in Japan may not reflect the overall situation. In 2003, the Diagnosis Procedure Combination Per-Diem Payment System (DPC/PDPS) for hospital reimbursement was introduced in Japan; more than half of Japan's acute care beds are currently covered under this system. Administrative data produced under the DPC system include claims data and clinical summaries for each admission. Due to the large amount of data spanning multiple institutions, this database may have applications in providing a more general and inclusive overview of healthcare. Here, we investigate the use of administrative data for analyses of long-term survival in cancer patients. We analyzed postoperative survival in 7,064 patients with primary non-small cell lung cancer admitted to 102 hospitals between April 2008 and March 2013 using DPC data. Survival was defined at the last date of examination or discharge within the study period, and the event was mortality during the same period. Overall survival rates for different cancer stages were calculated using the Kaplan-Meier method. Additionally, survival rates of cancer patients at clinical stage IA were compared between low- and high-volume hospitals using the Log-rank test. Postoperative 5-year survival for patients at stage IA was 85.8% (95% CI = 78.6%-93.0%). High-volume hospitals had higher survival rates than hospitals with lower volume. Our findings using large-scale administrative data were similar to previous clinical registry reports, showing potential applications as a new method in analyzing up-to-date healthcare information.

Off-hour presentation and outcomes in patients with acute ischemic stroke: a systematic review and meta-analysis

BACKGROUND

Studies have suggested that patients with acute ischemic stroke who present to the hospital during off-hours (weekends and nights) may or may not have worse clinical outcomes compared to patients who present during regular hours.

METHODS

We searched Medline In-Process & Other Non-Indexed Citations, MEDLINE, EMBASE, Cochrane Database of Systematic Reviews, and Scopus through August 2013, and included any study that evaluated the association between time of patient presentation to a healthcare facility and mortality or modified Rankin Scale in acute ischemic stroke. Quality of studies was assessed with the Newcastle-Ottawa Scale. A random-effect meta-analysis model was applied. Heterogeneity was assessed using the Q statistic and I(2). A priori subgroup analyses were used to explain observed heterogeneity.

RESULTS

A total of 21 cohort studies (23 cohorts) with fair quality enrolling 1,421,914 patients were included. Off-hour presentation for patients with acute ischemic stroke was associated with significantly higher short-term mortality (OR, 1.11, 95% CI 1.06-1.17). Presenting at accredited stroke centers (OR 1.04, 95% CI 0.98-1.11) and countries in North America (OR 1.05, 95% CI 1.01-1.09) were associated with smaller increase in mortality during off-hours. The results were not significantly different between adjusted (OR, 1.11, 95% CI 1.05-1.16) and unadjusted (OR, 1.13, 95% CI 0.95-1.35) outcomes. The proportion of patients with modified Rankin Scale at discharge ≥ 2-3 was higher in patients presenting during off-hours (OR, 1.14, 95% CI 1.06-1.22).

DISCUSSION

The evidence suggests that patients with acute ischemic stroke presenting during off-hours have higher short-term mortality and greater disability at discharge.