Explore the vast range of titles available.

Missing observational data pose an unavoidable problem in the hydrological field. Deep learning technology has recently been developing rapidly, and has started to be applied in the hydrological field. Being one of the network architectures used in deep learning, Long Short-Term Memory (LSTM) has been applied largely in related research, such as flood forecasting and discharge prediction, and the performance of an LSTM model has been compared with other deep learning models. Although the tuning of hyperparameters, which influences the performance of an LSTM model, is necessary, no sufficient knowledge has been obtained. In this study, we tuned the hyperparameters of an LSTM model to investigate the influence on the model performance, and tried to obtain a more suitable hyperparameter combination for the imputation of missing discharge data of the Daihachiga River. A traditional method, linear regression with an accuracy of 0.903 in Nash–Sutcliffe Efficiency (NSE), was chosen as the comparison target of the accuracy. The results of most of the trainings that used the discharge data of both neighboring and estimation points had better accuracy than the regression. Imputation of 7 days of the missing period had a minimum value of 0.904 in NSE, and 1 day of the missing period had a lower quartile of 0.922 in NSE. Dropout value indicated a negative correlation with the accuracy. Setting dropout as 0 had the best accuracy, 0.917 in the lower quartile of NSE. When the missing period was 1 day and the number of hidden layers were more than 100, all the compared results had an accuracy of 0.907–0.959 in NSE. Consequently, the case, which used discharge data with backtracked time considering the missing period of 1 day and 7 days and discharge data of adjacent points as input data, indicated better accuracy than other input data combinations. Moreover, the following information is obtained for this LSTM model: 100 hidden layers are better, and dropout and recurrent dropout levels equaling 0 are also better. The obtained optimal combination of hyperparameters exceeded the accuracy of the traditional method of regression analysis.

Improving the collection and quality of race and ethnicity reported in hospital data is a key step in identifying disparities in health service utilization and outcomes and opportunities for quality improvement. The objective of this study was to assess the quality of race/ethnicity reported in hospital discharge data and examine the impact on the identification of disparities in select health outcomes in New York City. Using the birth certificate as a gold standard, we examined the quality of hospital discharge race/ethnicity and estimated the impact of misclassification on racial/ethnic disparities in severe maternal morbidity and preventable hospitalizations. Delivery hospitalizations from the New York State hospital discharge data (Statewide Planning and Research Cooperative System) linked with 2015 New York City birth certificates. Sensitivity and positive predictive value (PPV). The non-Hispanic white and black race had relatively high sensitivity and PPV. Hispanic ethnicity and Asian race had moderate sensitivity and high PPV, but were often misclassified as \"Other.\" As a result, health disparities may be underestimated for those of Hispanic ethnicity and Asian race, particularly for indicators that use population denominators drawn from another source. The quality of hospital discharge data varies by race/ethnicity and may underestimate disparities in some groups. Future research should validate findings with other data sources, identify driving factors, and evaluate progress over time.

To estimate the accuracy and reliability of the reporting of diagnoses and procedures related to severe acute maternal morbidity in French hospital discharge data. The study, conducted in four French tertiary teaching hospitals, covered the years 2006 and 2007 and 30,607 deliveries. We identified severe maternal morbid events—eclampsia, pulmonary embolism, procedures related to postpartum hemorrhages, and intensive care—in administrative hospital discharge data and medical records and compared their recording. Information from medical records was the gold standard. Sensitivity, specificity, positive predictive value (PPV), and negative predictive value of the hospital discharge data for these events were calculated. False positives and false negatives were examined to identify the reasons for misrecorded information. The PPV of the hospital discharge data was 20% for eclampsia. For procedures related to postpartum hemorrhages, the PPVs were high, but sensitivities were lower; however, 95% of recording errors could be corrected. All indicators for intensive care exceeded 98%. Intensive care and procedures seem reliably reported in the hospital administrative database, which, therefore, can be used to monitor them. Using these data for monitoring diagnoses will require a greater investment by clinicians in the accuracy of their reporting.

Background In international comparisons, rates of amputations of the lower limb are relatively high in Germany. This study aims to analyze trends in lower limb amputations over time, as well as outcomes of care concerning in-hospital mortality and reamputation rates during the same hospital stay which might indicate the quality of surgical and perioperative health care processes. Methods This work is an observational population-based study using complete national hospital discharge data (Diagnosis-Related Group Statistics (DRG Statistics)) from 2005 to 2015. All inpatient cases with lower limb amputation were identified and stratified by eight amputation levels. Time trends of case numbers and in-hospital mortality were studied age-sex standardized. For inpatient cases with reamputation during the same hospital stay, first and last amputation levels were cross tabulated. Results A total of 55,595 amputations of the lower limb in 2015 (52,096 in 2005) were identified. After age-sex standardization to the demographic structure of 2005, a relative decrease of − 11.1% was revealed (men − 2.6%, women − 25.0%). The stratified analysis by amputation levels showed that the decreases were induced by higher amputation levels, whereas the amputation levels of toe/foot ray after standardization still showed a relative increase of + 12.8%. In-hospital mortality of all cases with lower limb amputation fell from 19.8% in 2005 to 17.4% in 2015 (SMR 0.89 [95% CI 0.86; 0.92]). The percentage of reamputations during the same hospital stay declined from 13.2 to 10.2%. Conclusions The number of lower limb amputations declined in Germany, however distinctly stronger in women than in men. The observed decreases of in-hospital mortality as well as of reamputation rates point to improvements in perioperative health care. Despite these indications of improvements, the distinct increase in case numbers at the level of toe/foot ray calls for additional targeted prevention efforts, especially for patients with diabetes.

Population-based stroke registries can provide valid stroke incidence because they ensure exhaustiveness of case ascertainment. However, their results are difficult to extrapolate because they cover a small population. The French Hospital Discharge Database (FHDDB), which routinely collects administrative data, could be a useful tool for providing data on the nationwide burden of stroke. The aim of our pilot study was to assess the validity of stroke diagnosis reported in the FHDDB. All records of patients with a diagnosis of stroke between 2004 and 2008 were retrieved from the FHDDB of Dijon Teaching Hospital. The Dijon Stroke Registry was considered as the gold standard. The sensitivity, positive predictive value (PPV), and weighted kappa were calculated. The Dijon Stroke Registry identified 811 patients with a stroke, among whom 186 were missed by the FHDDB and thus considered false-negatives. The FHDDB identified 903 patients discharged following a stroke including 625 true-positives confirmed by the registry and 278 false-positives. The overall sensitivity and PPV of the FHDDB for the diagnosis of stroke were, respectively, 77.1 % (95 % CI 74.2–80) and 69.2 % (95 % CI 66.1–72.2). For cardioembolic and lacunar strokes, the FHDDB yielded higher PPVs (respectively 86.7 and 84.6 %; p  < 0.0001) than those of other stroke subtypes. The PPV but not sensitivity significantly increased over the years ( p  < 0.0001). Agreement with the stroke registry was moderate (kappa 52.8; 95 % CI 46.8–58.9). The FHDDB-based stroke diagnosis showed moderate validity compared with the Dijon Stroke Registry as the gold standard. However, its accuracy (PPV) increased with time and was higher for some stroke subtypes.

Background: Canada has maintained a low incidence of toxigenic diphtheria since the 1990s, supported by continued commitment to publicly funded vaccination programs. Objective: To determine whether hospitalization data, complemented with notifiable disease data, can describe the toxigenic respiratory and cutaneous diphtheria burden in Canada, and to assess if Canada is meeting its diphtheria vaccine–preventable disease-reduction target of zero annual cases of locally transmitted respiratory diphtheria. Methods: Diphtheria-related hospital discharge data from 2006 to 2017 were extracted from the Discharge Abstract Database (DAD), and diphtheria case counts for the same period were retrieved from the Canadian Notifiable Disease Surveillance System (CNDSS), for descriptive analyses. As data from the province of Québec are not included in the DAD, CNDSS cases from Québec were excluded. Results: A total of 233 diphtheria-related hospitalizations were recorded in the DAD. Of these, diphtheria was the most responsible diagnosis in 23. Half the patients were male (52%), and 57% were 60 years and older. Central region (Ontario) accounted for the most discharge records (61%), followed by Prairie region (Alberta, Manitoba and Saskatchewan; 23%). Cutaneous diphtheria accounted for 43% of records, and respiratory diphtheria accounted for 3%, with the remainder being other diphtheria complications or site unspecified. Two records with diphtheria as the most responsible diagnosis resulted in inpatient deaths. Eighteen cases of diphtheria were reported through CNDSS. Cases occurred in all age groups, with the largest proportions among those aged 20 to 59 years (39%) and those aged 19 years and younger (33%). Cases were only reported in the Prairie (89%) and West Coast (British Columbia; 11%) regions. Conclusion: Hospital administrative data are consistent with the low incidence of diphtheria reported in CNDSS, and a low burden of respiratory diphtheria in Canada. Although Canada appears to be on track to meet its disease-reduction target, information on endemic transmission is not available.

Background The purpose of this study was to assess whether or not the change in coding classification had an impact on diagnosis and comorbidity coding in hospital discharge data across Canadian provinces. Methods This study examined eight years (fiscal years 1998 to 2005) of hospital records from the Hospital Person-Oriented Information database (HPOI) derived from the Canadian national Discharge Abstract Database. The average number of coded diagnoses per hospital visit was examined from 1998 to 2005 for provinces that switched from International Classifications of Disease 9 th version (ICD-9-CM) to ICD-10-CA during this period. The average numbers of type 2 and 3 diagnoses were also described. The prevalence of the Charlson comorbidities and distribution of the Charlson score one year before and one year after ICD-10 implementation for each of the 9 provinces was examined. The prevalence of at least one of the seventeen Charlson comorbidities one year before and one year after ICD-10 implementation were described by hospital characteristics (teaching/non-teaching, urban/rural, volume of patients). Results Nine Canadian provinces switched from ICD-9-CM to ICD-I0-CA over a 6 year period starting in 2001. The average number of diagnoses coded per hospital visit for all code types over the study period was 2.58. After implementation of ICD-10-CA a decrease in the number of diagnoses coded was found in four provinces whereas the number of diagnoses coded in the other five provinces remained similar. The prevalence of at least one of the seventeen Charlson conditions remained relatively stable after ICD-10 was implemented, as did the distribution of the Charlson score. When stratified by hospital characteristics, the prevalence of at least one Charlson condition decreased after ICD-10-CA implementation, particularly for low volume hospitals. Conclusion In conclusion, implementation of ICD-10-CA in Canadian provinces did not substantially change coding practices, but there was some coding variation in the average number of diagnoses per hospital visit across provinces.

Discharge instructions are a key form of documentation and patient communication in the time of transition from the emergency department (ED) to home. Discharge instructions are time-consuming and often underprioritized, especially in the ED, leading to discharge delays and possibly impersonal patient instructions. Generative artificial intelligence and large language models (LLMs) offer promising methods of creating high-quality and personalized discharge instructions; however, there exists a gap in understanding patient perspectives of LLM-generated discharge instructions. We aimed to assess the use of LLMs such as ChatGPT in synthesizing accurate and patient-accessible discharge instructions in the ED. We synthesized 5 unique, fictional ED encounters to emulate real ED encounters that included a diverse set of clinician history, physical notes, and nursing notes. These were passed to GPT-4 in Azure OpenAI Service (Microsoft) to generate LLM-generated discharge instructions. Standard discharge instructions were also generated for each of the 5 unique ED encounters. All GPT-generated and standard discharge instructions were then formatted into standardized after-visit summary documents. These after-visit summaries containing either GPT-generated or standard discharge instructions were randomly and blindly administered to Amazon MTurk respondents representing patient populations through Amazon MTurk Survey Distribution. Discharge instructions were assessed based on metrics of interpretability of significance, understandability, and satisfaction. Our findings revealed that survey respondents' perspectives regarding GPT-generated and standard discharge instructions were significantly (P=.01) more favorable toward GPT-generated return precautions, and all other sections were considered noninferior to standard discharge instructions. Of the 156 survey respondents, GPT-generated discharge instructions were assigned favorable ratings, \"agree\" and \"strongly agree,\" more frequently along the metric of interpretability of significance in discharge instruction subsections regarding diagnosis, procedures, treatment, post-ED medications or any changes to medications, and return precautions. Survey respondents found GPT-generated instructions to be more understandable when rating procedures, treatment, post-ED medications or medication changes, post-ED follow-up, and return precautions. Satisfaction with GPT-generated discharge instruction subsections was the most favorable in procedures, treatment, post-ED medications or medication changes, and return precautions. Wilcoxon rank-sum test of Likert responses revealed significant differences (P=.01) in the interpretability of significant return precautions in GPT-generated discharge instructions compared to standard discharge instructions but not for other evaluation metrics and discharge instruction subsections. This study demonstrates the potential for LLMs such as ChatGPT to act as a method of augmenting current documentation workflows in the ED to reduce the documentation burden of physicians. The ability of LLMs to provide tailored instructions for patients by improving readability and making instructions more applicable to patients could improve upon the methods of communication that currently exist.

ObjectivesTo explore the existence and strength of a relationship between hospital volume and mortality, to estimate minimum volume thresholds and to assess the potential benefit of centralisation of services.DesignObservational population-based study using complete German hospital discharge data (Diagnosis-Related Group Statistics (DRG Statistics)).SettingAll acute care hospitals in Germany.ParticipantsAll adult patients hospitalised for 1 out of 25 common or medically important types of inpatient treatment from 2009 to 2014.Main outcome measureRisk-adjusted inhospital mortality.ResultsLower inhospital mortality in association with higher hospital volume was observed in 20 out of the 25 studied types of treatment when volume was categorised in quintiles and persisted in 17 types of treatment when volume was analysed as a continuous variable. Such a relationship was found in some of the studied emergency conditions and low-risk procedures. It was more consistently present regarding complex surgical procedures. For example, about 22 000 patients receiving open repair of abdominal aortic aneurysm were analysed. In very high-volume hospitals, risk-adjusted mortality was 4.7% (95% CI 4.1 to 5.4) compared with 7.8% (7.1 to 8.7) in very low volume hospitals. Theminimum volume above which risk of death would fall below the average mortality was estimated as 18 cases per year. If all hospitals providing this service would perform at least 18 cases per year, one death among 104 (76 to 166) patients could potentially be prevented.ConclusionsBased on complete national hospital discharge data, the results confirmed volume–outcome relationships for many complex surgical procedures, as well as for some emergency conditions and low-risk procedures. Following these findings, the study identified areas where centralisation would provide a benefit for patients undergoing the specific type of treatment in German hospitals and quantified the possible impact of centralisation efforts.

ObjectiveTo describe and analyse hospitalisations for adverse drug reactions (ADRs) involving anticoagulants. We also analysed the progress of the reactions over time, the factors related with ADRs.DesignA retrospective, descriptive, epidemiological study.SettingThis study used the Spanish National Hospital Discharge Database (Conjunto Mínimo Básico de Datos, CMBD), over a 4-year period.ParticipantsWe selected CMBD data corresponding to hospital discharges with a diagnosis of ADRs to anticoagulants (International Classification of Diseases-Ninth Revision, Clinical Modification (ICD-9-CM) code E934.2) in any diagnostic field during the study period.Main outcome measuresWe calculated the annual incidence of ADRs to anticoagulants according to sex and age groups. The median lengths of hospital stay and in-hospital mortality (IHM) were also estimated for each year studied. Bivariate analyses of the changes in variables according to year were based on Poisson regression. IHM was analysed using logistic regression models. The estimates were expressed as ORs and their 95% CI.ResultsDuring the study period, 50 042 patients were hospitalised because of ADRs to anticoagulants (6.38% of all ADR-related admissions). The number of cases increased from 10 415 in 2010 to 13 891 in 2013. Cumulative incidence of ADRs to anticoagulants was significantly higher for men than women and in all age groups. An adjusted multivariate analysis revealed that IHM did not change significantly over time. We observed a statistically significant association between IHM and age, with the highest risk for the ≥85 age group (OR 2.67; 95% CI 2.44 to 2.93).ConclusionsThe incidence of ADRs to anticoagulants in Spain increased from 2010 to 2013, and was significantly higher for men than women and in all age groups. Older patients were particularly susceptible to being hospitalised with an adverse reaction to an anticoagulant.