Hospital discharge records under-report the prevalence of diabetes in inpatients

Higher mortality and longer length of stay in hospitalized patients with newly diagnosed diabetes

AIMS

We investigated the association between diabetes status at admission and in-hospital outcomes in all hospitalized patients, regardless of the reason for admission.

METHODS

All individuals aged 20 years or older who were admitted to Yongin Severance Hospital between March 2020 and February 2022 were included in study. Subjects were categorized into three groups: non-DM, known DM, and newly diagnosed DM, based on medical history, anti-diabetic medications use, and laboratory test. Hospitalization-related outcomes, including in-hospital mortality and length of hospital stay, were compared between groups.

RESULTS

33,166 participants were enrolled. At hospitalization, 6,572 (19.8 %) subjects were classified as known DM, and another 2,634 (7.9 %) subjects were classified as newly diagnosed DM. In-hospital mortality was highest in newly diagnosed DM (HR 1.89, 95% CI 1.58-2.26, p < 0.001) followed by known DM (HR 1.41, 95% CI 1.18-1.69, p < 0.001) compared to non-DM. Length of hospital stay was significantly longer in newly diagnosed DM (median [IQR] 9.0 [5.0-18.0],days) than known DM (median [IQR] 5.0 [3.0-10.0],days)(p < 0.001) and non-DM (median [IQR] 4.0 [2.0-7.0],days). After adjusting for multiple covariates, newly diagnosed diabetes was independently associated with increased in-hospital mortality (p < 0.001).

CONCLUSIONS

Diabetes status at admission was closely linked to hospitalization-related outcomes. Notably, individuals with newly diagnosed diabetes demonstrated a higher risk of in-hospital mortality and a prolonged length of hospital stay.

Influence of nutritional status on hospital length of stay in patients with type 2 diabetes

INTRODUCTION

In the hospitalized patient, Diabetes mellitus type 2 (DM2) may result in a worse nutritional status due to its pathophysiology and dietary treatment.

OBJECTIVES

The aim of this study was to know if a hospitalized diabetic patient has a worse nutritional status, and to establish the influence of DM2 on the hospital length of stay in patients with malnutrition.

MATERIAL AND METHODS

This was a transveral study from January 2014 to October 2016; 1017 patients were included who were assessed by the Endocrinology and Nutrition Department. The data collected included anthropometry, plasma albumin, delay in performing the nutrition interconsultation and hospital length of stay. Nutritional status was evaluated using the Mini Nutritional Assesment (MNA) questionnaire and the nutritional risk score (NRS).

RESULTS

24.4% of the patients were diabetic and 75.6% were not. Diabetic patients had a higher body mass index (BMI) [23.18 (20.78-25.99) kg/m² vs. 22.31 (19.79-25.30) kg/m2, P˂.01], a lower total score in the MNA questionnaire [16.5(13.12-19) points vs. 17(14-20) points, P˂.01], and a lower NRS score [83.09(77.72-91.12) points vs. 85.78(79.27-92.83) points, p=0.03]. According to the MNA and the NRS, diabetic patients had an increased risk of malnutrition (<17.5 points) [OR=1.39, IC95%(1.04-1.86), p=0.02]; and NRS (<85 points) [OR=1.65, IC 95% (1.07-2.54) p=0.02], respectively. When adjusted for age these significant results disappeared. Diabetes combined with malnutrition showed that diabetic patients with malnutrition (MNA˂17.5) spent longer in hospital [21(12-36) days vs. 17(9-30) days, P=.01].

CONCLUSIONS

Diabetic patients have a worse nutritional status than non-diabetic patients. Diabetic patients with a poor nutritional status spend a longer period in hospital.

Estimating disease prevalence from drug utilization data using the Random Forest algorithm

Background

Aggregated claims data on medication are often used as a proxy for the prevalence of diseases, especially chronic diseases. However, linkage between medication and diagnosis tend to be theory based and not very precise. Modelling disease probability at an individual level using individual level data may yield more accurate results.

Methods

Individual probabilities of having a certain chronic disease were estimated using the Random Forest (RF) algorithm. A training set was created from a general practitioners database of 276 723 cases that included diagnosis and claims data on medication. Model performance for 29 chronic diseases was evaluated using Receiver-Operator Curves, by measuring the Area Under the Curve (AUC).

Results

The diseases for which model performance was best were Parkinson’s disease (AUC = .89, 95% CI = .77–1.00), diabetes (AUC = .87, 95% CI = .85–.90), osteoporosis (AUC = .87, 95% CI = .81–.92) and heart failure (AUC = .81, 95% CI = .74–.88). Five other diseases had an AUC >.75: asthma, chronic enteritis, COPD, epilepsy and HIV/AIDS. For 16 of 17 diseases tested, the medication categories used in theory-based algorithms were also identified by our method, however the RF models included a broader range of medications as important predictors.

Conclusion

Data on medication use can be a useful predictor when estimating the prevalence of several chronic diseases. To improve the estimates, for a broader range of chronic diseases, research should use better training data, include more details concerning dosages and duration of prescriptions, and add related predictors like hospitalizations.

Type-2 diabetes mellitus in schizophrenia: Increased prevalence and major risk factor of excess mortality in a naturalistic 7-year follow-up

Objective

Physical co-morbidity including type 2 diabetes mellitus is more prevalent in patients with schizophrenia compared to the general population. However, there is little consistent evidence that co-morbidity with diabetes mellitus and/or other diseases leads to excess mortality in schizophrenia. Thus, we investigated whether co-morbidity with diabetes and other somatic diseases is increased in schizophrenics, and if these are equally or more relevant predictors of mortality in schizophrenia than in age- and gender-matched hospitalised controls.

Methods

During 2000–2007, 679 patients with schizophrenia were admitted to University Hospital Birmingham NHS Trust. Co-morbidities were compared with 88,778 age- and gender group-matched hospital controls. Predictors of mortality were identified using forward Cox regression models.

Results

The prevalence of type 2 diabetes mellitus was increased in schizophrenia compared to hospitalised controls (11.3% versus 6.3%). The initial prevalence of type 2 diabetes mellitus was significantly higher in the 100 later deceased schizophrenic patients (24.0%) than in those 579 surviving over 7 years (9.2%). Predictors of mortality in schizophrenia were found to be age (relative risk [RR] = 1.1/year), type 2 diabetes mellitus (RR = 2.2), pneumonia (RR = 2.7), heart failure (RR = 2.9) and chronic renal failure (RR = 3.2). The impact of diabetes mellitus on mortality was significantly higher in schizophrenia than in hospital controls (RR = 2.2 versus RR = 1.1). In agreement, deceased schizophrenics had significantly suffered more diabetes mellitus than deceased controls (24.0 versus 10.5%). The relative risks of mortality for other disorders and their prevalence in later deceased subjects did not significantly differ between schizophrenia and controls.

Conclusion

Schizophrenics have more and additionally suffer more from diabetes: co-morbidity with diabetes mellitus is increased in schizophrenia in comparison with hospital controls; type 2 diabetes mellitus causes significant excess mortality in schizophrenia. Thus, monitoring for and prevention of type 2 diabetes mellitus is of utmost relevance in hospitalised patients with schizophrenia.

Prevalence of diabetes mellitus in patients with home enteral nutrition

BACKGROUND

There are few data available in the literature on the prevalence of diabetes mellitus (DM) in patients with home enteral nutrition (HEN) via tube feeding. The objective was to analyze the prevalence of DM in patients receiving HEN, as well as evaluating the complications, the prescribed antidiabetic treatments and the nutrition regimen selected.

DESIGN

This was a retrospective, single-center, observational study reviewing clinical histories. The population consisted of patients over 18 years of age who started HEN by tube between January 2016 and January 2018. Sociodemographic variables were recorded, as well as variables related to HEN. Additional variables were recorded in patients with DM.

RESULTS

In the 198 study patients, followed up for a median of 104 days, the prevalence of DM was 31.8%, and patients with DM were older (71.3±11.5 vs. 64.2±15.8; p=0.002) than those without DM. There were no differences between patients with and without DM as regards the prescription of HEN, its route and form of administration, and its complications. One hundred and thirty-two patients (66.7%) died during follow-up. The presence of DM did not increase the risk of death during follow-up (after adjusting for age, gender, and diagnosis). More than 85% of patients with DM received a specific formula for diabetes, and 84.1% of these patients received drug treatment.

CONCLUSION

The prevalence of DM was high in patients receiving HEN, most of whom were prescribed specific enteral nutrition formulas. The presence of DM was not associated with greater morbidity and mortality or with differences in HEN regimens or indications.

A Population Dynamic Model to Assess the Diabetes Screening and Reporting Programs and Project the Burden of Undiagnosed Diabetes in Thailand

Diabetes mellitus (DM) is rising worldwide, exacerbated by aging populations. We estimated and predicted the diabetes burden and mortality due to undiagnosed diabetes together with screening program efficacy and reporting completeness in Thailand, in the context of demographic changes. An age and sex structured dynamic model including demographic and diagnostic processes was constructed. The model was validated using a Bayesian Markov Chain Monte Carlo (MCMC) approach. The prevalence of DM was predicted to increase from 6.5% (95% credible interval: 6.3-6.7%) in 2015 to 10.69% (10.4-11.0%) in 2035, with the largest increase (72%) among 60 years or older. Out of the total DM cases in 2015, the percentage of undiagnosed DM cases was 18.2% (17.4-18.9%), with males higher than females (p-value < 0.01). The highest group with undiagnosed DM was those aged less than 39 years old, 74.2% (73.7-74.7%). The mortality of undiagnosed DM was ten-fold greater than the mortality of those with diagnosed DM. The estimated coverage of diabetes positive screening programs was ten-fold greater for elderly compared to young. The positive screening rate among females was estimated to be significantly higher than those in males. Of the diagnoses, 87.4% (87.0-87.8%) were reported. Targeting screening programs and good reporting systems will be essential to reduce the burden of disease.

Influence of Hyperglycemia Associated with Enteral Nutrition on Mortality in Patients with Stroke

Objectives: To evaluate in patients admitted for stroke: (1) The frequency of hyperglycaemia associated with enteral nutrition (EN). (2) The risk of morbidity and mortality associated with the development of this type of hyperglycaemia.

METHODS

A longitudinal observational study was conducted in 115 non-diabetic patients admitted for stroke with EN. Age, functional capacity (Rankin scale), and blood plasma glucose (BPG) were recorded. Hyperglycaemia was considered as: a value higher than 126 mg/dL before the EN and/or a value higher than 150 mg/dL after a week of enteral nutrition. According to this, three groups were differentiated: HyperES: Those who had hyperglycemia before the beginning of the EN (33% patients); NoHyper: those who did not have hyperglycemia before or after (47.8% patients); and HyperEN: Those who did not have hyperglycemia before but suffered it after the beginning of the EN (19.1% patients).

RESULTS

The age was 72.72 (15.32) years. A higher rate of mortality was observed in the HyperEN group 45.50%, than HyperES 15.80% or NoHyper: 10.90%). A lower recovery of the oral feeding was observed in those patients of the HyperEN group 27.30%, than HyperES: 42.10% or NoHyper: 61.80%). In the multivariate analysis adjusting for age, sex, and Rankin scale the development of hyperglycemia in those who did not have it at the beginning (HyperEN) was an independent risk factor for non-recovery of the oral feeding (OR: 4.21 (1.20-14.79), p = 0.02); and mortality adjusted for age, sex and Rankin scale (OR: 6.83 (1.76-26.47), p < 0.01).

CONCLUSIONS

In non-diabetic patients admitted for stroke with EN, the development of hyperglycaemia in relation to enteral nutrition supposes an independent risk factor for mortality and for the non-recovery of the oral feeding.

Comparing external and internal validation methods in correcting outcome misclassification bias in logistic regression: A simulation study and application to the case of postsurgical venous thromboembolism following total hip and knee arthroplasty

PURPOSE

We assessed the validity of postsurgery venous thromboembolism (VTE) diagnoses identified from administrative databases and compared Bayesian and multiple imputation (MI) approaches in correcting for outcome misclassification in logistic regression models.

METHODS

Sensitivity and specificity of postsurgery VTE among patients undergoing total hip or knee replacement (THR/TKR) were assessed against chart review in six Montreal hospitals in 2009 to 2010. Administrative data on all THR/TKR Quebec patients in 2009 to 2010 were obtained. The performance of Bayesian external, Bayesian internal, and MI approaches to correct the odds ratio (OR) of postsurgery VTE in tertiary versus community hospitals was assessed using simulations. Bayesian external approach used prior information from external sources, while Bayesian internal and MI approaches used chart review.

RESULTS

In total, 17 319 patients were included, 2136 in participating hospitals, among whom 75 had VTE in administrative data versus 81 in chart review. VTE sensitivity was 0.59 (95% confidence interval, 0.48-0.69) and specificity was 0.99 (0.98-0.99), overall. The adjusted OR of VTE in tertiary versus community hospitals was 1.35 (1.12-1.64) using administrative data, 1.45 (0.97-2.19) when MI was used for misclassification correction, and 1.53 (0.83-2.87) and 1.57 (0.39-5.24) when Bayesian internal and external approaches were used, respectively. In simulations, all three approaches reduced the OR bias and had appropriate coverage for both nondifferential and differential misclassification.

CONCLUSION

VTE identified from administrative data had low sensitivity and high specificity. The Bayesian external approach was useful to reduce outcome misclassification bias in logistic regression; however, it required accurate specification of the misclassification properties and should be used with caution.

Verification of Data Accuracy in Japan Congenital Cardiovascular Surgery Database Including Its Postprocedural Complication Reports

Background:

Japan Congenital Cardiovascluar Surgical Database (JCCVSD) is a nationwide registry whose data are used for health quality assessment and clinical research in Japan. We evaluated the completeness of case registration and the accuracy of recorded data components including postprocedural mortality and complications in the database via on-site data adjudication.

Methods:

We validated the records from JCCVSD 2010 to 2012 containing congenital cardiovascular surgery data performed in 111 facilities throughout Japan. We randomly chose nine facilities for site visit by the auditor team and conducted on-site data adjudication. We assessed whether the records in JCCVSD matched the data in the source materials.

Results:

We identified 1,928 cases of eligible surgeries performed at the facilities, of which 1,910 were registered (99.1% completeness), with 6 cases of duplication and 1 inappropriate case registration. Data components including gender, age, and surgery time (hours) were highly accurate with 98% to 100% concordance. Mortality at discharge and at 30 and 90 postoperative days was 100% accurate. Among the five complications studied, reoperation was the most frequently observed, with 16 and 21 cases recorded in the database and source materials, respectively, having a sensitivity of 0.67 and a specificity of 0.99.

Conclusions:

Validation of JCCVSD database showed high registration completeness and high accuracy especially in the categorical data components. Adjudicated mortality was 100% accurate. While limited in numbers, the recorded cases of postoperative complications all had high specificities but had lower sensitivity (0.67-1.00). Continued activities for data quality improvement and assessment are necessary for optimizing the utility of these registries.

Impact of bariatric surgery on outcomes of patients with nonalcoholic fatty liver disease: a nationwide inpatient sample analysis, 2004-2012

BACKGROUND

Bariatric surgery in eligible morbidly obese individuals may improve liver steatosis, inflammation, and fibrosis; however, population-based data on the clinical benefits of bariatric surgery in patients with nonalcoholic fatty liver disease (NAFLD) are lacking.

OBJECTIVES

To assess the relationship between bariatric surgery and clinical outcomes in hospitalized patients with NAFLD.

SETTING

United States inpatient care database.

METHODS

The Nationwide Inpatient Sample database was queried from 2004 to 2012 with co-diagnoses of NAFLD and morbid obesity. Hospitalizations with a history of prior bariatric surgery (Roux-en-Y gastric bypass, gastric band, and sleeve gastrectomy) were also identified. The primary outcome was in-hospital mortality. Secondary outcomes included cirrhosis, myocardial infarction, stroke, and renal failure. Poisson regression was used to derive adjusted incidence risk ratios for clinical outcomes in patients with prior bariatric surgery compared with those without bariatric surgery.

RESULTS

Among 45,462 patients with a discharge diagnosis of NAFLD and morbid obesity, 18,618 patients (41.0%) had prior bariatric surgery. There was a downward trend in bariatric surgery procedures (percent annual change of -5.94% from 2004 to 2012). In a multivariable analysis, prior bariatric surgery was associated with decreased inpatient mortality compared with no bariatric surgery (incidence risk ratios = .08; 95% confidence interval, .03-.20, P<.001). Prior bariatric surgery was also associated with decreased incidence risk ratios for cirrhosis, myocardial infarction, stroke, and renal failure (all P<.001).

CONCLUSIONS

Prior bariatric surgery is associated with decreased in-hospital morbidity and mortality in morbidly obese NAFLD patients. Despite this, the proportion of NAFLD patients with bariatric surgery has declined from 2004 to 2012.

Correcting hazard ratio estimates for outcome misclassification using multiple imputation with internal validation data

OBJECTIVE

Outcome misclassification may occur in observational studies using administrative databases. We evaluated a two-step multiple imputation approach based on complementary internal validation data obtained from two subsamples of study participants to reduce bias in hazard ratio (HR) estimates in Cox regressions.

METHODS

We illustrated this approach using data from a surveyed sample of 6247 individuals in a study of statin-diabetes association in Quebec. We corrected diabetes status and onset assessed from health administrative data against self-reported diabetes and/or elevated fasting blood glucose (FBG) assessed in subsamples. The association between statin use and new onset diabetes was evaluated using administrative data and the corrected data. By simulation, we assessed the performance of this method varying the true HR, sensitivity, specificity, and the size of validation subsamples.

RESULTS

The adjusted HR of new onset diabetes among statin users versus non-users was 1.61 (95% confidence interval: 1.09-2.38) using administrative data only, 1.49 (0.95-2.34) when diabetes status and onset were corrected based on self-report and undiagnosed diabetes (FBG ≥ 7 mmol/L), and 1.36 (0.92-2.01) when corrected for self-report and undiagnosed diabetes/impaired FBG (≥ 6 mmol/L). In simulations, the multiple imputation approach yielded less biased HR estimates and appropriate coverage for both non-differential and differential misclassification. Large variations in the corrected HR estimates were observed using validation subsamples with low participation proportion. The bias correction was sometimes outweighed by the uncertainty introduced by the unknown time of event occurrence.

CONCLUSION

Multiple imputation is useful to correct for outcome misclassification in time-to-event analyses if complementary validation data are available from subsamples. Copyright © 2017 John Wiley & Sons, Ltd.

Evidence-based recommendations and expert consensus on enteral nutrition in the adult patient with diabetes mellitus or hyperglycemia

OBJECTIVE

The aim of this study was to develop evidence-based recommendations for glycemic control of patients with diabetes mellitus or stress hyperglycemia who are receiving enteral nutrition (EN).

METHODS

A Delphi survey method using Grading Recommendations Assessment, Development and Evaluation criteria was utilized for evaluation of suitable studies.

RESULTS

In patients with diabetes or stress hyperglycemia who were on EN support, the following results were found: CONCLUSIONS: These recommendations and suggestions regarding enteral feeding in patients with diabetes and hyperglycemia have direct clinical applicability.

Prevalence of diabetes, prediabetes, and stress hyperglycemia: insulin therapy and metabolic control in patients on total parenteral nutrition (prospective multicenter study)

OBJECTIVE

The prevalence of carbohydrate metabolism disorders in patients who receive total parenteral nutrition (TPN) is not well known. These disorders can affect the treatment, metabolic control, and prognosis of affected patients. The aims of this study were to determine the prevalence in noncritically ill patients on TPN of diabetes, prediabetes, and stress hyperglycemia; the factors affecting hyperglycemia during TPN; and the insulin therapy provided and the metabolic control achieved.

METHODS

We undertook a prospective multicenter study involving 19 Spanish hospitals. Noncritically ill patients who were prescribed TPN were included, and data were collected on demographic, clinical, and laboratory variables (glycated hemoglobin, C-reactive protein [CRP], capillary blood glucose) as well as insulin treatment.

RESULTS

The study included 605 patients. Before initiation of TPN, the prevalence of known diabetes was 17.4%, unknown diabetes 4.3%, stress hyperglycemia 7.1%, and prediabetes 27.8%. During TPN therapy, 50.9% of patients had at least one capillary blood glucose of >180 mg/dL. Predisposing factors were age, levels of CRP and glycated hemoglobin, the presence of diabetes, infectious complications, the number of grams of carbohydrates infused, and the administration of glucose-elevating drugs. Most (71.6%) patients were treated with insulin. The mean capillary blood glucose levels during TPN were: known diabetes (178.6 ± 46.5 mg/dL), unknown diabetes (173.9 ± 51.9), prediabetes (136.0 ± 25.4), stress hyperglycemia (146.0 ± 29.3), and normal (123.2 ± 19.9) (P<.001).

CONCLUSION

The prevalence of carbohydrate metabolism disorders is very high in noncritically ill patients on TPN. These disorders affect insulin treatment and the degree of metabolic control achieved.

Systematic review of validated case definitions for diabetes in ICD-9-coded and ICD-10-coded data in adult populations

OBJECTIVES

With steady increases in 'big data' and data analytics over the past two decades, administrative health databases have become more accessible and are now used regularly for diabetes surveillance. The objective of this study is to systematically review validated International Classification of Diseases (ICD)-based case definitions for diabetes in the adult population.

SETTING, PARTICIPANTS AND OUTCOME MEASURES

Electronic databases, MEDLINE and Embase, were searched for validation studies where an administrative case definition (using ICD codes) for diabetes in adults was validated against a reference and statistical measures of the performance reported.

RESULTS

The search yielded 2895 abstracts, and of the 193 potentially relevant studies, 16 met criteria. Diabetes definition for adults varied by data source, including physician claims (sensitivity ranged from 26.9% to 97%, specificity ranged from 94.3% to 99.4%, positive predictive value (PPV) ranged from 71.4% to 96.2%, negative predictive value (NPV) ranged from 95% to 99.6% and κ ranged from 0.8 to 0.9), hospital discharge data (sensitivity ranged from 59.1% to 92.6%, specificity ranged from 95.5% to 99%, PPV ranged from 62.5% to 96%, NPV ranged from 90.8% to 99% and κ ranged from 0.6 to 0.9) and a combination of both (sensitivity ranged from 57% to 95.6%, specificity ranged from 88% to 98.5%, PPV ranged from 54% to 80%, NPV ranged from 98% to 99.6% and κ ranged from 0.7 to 0.8).

CONCLUSIONS

Overall, administrative health databases are useful for undertaking diabetes surveillance, but an awareness of the variation in performance being affected by case definition is essential. The performance characteristics of these case definitions depend on the variations in the definition of primary diagnosis in ICD-coded discharge data and/or the methodology adopted by the healthcare facility to extract information from patient records.

Impact of screening and monitoring of capillary blood glucose in the detection of hyperglycemia and hypoglycemia in non-critical inpatients

OBJECTIVE

To evaluate the impact of screening hyper and hypoglycemia measured by capillary glycemia and standard monitorization of hyperglycemic patients hospitalized in regular care units of Hospital Israelita Albert Einstein.

METHODS

The capillary glycemia was measured by the Precision PCx (Abbott) glucosimeter, using the PrecisionWeb (Abbott) software. The detection of hyper and hypoglycemia during the months of May/June were compared to those of March/April in 2009 and to the frequency of the diagnosis of diabetes in 2007.

RESULTS

There was an increase in the glycemia screening from 27.7 to 77.5% of hospitalized patients (p < 0.001), of hyperglycemia detection (from 9.3 to 12.2%; p < 0.001) and of hypoglycemia (from 1.5 to 3.3%; p < 0.001) during the months of May/June 2009. According to this action 14 patients for each additional case of hyperglycemia and 26 cases for each case of hypoglycemia were identified. The detection of hyperglycemia was significantly higher (p < 0.001) than the frequency of registered diagnosis related do diabetes in the year of 2007.

CONCLUSIONS

the adoption of an institutional program of glycemia monitorization improves the detection of hyper and hypoglycemia and glycemia control in hospitalized patients in regular care units.

The Burden of Diagnosed and Undiagnosed Diabetes in Native Hawaiian and Asian American Hospitalized Patients

•

38% of Native Hawaiian, 34% of Japanese, 33% of Filipino, 23% of Chinese and 19% of White inpatients had diagnosed diabetes.

•

3% Native Hawaiian, 3% Japanese, 4% Filipino, 4% Chinese and 4% White inpatients had potentially undiagnosed diabetes.

•

Few potentially undiagnosed diabetes cases were found. Prospective methods are recommended.

•

Potentially undiagnosed diabetes was associated with a longer hospital stay and higher mortality.

•

Inpatient diabetes screening with HbA1c is suggested, which should be evaluated prospectively for improved outcomes.

Aims

Little is known about diabetes in hospitalized Native Hawaiians and Asian Americans. We determined the burden of diabetes (both diagnosed and undiagnosed) among hospitalized Native Hawaiian, Asian (Filipino, Chinese, Japanese), and White patients.

Methods

Diagnosed diabetes was determined from discharge data from a major medical center in Hawai‘i during 2007–2008. Potentially undiagnosed diabetes was determined by Hemoglobin A1c ≥ 6.5% or glucose ≥ 200 mg/dl values for those without diagnosed diabetes. Multivariable log-binomial models predicted diabetes (potentially undiagnosed and diagnosed, separately) controlling for socio-demographic factors.

Results

Of 17,828 hospitalized patients, 3.4% had potentially undiagnosed diabetes and 30.5% had diagnosed diabetes. In multivariable models compared to Whites, Native Hawaiian and all Asian subgroups had significantly higher percentages of diagnosed diabetes, but not of potentially undiagnosed diabetes. Potentially undiagnosed diabetes was associated with significantly more hospitalizations during the study period compared to both those without diabetes and those with diagnosed diabetes. In all racial/ethnic groups, those with potentially undiagnosed diabetes also had the longest length of stay and were more likely to die during the hospitalization.

Conclusions

Hospitalized Native Hawaiians (41%) and Asian subgroups had significantly higher overall diabetes burdens compared to Whites (23%). Potentially undiagnosed diabetes was associated with poor outcomes. Hospitalized patients, irrespective of race/ethnicity, may require more effective inpatient identification and management of previously undiagnosed diabetes to improve clinical outcomes.

[Predictors of mean blood glucose control and its variability in diabetic hospitalized patients]

INTRODUCTION

This study was intended to assess the effectiveness and predictors factors of inpatient blood glucose control in diabetic patients admitted to medical departments.

MATERIAL AND METHODS

A retrospective, analytical cohort study was conducted on patients discharged from internal medicine with a diagnosis related to diabetes. Variables collected included demographic characteristics, clinical data and laboratory parameters related to blood glucose control (HbA1c, basal plasma glucose, point-of-care capillary glucose). The cumulative probability of receiving scheduled insulin regimens was evaluated using Kaplan-Meier analysis. Multivariate regression models were used to select predictors of mean inpatient glucose (MHG) and glucose variability (standard deviation [GV]).

RESULTS

The study sample consisted of 228 patients (mean age 78.4 (SD 10.1) years, 51% women). Of these, 96 patients (42.1%) were treated with sliding-scale regular insulin only. Median time to start of scheduled insulin therapy was 4 (95% CI, 2-6) days. Blood glucose control measures were: MIG 181.4 (SD 41.7) mg/dL, GV 56.3 (SD 22.6). The best model to predict MIG (R(2): .376; P<.0001) included HbA1c (b=4.96; P=.011), baseline plasma glucose (b=.056; P=.084), mean capillary blood glucose in the first 24hours (b=.154; P<.0001), home treatment (versus oral agents) with basal insulin only (b=13.1; P=.016) or more complex (pre-mixed insulin or basal-bolus) regimens (b=19.1; P=.004), corticoid therapy (b=14.9; P=.002), and fasting on admission (b=10.4; P=.098).

CONCLUSION

Predictors of inpatient blood glucose control which should be considered in the design of DM management protocols include home treatment, HbA1c, basal plasma glucose, mean blood glucose in the first 24hours, fasting, and corticoid therapy.

Physical comorbidity and its relevance on mortality in schizophrenia: a naturalistic 12-year follow-up in general hospital admissions

Schizophrenia is a major psychotic disorder with significant comorbidity and mortality. Patients with schizophrenia are said to suffer more type-2 diabetes mellitus (T2DM) and diabetogenic complications. However, there is little consistent evidence that comorbidity with physical diseases leads to excess mortality in schizophrenic patients. Consequently, we investigated whether the burden of physical comorbidity and its relevance on hospital mortality differed between patients with and without schizophrenia in a 12-year follow-up in general hospital admissions. During 1 January 2000 and 31 June 2012, 1418 adult patients with schizophrenia were admitted to three General Manchester NHS Hospitals. All comorbid diseases with a prevalemce ≥1% were compared with those of 14,180 age- and gender-matched hospital controls. Risk factors, i.e. comorbid diseases that were predictors for general hospital mortality were identified using multivariate logistic regression analyses. Compared with controls, schizophrenic patients had a higher proportion of emergency admissions (69.8 vs. 43.0%), an extended average length of stay at index hospitalization (8.1 vs. 3.4 days), a higher number of hospital admissions (11.5 vs. 6.3), a shorter length of survival (1895 vs. 2161 days), and a nearly twofold increased mortality rate (18.0 vs. 9.7%). Schizophrenic patients suffered more depression, T2DM, alcohol abuse, asthma, COPD, and twenty-three more diseases, many of them diabetic-related complications or other environmentally influenced conditions. In contrast, hypertension, cataract, angina, and hyperlipidaemia were less prevalent in the schizophrenia population compared to the control population. In deceased schizophrenic patients, T2DM was the most frequently recorded comorbidity, contributing to 31.4% of hospital deaths (only 14.4% of schizophrenic patients with comorbid T2DM survived the study period). Further predictors of general hospital mortality in schizophrenia were found to be alcoholic liver disease (OR = 10.3), parkinsonism (OR = 5.0), T1DM (OR = 3.8), non-specific renal failure (OR = 3.5), ischaemic stroke (OR = 3.3), pneumonia (OR = 3.0), iron-deficiency anaemia (OR = 2.8), COPD (OR = 2.8), and bronchitis (OR = 2.6). There were no significant differences in their impact on hospital mortality compared to control subjects with the same diseases except parkinsonism which was associated with higher mortality in the schizophrenia population compared with the control population. The prevalence of parkinsonism was significantly elevated in the 255 deceased schizophrenic patients (5.5 %) than in those 1,163 surviving the study period (0.8 %, OR = 5.0) and deceased schizophrenic patients had significantly more suffered extrapyramidal symptoms than deceased control subjects (5.5 vs. 1.5 %). Therefore patients with schizophrenia have a higher burden of physical comorbidity that is associated with a worse outcome in a 12-year follow-up of mortality in general hospitals compared with hospital controls. However, schizophrenic patients die of the same physical diseases as their peers without schizophrenia. The most relevant physical risk factors of general hospital mortality are T2DM, COPD and infectious respiratory complications, iron-deficiency anaemia, T1DM, unspecific renal failure, ischaemic stroke, and alcoholic liver disease. Additionally, parkinsonism is a major risk factor for general hospital mortality in schizophrenia. Thus, optimal monitoring and management of acute T2DM and COPD with its infectious respiratory complications, as well as the accurate detection and management of iron-deficiency anaemia, of diabetic-related long-term micro- and macrovascular complications, of alcoholic liver disease, and of extrapyramidal symptoms are of utmost relevance in schizophrenia.

Systematic review and meta-analysis of validation studies on a diabetes case definition from health administrative records

OBJECTIVES

Health administrative data are frequently used for diabetes surveillance. We aimed to determine the sensitivity and specificity of a commonly-used diabetes case definition (two physician claims or one hospital discharge abstract record within a two-year period) and their potential effect on prevalence estimation.

METHODS

Following Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, we searched Medline (from 1950) and Embase (from 1980) databases for validation studies through August 2012 (keywords: "diabetes mellitus"; "administrative databases"; "validation studies"). Reviewers abstracted data with standardized forms and assessed quality using Quality Assessment of Diagnostic Accuracy Studies (QUADAS) criteria. A generalized linear model approach to random-effects bivariate regression meta-analysis was used to pool sensitivity and specificity estimates. We applied correction factors derived from pooled sensitivity and specificity estimates to prevalence estimates from national surveillance reports and projected prevalence estimates over 10 years (to 2018).

RESULTS

The search strategy identified 1423 abstracts among which 11 studies were deemed relevant and reviewed; 6 of these reported sensitivity and specificity allowing pooling in a meta-analysis. Compared to surveys or medical records, sensitivity was 82.3% (95%CI 75.8, 87.4) and specificity was 97.9% (95%CI 96.5, 98.8). The diabetes case definition underestimated prevalence when it was ≤10.6% and overestimated prevalence otherwise.

CONCLUSION

The diabetes case definition examined misses up to one fifth of diabetes cases and wrongly identifies diabetes in approximately 2% of the population. This may be sufficiently sensitive and specific for surveillance purposes, in particular monitoring prevalence trends. Applying correction factors to adjust prevalence estimates from this definition may be helpful to increase accuracy of estimates.

High prevalence of diabetes mellitus and hospital-related hyperglycaemia in French general wards

AIM

The study evaluated the in-hospital prevalence of diabetes and hospital-related hyperglycaemia in a variety of French general wards.

METHODS

The multicentre cross-sectional study involving nine French hospitals measured venous fasting plasma glucose (FPG) on a single day in patients hospitalized in adult medical and surgical short-term wards. Diabetes status and length of stay were recorded.

RESULTS

Of the 2141 inpatients included in the study, 355 (16.5%) had known diabetes, 156 (7.3%) had screened diabetes (FPG ≥7 mmol/L with no diabetes history), 515 (24.1%) had impaired fasting glucose (IFG; FPG 5.5-6.9 mmol/L) and 1115 (52.1%) had normal glucose values (FPG < 5.5 mmol/L). Diabetes prevalence varied from 11% in hospitals in the west of France to 21% in hospitals in northern and eastern regions. The highest known diabetes prevalence was observed in units for cardiovascular surgery (33%), infectious diseases (27%) and kidney disorders (26%). In cancer units, one-fifth of patients had screened diabetes and one-sixth had known diabetes. Among the known diabetes patients, 127 (36%) were already being treated with insulin, while an additional 41 (12%) started insulin therapy during their hospital stay. Patients with known and screened diabetes were older (70.8 ± 12.2 and 71.1 ± 15.6 years, respectively) than the normal-glucose patients (65.6 ± 18.9 years; P<0.001). Average length of stay was no different between known diabetes and normal-glucose patients after adjusting for age (11.3 ± 7.7 vs 10.0 ± 7.4 days; NS).

CONCLUSION

Overall, metabolic glucose disorders (known or screened diabetes and IFG) were found in 48% of inpatients in various French hospital general wards.

Parenteral nutrition-associated hyperglycemia in non-critically ill inpatients increases the risk of in-hospital mortality (multicenter study)

OBJECTIVE

Hyperglycemia may increase mortality in patients who receive total parenteral nutrition (TPN). However, this has not been well studied in noncritically ill patients (i.e., patients in the nonintensive care unit setting). The aim of this study was to determine whether mean blood glucose level during TPN infusion is associated with increased mortality in noncritically ill hospitalized patients.

RESEARCH DESIGN AND METHODS

This prospective multicenter study involved 19 Spanish hospitals. Noncritically ill patients who were prescribed TPN were included prospectively, and data were collected on demographic, clinical, and laboratory variables as well as on in-hospital mortality.

RESULTS

The study included 605 patients (mean age 63.2 ± 15.7 years). The daily mean TPN values were 1.630 ± 323 kcal, 3.2 ± 0.7 g carbohydrates/kg, 1.26 ± 0.3 g amino acids/kg, and 0.9 ± 0.2 g lipids/kg. Multiple logistic regression analysis showed that the patients who had mean blood glucose levels >180 mg/dL during the TPN infusion had a risk of mortality that was 5.6 times greater than those with mean blood glucose levels <140 mg/dL (95% CI 1.47-21.4 mg/dL) after adjusting for age, sex, nutritional state, presence of diabetes or hyperglycemia before starting TPN, diagnosis, prior comorbidity, carbohydrates infused, use of steroid therapy, SD of blood glucose level, insulin units supplied, infectious complications, albumin, C-reactive protein, and HbA1c levels.

CONCLUSIONS

Hyperglycemia (mean blood glucose level >180 mg/dL) in noncritically ill patients who receive TPN is associated with a higher risk of in-hospital mortality.

[Quality of hospital discharge reports in terms of current legislation and expert recommendations]

OBJECTIVE

To determine the quality of hospital discharge reports (HDRs) taking into account current legislation and the conclusions of the consensus on hospital discharge reports in medical specialities in 11 community hospitals in Andalusia (Spain).

MATERIAL AND METHODS

A cross-sectional study of 1,708 HDRs was carried out. We determined the presence or absence of the various items required by current legislation and by the recommendations of the above-mentioned consensus.

RESULTS

A total of 97.4% (95% confidence interval [95% CI]: 96.5-98.2) of the HDRs were classified as satisfactory according to the stipulations of current legislation. However, when the assessment was based on the consensus, the rate of adequacy fell to 72.1% (95% CI: 70.0-74.3). A notable finding was the absence of the duration of treatment after hospital discharge in 39.4% of the HDRs.

CONCLUSIONS

HDRs show an excellent level of compliance with the data required by current regulations, but their intrinsic quality needs to be improved.

[Consensus document on the management after hospital discharge of patient with hyperglycaemia]

The present document intends to adapt the general recommendations set up in a consensus to elaborate the hospital discharge report in medical specialties to the specific needs of the hospitalized diabetic population. Diabetes is an illness with a very high health cost, being the global risk of death in people with diabetes almost double than in non-diabetes people, justifying the fact that diabetes constitutes one of the most frequent diagnoses in hospitalized patients and the growing interest upon hyperglycaemia management during hospitalization and at discharge. To set up an adequate treatment plan at discharge suitable for each patient, the most important elements to take into account are the etiology and prior hyperglycaemia treatment, the patient's clinical situation and the degree of glycaemia control. Due to instability of glycaemia control, it is also needed to anticipate the educational needs for each patient, as well as to set up the monitoring schedule and follow-up at discharge, and an adequate treatment plan at discharge.

Characteristics and hospitalization costs of patients with diabetes in Spain

Using data from the Minimum Basic Dataset (MBDS) we described the hospitalizations in Spanish patients with diabetes. In 2007, acute diabetes-related complications accounted for 3.9% of admissions and chronic complications accounted for 30.3%. Mean cost per patient was euro4339. We conclude that diabetes had a high impact on hospitalizations including costs.

Prevalence of elevated hemoglobin A1c among patients admitted to the hospital without a diagnosis of diabetes

CONTEXT

One in four hospitalized patients has diagnosed diabetes. The prevalence of unrecognized, or undiagnosed, diabetes among hospitalized patients is not well established.

OBJECTIVE

Our objective was to determine the prevalence of unrecognized probable diabetes in this patient population determined by elevated hemoglobin A1c (HbA1c) level.

DESIGN

We conducted a prospective observational cohort trial with retrospective follow-up of patients with elevated HbA1c levels and no diagnosis of diabetes. HbA1c levels were obtained for all patients.

SETTING

The study was conducted at an acute care general hospital.

PATIENTS

Patients included 695 adult, nonobstetric patients admitted on 11 d in 2006.

MAIN OUTCOME MEASURES

Outcome measures included rate of unrecognized probable diabetes, defined as admission HbA1c of more than 6.1% and no diagnosis of diabetes or treatment with antidiabetic medications before or during their admission and rate of unrecognized diabetes 1 yr after discharge.

RESULTS

Eighteen percent of hospitalized patients had elevated HbA1c levels without a diagnosis of diabetes. Random glucose levels poorly predicted elevated HbA1c levels (area under receiver operating characteristic curve, 0.60). Neither diagnosed diabetes nor HbA1c level was associated with length of stay or costs (P>0.1 for all comparisons). Only 15% of patients with elevated HbA1c levels who continued to receive care within the system studied had diabetes diagnosed in the year after the index admission.

CONCLUSIONS

Nearly one in five adult patients admitted to a large general hospital had unrecognized probable diabetes, based on elevated HbA1c levels. Random glucose levels during the hospital stay were poorly predictive of this condition. Few hospitalized patients with elevated HbA1c levels were diagnosed within the year after admission.

Diabetes team consultation: impact on length of stay of diabetic patients admitted to a short-stay unit

OBJECTIVE

To assess the impact of endocrinology team consultation on hospital stay and clinical outcomes of diabetic patients admitted with a primary non-diabetes-related diagnosis in a short stay unit (SSU).

METHODS

Patients admitted to the SSU between 2001 and 2005. Between 2001 and 2003 there was no endocrinology team consultation available and the management of hyperglycemia was handled by the SSU team alone. From 2003 until 2005 an endocrinology team was in charge of diabetes care. We compared in both periods: prevalence of diabetes, length of hospital stay, mortality, early readmissions and number of patients requiring conventional hospitalization.

RESULTS

In period 2001-2003, 1023 patients were admitted, among which 212 were diabetic (20.7%). Over the years 2003-2005, 892 patients were hospitalized, 223 were diabetic (25%). Clinical characteristics of diabetic patients from both periods were comparable, but glycaemia at admission was higher on the second period (217 mg/dl versus 198 mg/dl). The length of stay of diabetic patients in the second period decreased from 5.49 to 4.90 days. There were no significant differences in mortality (1.4% versus 0.4%) or in early re-admissions among the two periods.

CONCLUSIONS

The intervention of a diabetes team diminished the average length of stay of diabetic patients.

Diabetes care in hospitalized noncritically ill patients: More evidence for clinical inertia and negative therapeutic momentum

BACKGROUND

Little is known about management of hyperglycemia in inpatients.

OBJECTIVE

To gain insight into caring for hospitalized patients with hyperglycemia.

DESIGN

Retrospective analysis.

SETTING

Teaching hospital.

PATIENTS

Data on all patients discharged between January 1, 2001, and December 31, 2004 with a diagnosis of diabetes or hyperglycemia were extracted and linked to laboratory and pharmacy databases. Only the data on patients who did not require intensive care and who were hospitalized for at least 3 days were analyzed.

MEASUREMENTS

Average bedside glucose during the first and last 24 hours of hospital stay and for the entire length of stay; assessment of changes in insulin regimen and dose.

RESULTS

The average age of patients included in the study (n = 2916) was 69 years. Fifty-seven percent of the patients were men, 90% were white, and average length of stay was 5.7 days. More than 20% of the patients had evidence of sustained hyperglycemia. Forty-two percent of the patients who showed poor control of glycemia (glucose > 200 mg/dL) during the first 24 hours were discharged in poor control. The frequency of hypoglycemia was low (only 2.2 of 100 measurements per person) compared with hyperglycemia (25.5 of 100 measurements per person). Most patients (72%) received insulin during hospitalization, but there was high use of short-acting insulin and less than optimal intensification of therapy (clinical inertia); many patients had insulin therapy decreased despite persistent hyperglycemia (negative therapeutic momentum).

CONCLUSIONS

Glycemic control in the hospital was frequently poor, and there was suboptimal use of insulin, even among patients with sustained hyperglycemia. Educational programs directed at practitioners should focus on the importance of inpatient glucose control and provide guidelines on how and when to change therapy.

Common Reasons for Hospitalization Among Adult Patients with Diabetes

OBJECTIVE

To determine reasons for hospitalization among adult patients with diabetes.

METHODS

A cross-sectional analysis was conducted of hospital discharges in the state of Georgia for the years 1998 through 2001 that contained either a primary or a coexisting diagnosis of diabetes. With use of the Clinical Classification Software of the Agency for Healthcare Research and Quality, the principal diagnoses among diabetes-related hospital discharges were organized into diagnostic categories.

RESULTS

Diabetes was listed as a diagnosis in 14% of all Georgia hospital discharges of adult patients during our study period (57% women; 62% non-Hispanic white; mean age, 64 years; mean length of stay, 5.7 days; and mean hospital charge, 13,540 dollars). Among patients with a diagnosis of diabetes, the 3 most common categories of discharges were "diseases of the circulatory system" (33%), "endocrine, nutritional, and metabolic; immunity disorders" (13%), and "diseases of the respiratory system: (11%). When infections were identified and aggregated, however, these conditions became the second most frequent discharge category (14% of all hospital discharges among patients with diabetes). "Congestive heart failure," "coronary atherosclerosis," and "acute myocardial infarction" were the first, second, and fifth most frequently found unique diagnoses, respectively, among patients with diabetes.

CONCLUSION

In this study, diseases of the circulatory system were the most common diagnoses in hospital discharge data for adult patients with diabetes in Georgia. Hospitals should be cognizant of the increased burden placed on them by diabetes, and outpatient treatment of diabetes should focus on prevention of cardiovascular diseases to avoid hospitalizations.

Diagnosing diabetes and preventing rehospitalizations: the urban diabetes study

BACKGROUND

Patients with diabetes frequently are hospitalized, and quality of inpatient care for diabetes is of great concern. Rehospitalization after hospital discharge is a frequent adverse outcome experienced by patients with diabetes.

OBJECTIVES

We assessed the frequency of and risk factors for rehospitalization among all Philadelphia residents with diabetes.

METHODS

Individual histories of hospitalization were ascertained from hospital discharge summaries for Philadelphia residents ages 25-84 who had at least 1 diabetes hospitalization from 1994 through 2001. Logistic regression was used to assess predictors of nonelective rehospitalization within 30 days of discharge, including recording of diabetes diagnosis.

RESULTS

Nonelective rehospitalizations within 30 days of hospital discharge were ascertained for 58,308 (20.0%) of 291,752 discharges. The proportion rehospitalized was 9.4% after a patient's first diabetes diagnosis hospitalization; after later discharges for which a diabetes diagnosis was not recorded, rehospitalizations occurred in 30.6% of all cases. The absence of a diabetes diagnosis was a highly significant predictor of rehospitalization after adjustment for age, year, gender, race/ethnicity, insurance status, admission type, severity code, length of stay, discharge status, and number of previous hospitalizations.

CONCLUSION

Failure to record a diabetes diagnoses in administrative hospital discharge data may reflect lack of attention to the critical needs of patients with diabetes who are being treated for other conditions, whereas the attention to patient education and follow-up planning for patients with incident diabetes diagnoses may reduce the risk of rehospitalization.

Hospital in-patients with diabetes: increasing prevalence and management problems

AIMS

To re-assess the prevalence, management problems, clinical outcomes and discharge summaries of hospital in-patients with diabetes.

METHODS

Case records of all patients occupying in-patient beds were audited on a single weekday in 2003 in a large urban hospital and repeated after 3 months. Data was compared with an identical audit 12 years previously.

RESULTS

Over 12 years the number of beds available for admission (1191) had reduced by 25% with a bed occupancy of 97%. Diabetes prevalence had increased from 7.0% to 11.1% (P < 0.01) (97% Type 2). Diabetes management was considered inappropriate in 29%, more than in 1991 (20%). After 3 months, discharge summaries had been completed on 75% of patients but diabetes was mentioned in only 53%.

CONCLUSION

The prevalence of in-patient diabetes (11.1%) was over 50% greater and diabetes management was suboptimal in more patients than in 1991. In many length of stay was prolonged and almost half of the discharge summaries did not mention diabetes. These findings have major implications for service delivery and resource planning.

Excess hospitalizations, hospital days, and inpatient costs among people with diabetes in Andalusia, Spain

OBJECTIVE

The goal of this study was to estimate the excess hospitalizations, hospital days, and inpatient costs attributable to diabetes in Andalusia, Spain (37 hospitals, 7,236,459 inhabitants), during 1999 compared with those without diabetes.

RESEARCH DESIGN AND METHODS

This study was an analysis of all hospital discharges. Those with an ICD-9-CM code of 250 as either the main or secondary diagnosis were considered to have been admissions of individuals with diabetes. An estimate of costs was applied to each inpatient admission by assigning a cost weight based on the diagnostic-related group (DRG) related to each admission.

RESULTS

A total of 538,580 admissions generated 4,310,654 hospital bed-days and total costs of 940,026,949 euro. People with diabetes accounted for 9.7% of all hospital discharges, 13.8% of total stays, and 14.1% of the total cost. Of the total cost for individuals with diabetes (132,509,217 euro), 58.3% were excess costs, of which 47% was attributable to cardiovascular complications and 43% to admissions for comorbid diseases. Individuals 45-75 years of age accounted for 75% of the excess costs. The rate of admissions during the study year was 145 per 1,000 inhabitants for individuals with diabetes compared with 70 admissions per 1,000 inhabitants for individuals without diabetes.

CONCLUSIONS

The costs arising from hospitalization of individuals with diabetes are disproportionate in relation to their prevalence. For those aged >or=45 years, cardiovascular complications were clearly the most important factor determining increased costs from diabetes.