Your browser doesn't support javascript.
Mostrar: 20 | 50 | 100
Resultados 1 - 20 de 53
Filtrar
Mais filtros










Base de dados
Intervalo de ano de publicação
1.
JAMA Netw Open ; 3(1): e1919099, 2020 Jan 03.
Artigo em Inglês | MEDLINE | ID: mdl-31922562

RESUMO

Importance: Severe hypoglycemia is a serious and potentially preventable complication of diabetes, with some of the most severe episodes requiring emergency department (ED) care or hospitalization. A variety of health conditions increase the risk of hypoglycemia. People with diabetes often have multiple comorbidities, and the association of such multimorbidity with hypoglycemia risk in the context of other risk factors is uncertain. Objective: To examine the associations of age, cumulative multimorbidity, glycated hemoglobin (HbA1c) level, and use of glucose level-lowering medication with hypoglycemia-related ED visits and hospitalizations. Design, Setting, and Participants: Cohort study of claims and laboratory data from OptumLabs Data Warehouse, an administrative claims database of commercially insured and Medicare Advantage beneficiaries in the United States. Participants were adults (aged ≥18 years) with diabetes who had an available HbA1c level result in 2015. Data from January 1, 2014, to December 31, 2016, were analyzed. Final analyses were conducted from December 2017 to September 2018. Main Outcomes and Measures: This study calculated rates of hypoglycemia-related ED visits and hospitalizations during the year after the index HbA1c level was obtained, stratified by patient demographic characteristics, diabetes type, comorbidities (from 16 guideline-specified high-risk conditions), index HbA1c level, and glucose level-lowering medication use. The association of each variable with hypoglycemia-related ED and hospital care was examined using multivariable Poisson regression analysis overall and by diabetes type. Results: The study cohort was composed of 201 705 adults with diabetes (mean [SD] age, 65.8 [12.1] years; 102 668 [50.9%] women; 118 804 [58.9%] white; mean [SD] index HbA1c level, 7.2% [1.5%]). Overall, there were 9.06 (95% CI, 8.64-9.47) hypoglycemia-related ED visits and hospitalizations per 1000 persons per year. The risk of hypoglycemia-related ED visits and hospitalizations was increased by age 75 years or older (incidence rate ratio [IRR], 1.56 [95% CI, 1.23-2.02] vs 18-44 years), black race/ethnicity (IRR, 1.30 [95% CI, 1.16-1.46] vs white race/ethnicity), lower annual household income (IRR, 0.63 [95% CI, 0.53-0.74] for ≥$100 000 vs <$40 000), number of comorbidities (increasing from IRR of 1.66 [95% CI, 1.42-1.95] in the presence of 2 comorbidities to IRR of 4.12 [95% CI, 3.07-5.51] with ≥8 comorbidities compared with ≤1), prior hypoglycemia-related ED visit or hospitalization (IRR, 6.60 [95% CI, 5.77-7.56]), and glucose level-lowering treatment regimen (IRR, 6.73 [95% CI, 4.93-9.22] for sulfonylurea; 12.53 [95% CI, 8.90-17.64] for basal insulin; and 27.65 [95% CI, 20.32-37.63] for basal plus bolus insulin compared with other medications). Independent of these factors, having type 1 diabetes was associated with a 34% increase in the risk of hypoglycemia-related ED visits or hospitalizations (IRR, 1.34 [95% CI, 1.15-1.55]). The index HbA1c level was associated with hypoglycemia-related ED visits and hospitalizations when both low (IRR, 1.45 [95% CI, 1.12-1.87] for HbA1c level ≤5.6% vs 6.5%-6.9%) and high (IRR, 1.24 [95% CI, 1.02-1.50] for HbA1c level ≥10%). Conclusions and Relevance: In this cohort study of adults with diabetes, the risk of an ED visit or hospitalization for hypoglycemia appeared to be highest among patients with type 1 diabetes, multiple comorbidities, prior severe hypoglycemia, and sulfonylurea and/or insulin use. At-risk patients may benefit from individualized treatment regimens to decrease their risk of hypoglycemia.

2.
Am Heart J ; 219: 31-36, 2020 Jan.
Artigo em Inglês | MEDLINE | ID: mdl-31710842

RESUMO

BACKGROUND: A deep learning algorithm to detect low ejection fraction (EF) using routine 12-lead electrocardiogram (ECG) has recently been developed and validated. The algorithm was incorporated into the electronic health record (EHR) to automatically screen for low EF, encouraging clinicians to obtain a confirmatory transthoracic echocardiogram (TTE) for previously undiagnosed patients, thereby facilitating early diagnosis and treatment. OBJECTIVES: To prospectively evaluate a novel artificial intelligence (AI) screening tool for detecting low EF in primary care practices. DESIGN: The EAGLE trial is a pragmatic two-arm cluster randomized trial (NCT04000087) that will randomize >100 clinical teams (i.e., clusters) to either intervention (access to the new AI screening tool) or control (usual care) at 48 primary care practices across Minnesota and Wisconsin. The trial is expected to involve approximately 400 clinicians and 20,000 patients. The primary endpoint is newly discovered EF ≤50%. Eligible patients will include adults who undergo ECG for any reason and have not been previously diagnosed with low EF. Data will be pulled from the EHR, and no contact will be made with patients. A positive deviance qualitative study and a post-implementation survey will be conducted among select clinicians to identify facilitators and barriers to using the new screening report. SUMMARY: This trial will examine the effectiveness of the AI-enabled ECG for detection of asymptomatic low EF in routine primary care practices and will be among the first to prospectively evaluate the value of AI in real-world practice. Its findings will inform future implementation strategies for the translation of other AI-enabled algorithms.

3.
J Gen Intern Med ; 2019 Dec 02.
Artigo em Inglês | MEDLINE | ID: mdl-31792867

RESUMO

BACKGROUND: People with chronic kidney disease (CKD) are at risk for adverse events and/or CKD progression with use of renally eliminated or nephrotoxic medications. OBJECTIVE: To examine the prevalence of potentially inappropriate medication (PIM) use by U.S. adults by CKD stage and self-reported CKD awareness. DESIGN: Cross-sectional analysis of National Health and Nutrition Examination Surveys, 2011-2016 PARTICIPANTS: Non-pregnant adults with stages 3a (eGFR 45-59 mL/min/1.73 m2), 3b (eGFR 30-44), or 4-5 (eGFR < 30) CKD, stratified as CKD-aware/unaware. MAIN MEASURES: PIMs were identified on the basis of KDIGO guidelines, label information, and literature review. We calculated proportions using any and individual PIMs, assessing for differences over CKD awareness within each CKD stage. Analyses were adjusted for age, sex, race/ethnicity, education, comorbidities, and insurance type. KEY RESULTS: Adjusted proportions of U.S. adults taking any PIM(s) exceeded 50% for all CKD stages and awareness categories, and were highest among CKD-unaware patients with stages 4-5 CKD: 66.6% (95% CI, 55.5-77.8). Proton pump inhibitors, opioids, metformin, sulfonylureas, and non-steroidal anti-inflammatory drugs (NSAIDs) were all used frequently across CKD stages. NSAIDs were used less frequently when CKD-aware by patients with stage 3a CKD (2.2% [95% CI, - 0.3 to 4.7] vs. 10.7% [95% CI, 7.6 to 13.8]) and stages 4-5 CKD (0.8% [95% CI, - 0.9 to 2.5] vs. 16.5% [95% CI, 4.0 to 29.0]). Metformin was used less frequently when CKD-aware by patients with stage 3b CKD (8.1% [95% CI, 0.3-15.9] vs. 26.5% [95% CI, 17.4-35.7]) and stages 4-5 CKD (none vs. 20.8% [95% CI, 1.8-39.8]). The impact of CKD awareness was statistically significant after correction for multiple comparisons only for NSAIDs in stage 3a CKD. CONCLUSIONS: PIMs are frequently used by people with CKD, with some impact of CKD awareness on NSAID and metformin use. This may lead to adverse outcomes or hasten CKD progression, reinforcing the need for improved medication management among people with CKD.

4.
Endocrine ; 2019 Dec 04.
Artigo em Inglês | MEDLINE | ID: mdl-31802353

RESUMO

PURPOSE: To examine the proportion of diabetes-focused clinical encounters in primary care and endocrinology practices where the evaluation for hypoglycemia is documented; and when it is, identify clinicians' stated actions in response to patient-reported events. METHODS: A total of 470 diabetes-focused encounters among 283 patients nonpregnant adults (≥18 years) with type 1 or type 2 diabetes mellitus in this retrospective cohort study. Participants were randomly identified in blocks of treatment strategy and care location (95 and 52 primary care encounters among hypoglycemia-prone medications (i.e. insulin, sulfonylurea) and others patients, respectively; 94 and 42 endocrinology encounters among hypo-treated and others, respectively). Documentation of hypoglycemia and subsequent management plan in the electronic health record were evaluated. RESULTS: Overall, 132 (46.6%) patients had documentation of hypoglycemia assessment, significantly more prevalent among hypo-treated patients seen in endocrinology than in primary care (72.3% vs. 47.4%; P = 0.001). Hypoglycemia was identified by patient in 38.2% of encounters. Odds of hypoglycemia assessment documentation was highest among the hypo-treated (OR 13.6; 95% CI 5.5-33.74, vs. others) and patients seen in endocrine clinic (OR 4.48; 95% CI 2.3-8.6, vs. primary care). After documentation of hypoglycemia, treatment was modified in 30% primary care and 46% endocrine clinic encounters; P = 0.31. Few patients were referred to diabetes self-management education and support (DSMES). CONCLUSIONS: Continued efforts to improve hypoglycemia evaluation, documentation, and management are needed, particularly in primary care. This includes not only screening at-risk patients for hypoglycemia, but also modifying their treatment regimens and/or leveraging DSMES.

5.
BMJ ; 367: l5887, 2019 11 05.
Artigo em Inglês | MEDLINE | ID: mdl-31690574

RESUMO

Diabetes is a major and costly health concern worldwide, with high morbidity, disability, mortality, and impaired quality of life. The vast majority of people living with diabetes have type 2 diabetes. Historically, the main strategy to reduce complications of type 2 diabetes has been intensive glycemic control. However, the body of evidence shows no meaningful benefit of intensive (compared with moderate) glycemic control for microvascular and macrovascular outcomes important to patients, with the exception of reduced rates of non-fatal myocardial infarction. Intensive glycemic control does, however, increase the risk of severe hypoglycemia and incurs additional burden by way of polypharmacy, side effects, and cost. Additionally, data from cardiovascular outcomes trials showed that cardiovascular, kidney, and mortality outcomes may be improved with use of specific classes of glucose lowering drugs largely independently of their glycemic effects. Therefore, delivering evidence based, patient centered care to people with type 2 diabetes requires a paradigm shift and departure from the predominantly glucocentric view of diabetes management. Instead of prioritizing intensive glycemic control, the focus needs to be on ensuring access to adequate diabetes care, aligning glycemic targets to patients' goals and situations, minimizing short term and long term complications, reducing the burden of treatment, and improving quality of life.


Assuntos
Complicações do Diabetes/prevenção & controle , Diabetes Mellitus Tipo 2/tratamento farmacológico , Hipoglicemia/epidemiologia , Hipoglicemiantes/administração & dosagem , Qualidade de Vida , Glicemia/análise , Glicemia/efeitos dos fármacos , Complicações do Diabetes/sangue , Diabetes Mellitus Tipo 2/sangue , Diabetes Mellitus Tipo 2/complicações , Medicina Baseada em Evidências/métodos , Medicina Baseada em Evidências/normas , Humanos , Hiperglicemia/sangue , Hiperglicemia/tratamento farmacológico , Hipoglicemia/induzido quimicamente , Hipoglicemiantes/efeitos adversos , Incidência , Metanálise como Assunto , Assistência Centrada no Paciente/métodos , Assistência Centrada no Paciente/normas , Guias de Prática Clínica como Assunto , Revisão Sistemática como Assunto , Resultado do Tratamento
6.
JAMA Netw Open ; 2(10): e1913249, 2019 Oct 02.
Artigo em Inglês | MEDLINE | ID: mdl-31603490

RESUMO

Importance: Differences in readmission rates among racial and ethnic minorities have been reported, but data among people with diabetes are lacking despite the high burden of diabetes and its complications in these populations. Objectives: To examine racial/ethnic differences in all-cause readmission among US adults with diabetes and categorize patient- and system-level factors associated with these differences. Design, Setting, and Participants: This retrospective cohort study includes 272 758 adult patients with diabetes, discharged alive from the hospital between January 1, 2009, and December 31, 2014, and stratified by race/ethnicity. An administrative claims data set of commercially insured and Medicare Advantage beneficiaries across the United States was used. Data analysis took place between October 2016 and February 2019. Main Outcomes and Measures: Unplanned all-cause readmission within 30 days of discharge and individual-, clinical-, economic-, index hospitalization-, and hospital-level risk factors for readmission. Results: A total of 467 324 index hospitalizations among 272 758 adults with diabetes (mean [SD] age, 67.7 [12.7]; 143 498 [52.6%] women) were examined. The rates of 30-day all-cause readmission were 10.2% (33 683 of 329 264) among white individuals, 12.2% (11 014 of 89 989) among black individuals, 10.9% (4151 of 38 137) among Hispanic individuals, and 9.9% (980 of 9934) among Asian individuals (P < .001). After adjustment for all factors, only black patients had a higher risk of readmission compared with white patients (odds ratio, 1.05; 95% CI, 1.02-1.08). This increased readmission risk among black patients was sequentially attenuated, but not entirely explained, by other demographic factors, comorbidities, income, reason for index hospitalization, or place of hospitalization. Compared with white patients, both black and Hispanic patients had the highest observed-to-expected (OE) readmission rate ratio when their income was low (annual household income <$40 000 among black patients: OE ratio, 1.11; 95% CI, 1.09-1.14; among Hispanic patients: OE ratio, 1.11; 95% CI, 1.07-1.16) and when they were hospitalized in nonprofit hospitals (black patients: OE ratio, 1.10; 95% CI, 1.08-1.12; among Hispanic patients: OE ratio, 1.08; 95% CI, 1.05-1.12), academic hospitals (black patients: OE ratio, 1.16; 95% CI, 1.13-1.20; Hispanic patients: OE ratio, 1.12; 95% CI, 1.06-1.19), or large hospitals (ie, with ≥400 beds; black patients: OE ratio, 1.11; 95% CI, 1.09-1.14; Hispanic patients: OE ratio, 1.09; 95% CI, 1.04-1.14). Conclusions and Relevance: In this study, black patients with diabetes had a significantly higher risk of readmission than members of other racial/ethnic groups. This increased risk was most pronounced among lower-income patients hospitalized in nonprofit, academic, or large hospitals. These findings reinforce the importance of identifying and addressing the many reasons for persistent racial/ethnic differences in health care quality and outcomes.

7.
8.
Diabetes Technol Ther ; 21(12): 702-712, 2019 Dec.
Artigo em Inglês | MEDLINE | ID: mdl-31418588

RESUMO

Background: High-quality diabetes care is evidence-based, timely, and equitable. Sodium-glucose cotransporter-2 inhibitors (SGLT2i) are the most recently approved class of glucose-lowering medications with additional cardio- and renal-protective benefits and low risk of hypoglycemia. Cardiovascular and kidney disease are among the most common chronic diabetes complications, whereas hypoglycemia is the most prevalent adverse effect of glucose-lowering therapy. We examine the sociodemographic and clinical factors associated with early SGLT2i initiation and appropriateness of use based on contemporaneous scientific evidence. Materials and Methods: Retrospective analysis of medical and pharmacy claims data from OptumLabs® Data Warehouse for commercially insured and Medicare Advantage adult beneficiaries with diabetes types 1 and 2, who filled any glucose-lowering medication between January 1, 2013 and December 31, 2016. Demographic (age, sex, race, income), clinical (comorbidities), and insurance-related factors affecting first prescription for a SGLT2i were examined using multivariable logistic regression. Results: Among 1,054,727 adults with pharmacologically treated diabetes, 7.2% (n = 75,500) initiated a SGLT2i. Patients with prior myocardial infarction (MI) (odds ratio [OR]: 0.94, 95% confidence interval [CI]: 0.91-0.96), heart failure (HF) (OR: 0.93, 95% CI: 0.91-0.94), kidney disease (OR: 0.80, 95% CI: 0.78-0.81), and severe hypoglycemia (OR: 0.96, 95% CI: 0.94-0.98) were all less likely to start a SGLT2i; P < 0.001 for all. SGLT2i were also less likely to be started by patients ≥75 years (OR: 0.57, 95% CI: 0.55-0.59, vs. 18-44 years), Black patients (OR: 0.93, 95% CI: 0.91-0.95, vs. White), and those with Medicare Advantage insurance (OR: 0.63, 95% CI: 0.62-0.64, vs. commercial). Conclusions: Younger, healthier, non-Black patients with commercial health insurance were most likely to start taking SGLT2i. Patients with MI, HF, kidney disease, and prior hypoglycemia were less likely to use SGLT2i, despite evidence supporting their preferential use in these patients. Efforts to address this treatment-risk paradox may help improve health outcomes among patients with type 2 diabetes.

9.
Mayo Clin Proc ; 94(9): 1731-1742, 2019 09.
Artigo em Inglês | MEDLINE | ID: mdl-31422897

RESUMO

OBJECTIVE: To estimate the contemporary prevalence of intensive glucose-lowering therapy among US adults with diabetes and model the number of hypoglycemia-related emergency department (ED) visits and hospitalizations that are attributable to such intensive treatment. PATIENTS AND METHODS: US adults with diabetes and glycated hemoglobin (HbA1c) levels less than 7.0% who were included in the National Health and Nutrition Examination Survey (NHANES) between 2011 and 2014. Participants were categorized as clinically complex if 75 years or older or with 2 or more activities of daily living limitations, end-stage renal disease, or 3 or more chronic conditions. Intensive treatment was defined as any glucose-lowering medications with HbA1c levels of 5.6% or less or 2 or more with HbA1c levels of 5.7% to 6.4%. First, we quantified the proportion of clinically complex and intensively treated individuals in the NHANES population. Then, we modeled the attributable hypoglycemia-related ED visits/hospitalizations over a 2-year period based on published data for event risk. RESULTS: Almost half (48.8% [10,719,057 of 21,980,034]) of US adults with diabetes (representing 10.7 million US adults) had HbA1c levels less than 7.0%. Among them, 32.3% (3,466,713 of 10,719,057) were clinically complex, and 21.6% (2,309,556 of 10,719,057) were intensively treated, with no difference by clinical complexity. Over a 2-year period, we estimated 31,511 hospitalizations and 30,954 ED visits for hypoglycemia in this population; of these, 4774 (95% CI, 954-9714) hospitalizations and 4804 (95% CI, 862-9851) ED visits were attributable to intensive treatment. CONCLUSION: Intensive glucose-lowering therapy, particularly among vulnerable clinically complex adults, is strongly discouraged because it may lead to hypoglycemia. However, intensive treatment was equally prevalent among US adults, irrespective of clinical complexity. Over a 2-year period, an estimated 9578 hospitalizations and ED visits for hypoglycemia could be attributed to intensive diabetes treatment, particularly among clinically complex patients. Patients at risk for hypoglycemia may benefit from treatment deintensification to reduce hypoglycemia risk and treatment burden.


Assuntos
Diabetes Mellitus Tipo 2/tratamento farmacológico , Serviço Hospitalar de Emergência/estatística & dados numéricos , Hipoglicemia/induzido quimicamente , Hipoglicemia/epidemiologia , Hipoglicemiantes/administração & dosagem , Atividades Cotidianas , Fatores Etários , Idoso , Glicemia/análise , Comorbidade , Diabetes Mellitus Tipo 2/sangue , Diabetes Mellitus Tipo 2/epidemiologia , Relação Dose-Resposta a Droga , Esquema de Medicação , Feminino , Seguimentos , Hemoglobina A Glicada/análise , Hospitalização/estatística & dados numéricos , Humanos , Hipoglicemia/fisiopatologia , Hipoglicemiantes/efeitos adversos , Falência Renal Crônica/epidemiologia , Masculino , Pessoa de Meia-Idade , Inquéritos Nutricionais , Prevalência , Estudos Retrospectivos , Medição de Risco , Índice de Gravidade de Doença , Fatores Sexuais , Fatores de Tempo , Estados Unidos
12.
Circulation ; 140(7): e294-e324, 2019 Aug 13.
Artigo em Inglês | MEDLINE | ID: mdl-31167558

RESUMO

Type 2 diabetes mellitus is a risk factor for incident heart failure and increases the risk of morbidity and mortality in patients with established disease. Secular trends in the prevalence of diabetes mellitus and heart failure forecast a growing burden of disease and underscore the need for effective therapeutic strategies. Recent clinical trials have demonstrated the shared pathophysiology between diabetes mellitus and heart failure, the synergistic effect of managing both conditions, and the potential for diabetes mellitus therapies to modulate the risk of heart failure outcomes. This scientific statement on diabetes mellitus and heart failure summarizes the epidemiology, pathophysiology, and impact of diabetes mellitus and its control on outcomes in heart failure; reviews the approach to pharmacological therapy and lifestyle modification in patients with diabetes mellitus and heart failure; highlights the value of multidisciplinary interventions to improve clinical outcomes in this population; and outlines priorities for future research.

13.
Hepatology ; 70(4): 1493-1494, 2019 Oct.
Artigo em Inglês | MEDLINE | ID: mdl-31152591
14.
J Card Fail ; 25(8): 584-619, 2019 Aug.
Artigo em Inglês | MEDLINE | ID: mdl-31174952

RESUMO

Type 2 diabetes mellitus is a risk factor for incident heart failure and increases the risk of morbidity and mortality in patients with established disease. Secular trends in the prevalence of diabetes mellitus and heart failure forecast a growing burden of disease and underscore the need for effective therapeutic strategies. Recent clinical trials have demonstrated the shared pathophysiology between diabetes mellitus and heart failure, the synergistic effect of managing both conditions, and the potential for diabetes mellitus therapies to modulate the risk of heart failure outcomes. This scientific statement on diabetes mellitus and heart failure summarizes the epidemiology, pathophysiology, and impact of diabetes mellitus and its control on outcomes in heart failure; reviews the approach to pharmacological therapy and lifestyle modification in patients with diabetes mellitus and heart failure; highlights the value of multidisciplinary interventions to improve clinical outcomes in this population; and outlines priorities for future research.

15.
Artigo em Inglês | MEDLINE | ID: mdl-31127823

RESUMO

CONTEXT: Although thyroid hormone replacement may improve outcomes in pregnant women with subclinical hypothyroidism (SCH), the extent to which they receive treatment is unknown. OBJECTIVE: To describe levothyroxine (LT4) treatment practices for pregnant women with SCH. DESIGN: Retrospective cohort study. SETTING: Large U.S. administrative claims database. PARTICIPANTS: Pregnant women with SCH defined by untreated TSH 2.5-10 mIU/L. MAIN OUTCOME MEASURE: Initiation of LT4 as a function of treating clinician specialty (endocrinology, obstetrics/gynecology, primary care, or other), baseline TSH, patient clinical and demographic factors, and U.S. region. RESULTS: We identified 7,990 pregnant women with SCH; only 1,214 (15.2%) received LT4. Treatment was significantly more likely in patients with higher TSH, obesity, recurrent pregnancy loss, thyroid disease, and cared by endocrinologists. Proportion of treated women increased over time; LT4 treatment was twice as likely in 2014 as in 2010. Women in Northeast and West U.S. were significantly more likely to receive LT4 compared to other regions. Asian women were more likely, while Hispanic women were less likely, to receive LT4 compared to White women. Endocrinologists started LT4 at lower TSH thresholds than other specialties, and treated women who were more likely to have had recurrent pregnancy loss and thyroid disease than women treated by other clinicians. CONCLUSIONS: We found large variation in the prescription of LT4 to pregnant women with SCH, though most treatment-eligible women remained untreated. Therapy initiation is associated with geographic, clinician, and patient characteristics. This evidence can inform quality improvement efforts to optimize care for pregnant women with SCH.

17.
J Am Med Dir Assoc ; 20(4): 444-450.e2, 2019 Apr.
Artigo em Inglês | MEDLINE | ID: mdl-30852170

RESUMO

OBJECTIVES: Patients discharged to a skilled nursing facility (SNF) for post-acute care have a high risk of hospital readmission. We aimed to develop and validate a risk-prediction model to prospectively quantify the risk of 30-day hospital readmission at the time of discharge to a SNF. DESIGN: Retrospective cohort study. SETTING: Ten independent SNFs affiliated with the post-acute care practice of an integrated health care delivery system. PARTICIPANTS: We evaluated 6032 patients who were discharged to SNFs for post-acute care after hospitalization. MEASUREMENTS: The primary outcome was all-cause 30-day hospital readmission. Patient demographics, medical comorbidity, prior use of health care, and clinical parameters during the index hospitalization were analyzed by using gradient boosting machine multivariable analysis to build a predictive model for 30-day hospital readmission. Area under the receiver operating characteristic curve (AUC) was assessed on out-of-sample observations under 10-fold cross-validation. RESULTS: Among 8616 discharges to SNFs from January 1, 2009, through June 30, 2014, a total of 1568 (18.2%) were readmitted to the hospital within 30 days. The 30-day hospital readmission prediction model had an AUC of 0.69, a 16% improvement over risk assessment using the Charlson Comorbidity Index alone. The final model included length of stay, abnormal laboratory parameters, and need for intensive care during the index hospitalization; comorbid status; and number of emergency department and hospital visits within the preceding 6 months. CONCLUSIONS AND IMPLICATIONS: We developed and validated a risk-prediction model for 30-day hospital readmission in patients discharged to a SNF for post-acute care. This prediction tool can be used to risk stratify the complex population of hospitalized patients who are discharged to SNFs to prioritize interventions and potentially improve the quality, safety, and cost-effectiveness of care.

19.
J Biomed Inform ; 89: 56-67, 2019 01.
Artigo em Inglês | MEDLINE | ID: mdl-30189255

RESUMO

Accurate and reliable prediction of clinical progression over time has the potential to improve the outcomes of chronic disease. The classical approach to analyzing longitudinal data is to use (generalized) linear mixed-effect models (GLMM). However, linear parametric models are predicated on assumptions, which are often difficult to verify. In contrast, data-driven machine learning methods can be applied to derive insight from the raw data without a priori assumptions. However, the underlying theory of most machine learning algorithms assume that the data is independent and identically distributed, making them inefficient for longitudinal supervised learning. In this study, we formulate an analytic framework, which integrates the random-effects structure of GLMM into non-linear machine learning models capable of exploiting temporal heterogeneous effects, sparse and varying-length patient characteristics inherent in longitudinal data. We applied the derived mixed-effect machine learning (MEml) framework to predict longitudinal change in glycemic control measured by hemoglobin A1c (HbA1c) among well controlled adults with type 2 diabetes. Results show that MEml is competitive with traditional GLMM, but substantially outperformed standard machine learning models that do not account for random-effects. Specifically, the accuracy of MEml in predicting glycemic change at the 1st, 2nd, 3rd, and 4th clinical visits in advanced was 1.04, 1.08, 1.11, and 1.14 times that of the gradient boosted model respectively, with similar results for the other methods. To further demonstrate the general applicability of MEml, a series of experiments were performed using real publicly available and synthetic data sets for accuracy and robustness. These experiments reinforced the superiority of MEml over the other methods. Overall, results from this study highlight the importance of modeling random-effects in machine learning approaches based on longitudinal data. Our MEml method is highly resistant to correlated data, readily accounts for random-effects, and predicts change of a longitudinal clinical outcome in real-world clinical settings with high accuracy.

20.
BMJ Open Diabetes Res Care ; 7(1): e000720, 2019.
Artigo em Inglês | MEDLINE | ID: mdl-31908790

RESUMO

Objective: Activin A, an inflammatory mediator implicated in cellular senescence-induced adipose tissue dysfunction and profibrotic kidney injury, may become a new target for the treatment of diabetic kidney disease (DKD) and chronic kidney diseases. We tested the hypothesis that human DKD-related injury leads to upregulation of activin A in blood and urine and in a human kidney cell model. We further hypothesized that circulating activin A parallels kidney injury markers in DKD. Research design and methods: In two adult diabetes cohorts and controls (Minnesota, USA; Galway, Ireland), the relationships between plasma (or urine) activin A, estimated glomerular filtration rate (eGFR) and DKD injury biomarkers were tested with logistic regression and correlation coefficients. Activin A, inflammatory, epithelial-mesenchymal-transition (EMT) and senescence markers were assayed in human kidney (HK-2) cells incubated in high glucose plus transforming growth factor-ß1 or albumin. Results: Plasma activin A levels were elevated in diabetes (n=206) compared with controls (n=76; 418.1 vs 259.3 pg/mL; p<0.001) and correlated inversely with eGFR (rs=-0.61; p<0.001; diabetes). After eGFR adjustment, only albuminuria (OR 1.56, 95% CI 1.16 to 2.09) and tumor necrosis factor receptor-1 (OR 6.40, 95% CI 1.08 to 38.00) associated with the highest activin tertile. Albuminuria also related to urinary activin (rs=0.65; p<0.001). Following in vitro HK-2 injury, activin, inflammatory, EMT genes and supernatant activin levels were increased. Conclusions: Circulating activin A is increased in human DKD and correlates with reduced kidney function and kidney injury markers. DKD-injured human renal tubule cells develop a profibrotic and inflammatory phenotype with activin A upregulation. These findings underscore the role of inflammation and provide a basis for further exploration of activin A as a diagnostic marker and therapeutic target in DKD.

SELEÇÃO DE REFERÊNCIAS
DETALHE DA PESQUISA