Dataset Information

Sentiment Measured in Hospital Discharge Notes Is Associated with Readmission and Mortality Risk: An Electronic Health Record Study.

ABSTRACT: Natural language processing tools allow the characterization of sentiment--that is, terms expressing positive and negative emotion--in text. Applying such tools to electronic health records may provide insight into meaningful patient or clinician features not captured in coded data alone. We performed sentiment analysis on 2,484 hospital discharge notes for 2,010 individuals from a psychiatric inpatient unit, as well as 20,859 hospital discharges for 15,011 individuals from general medical units, in a large New England health system between January 2011 and 2014. The primary measures of sentiment captured intensity of subjective positive or negative sentiment expressed in the discharge notes. Mean scores were contrasted between sociodemographic and clinical groups in mixed effects regression models. Discharge note sentiment was then examined for association with risk for readmission in Cox regression models. Discharge notes for individuals with greater medical comorbidity were modestly but significantly lower in positive sentiment among both psychiatric and general medical cohorts (p<0.001 in each). Greater positive sentiment at discharge was associated with significantly decreased risk of hospital readmission in each cohort (~12% decrease per standard deviation above the mean). Automated characterization of discharge notes in terms of sentiment identifies differences between sociodemographic groups, as well as in clinical outcomes, and is not explained by differences in diagnosis. Clinician sentiment merits investigation to understand why and how it reflects or impacts outcomes.

SUBMITTER: McCoy TH

PROVIDER: S-EPMC4547711 | biostudies-literature | 2015

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Sentiment Measured in Hospital Discharge Notes Is Associated with Readmission and Mortality Risk: An Electronic Health Record Study.

McCoy Thomas H TH Castro Victor M VM Cagan Andrew A Roberson Ashlee M AM Kohane Isaac S IS Perlis Roy H RH

PloS one 20150824 8

Natural language processing tools allow the characterization of sentiment--that is, terms expressing positive and negative emotion--in text. Applying such tools to electronic health records may provide insight into meaningful patient or clinician features not captured in coded data alone. We performed sentiment analysis on 2,484 hospital discharge notes for 2,010 individuals from a psychiatric inpatient unit, as well as 20,859 hospital discharges for 15,011 individuals from general medical units ...[more]

PMID: 26302085

Similar Datasets

Project description:ImportanceThe Affordable Care Act has led to US national reductions in hospital 30-day readmission rates for heart failure (HF), acute myocardial infarction (AMI), and pneumonia. Whether readmission reductions have had the unintended consequence of increasing mortality after hospitalization is unknown.ObjectiveTo examine the correlation of paired trends in hospital 30-day readmission rates and hospital 30-day mortality rates after discharge.Design, setting, and participantsRetrospective study of Medicare fee-for-service beneficiaries aged 65 years or older hospitalized with HF, AMI, or pneumonia from January 1, 2008, through December 31, 2014.ExposureThirty-day risk-adjusted readmission rate (RARR).Main outcomes and measuresThirty-day RARRs and 30-day risk-adjusted mortality rates (RAMRs) after discharge were calculated for each condition in each month at each hospital in 2008 through 2014. Monthly trends in each hospital's 30-day RARRs and 30-day RAMRs after discharge were examined for each condition. The weighted Pearson correlation coefficient was calculated for hospitals' paired monthly trends in 30-day RARRs and 30-day RAMRs after discharge for each condition.ResultsIn 2008 through 2014, 2 962 554 hospitalizations for HF, 1 229 939 for AMI, and 2 544 530 for pneumonia were identified at 5016, 4772, and 5057 hospitals, respectively. In January 2008, mean hospital 30-day RARRs and 30-day RAMRs after discharge were 24.6% and 8.4% for HF, 19.3% and 7.6% for AMI, and 18.3% and 8.5% for pneumonia. Hospital 30-day RARRs declined in the aggregate across hospitals from 2008 through 2014; monthly changes in RARRs were -0.053% (95% CI, -0.055% to -0.051%) for HF, -0.044% (95% CI, -0.047% to -0.041%) for AMI, and -0.033% (95% CI, -0.035% to -0.031%) for pneumonia. In contrast, monthly aggregate changes across hospitals in hospital 30-day RAMRs after discharge varied by condition: HF, 0.008% (95% CI, 0.007% to 0.010%); AMI, -0.003% (95% CI, -0.005% to -0.001%); and pneumonia, 0.001% (95% CI, -0.001% to 0.003%). However, correlation coefficients in hospitals' paired monthly changes in 30-day RARRs and 30-day RAMRs after discharge were weakly positive: HF, 0.066 (95% CI, 0.036 to 0.096); AMI, 0.067 (95% CI, 0.027 to 0.106); and pneumonia, 0.108 (95% CI, 0.079 to 0.137). Findings were similar in secondary analyses, including with alternate definitions of hospital mortality.Conclusions and relevanceAmong Medicare fee-for-service beneficiaries hospitalized for heart failure, acute myocardial infarction, or pneumonia, reductions in hospital 30-day readmission rates were weakly but significantly correlated with reductions in hospital 30-day mortality rates after discharge. These findings do not support increasing postdischarge mortality related to reducing hospital readmissions.

Project description:ImportanceHospital readmissions contribute to higher expenditures and may sometimes reflect suboptimal patient care. Individuals discharged against medical advice (AMA) are a vulnerable patient population and may have higher risk for readmission.ObjectivesTo determine odds of readmission and mortality for patients discharged AMA vs all others, to characterize patient and hospital-level factors associated with readmissions, and to quantify their overall cost burden.Design, setting, and participantsNationally representative, all-payer cohort study using the 2014 National Readmissions Database. Eligible index admissions were nonobstetrical/newborn hospitalizations for patients 18 years and older discharged between January 2014 and November 2014. Admissions were excluded if there was a missing primary diagnosis, discharge disposition, length of stay, or if the patient died during that hospitalization. Data were analyzed between January 2018 and June 2018.ExposuresDischarge AMA and non-AMA discharge.Main outcomes and measuresThirty-day all-cause readmission and in-hospital mortality rate.ResultsThere were 19.9 million weighted index admissions, of which 1.5% resulted in an AMA discharge. Within the AMA cohort, 85% were younger than 65 years, 63% were male, 55% had Medicaid or other (including uninsured) coverage, and 39% were in the lowest income quartile. Thirty-day all-cause readmission was 21.0% vs 11.9% for AMA vs non-AMA discharge (P < .001), and 30-day in-hospital mortality was 2.5% vs 5.6% (P < .001), respectively. Individuals discharged AMA were more likely to be readmitted to a different hospital compared with non-AMA patients (43.0% vs 23.9%; P < .001). Of all 30-day readmissions, 19.0% occurred within the first day after AMA discharge vs 6.1% for non-AMA patients (P < .001). On multivariable regression, AMA discharge was associated with a 2.01 (95% CI, 1.97-2.05) increased adjusted odds of readmission and a 0.80 (95% CI, 0.74-0.87) decreased adjusted odds of in-hospital mortality compared with non-AMA discharge. Nationwide readmissions after AMA discharge accounted for more than 400 000 inpatient hospitalization days at a total cost of $822 million in 2014.Conclusions and relevanceIndividuals discharged AMA have higher odds of 30-day readmission at significant cost to the health care system and lower in-hospital mortality rates compared with non-AMA patients. Patients discharged AMA are also more likely to be readmitted to different hospitals and to have earlier bounce-back readmissions, which may reflect dissatisfaction with their initial episode of care.

Project description:BackgroundObtaining accurate clinical information about recent acute care visits is extremely important for outpatient providers. However, documents used to communicate this information are often difficult to use. This puts patients at risk of adverse events. Elderly patients who are seen by more providers and have more care transitions are especially vulnerable.ObjectiveThis study aimed to (1) identify the information about elderly patients' recent acute care visits needed to coordinate their care, (2) use this information to assess discharge summaries, and (3) provide recommendations to help improve the quality of electronic health record (EHR)-generated discharge summaries, thereby increasing patient safety.MethodsA literature review, clinician interviews, and a survey of outpatient providers were used to identify and categorize data needed to coordinate care for recently discharged elderly patients. Based upon those data, 2 guidelines for creating useful discharge summaries were created. The new guidelines, along with 17 previously developed medical documentation usability heuristics, were applied to assess 4 simulated elderly patient discharge summaries.ResultsThe initial research effort yielded a list of 29 items that should always be included in elderly patient discharge summaries and a list of 7 "helpful, but not always necessary" items. Evaluation of 4 deidentified elderly patient discharge summaries revealed that none of the documents contained all 36 necessary items; between 14 and 18 were missing. The documents each had several other issues, and they differed significantly in organization, layout, and formatting.ConclusionsVariations in content and structure of discharge summaries in the United States make them unnecessarily difficult to use. Standardization would benefit both patients, by lowering the risk of care transition-related adverse events, and outpatient providers, by helping reduce frustration that can contribute to burnout. In the short term, acute care providers can help improve the quality of their discharge summaries by working with EHR vendors to follow recommendations based upon this study. Meanwhile, additional human factors work should determine the most effective way to organize and present information in discharge summaries, to facilitate effective standardization.

Dataset Information

Sentiment Measured in Hospital Discharge Notes Is Associated with Readmission and Mortality Risk: An Electronic Health Record Study.

Publications

Sentiment Measured in Hospital Discharge Notes Is Associated with Readmission and Mortality Risk: An Electronic Health Record Study.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets