Restricted use of copy and paste in electronic health records potentially improves healthcare quality.

ABSTRACT: The copy-and-paste feature is commonly used for clinical documentation, and a policy is needed to reduce overdocumentation. We aimed to determine if the restricted use of copy and paste by doctors could improve inpatient healthcare quality.Clinical documentation in an inpatient dataset compiled from 2016 to 2018 was used. Copied-and-pasted text was detected in word templates using natural language programming with a threshold of 70%. The prevalence of copying and pasting after the policy introduction was accessed by segmented regression for trend analysis. The rate of readmission for the same disease within 14 days was assessed to evaluate inpatient healthcare quality, and the completion of discharge summary notes within 3 days was assessed to determine the timeliness of note completion. The relationships between these factors were used cross-correlation to detect lag effect. Poisson regression was performed to identify the relative effect of the copy and paste restriction policy on the 14-day readmission rate or the discharge note completion rate within 3 days.The prevalence of copying and pasting initially decreased, then increased, and then flatly decreased. The cross-correlation results showed a significant correlation between the prevalence of copied-and-pasted text and the 14-day readmission rate (P < .001) and a relative risk of 1.105 (P < .005), with a one-month lag. The discharge note completion rate initially decreased and not affected long term after restriction policy.Appropriate policies to restrict the use of copying and pasting can lead to improvements in inpatient healthcare quality. Prospective research with cost analysis is needed.

Association between Electronic Medical Records and Healthcare Quality.

The implementation of electronic medical records (EMRs) has generally been thought to improve medical efficiency and safety, but consistent evidence of improved healthcare quality due to EMRs in population-based studies is lacking. We assessed the relationship between the degree of EMR adoption and patient outcomes.We performed an observational study using discharge data from Tri-service General Hospital from 2013 to 2018. The levels of EMR utilization were divided into no EMRs, partial EMRs and full EMRs. The primary healthcare quality indicators were inpatient mortality, readmission within 14 days, and 48-hour postoperative mortality. We performed a Cox proportional hazards regression analysis to evaluate the relationship between the EMR utilization level and healthcare quality.In total, 262,569 patients were included in this study. Compared with no EMRs, full EMR implementation led to lower inpatient mortality [adjusted hazard ratio (HR) 0.947, 95% confidence interval (CI): 0.897-0.999, P = ..049] and a lower risk of readmission within 14 days (adjusted HR 0.627, 95% CI: 0.577-0.681, P < .001). Full EMR implementation was associated was a lower risk of 48-hour postoperative mortality (adjusted HR 0.372, 95% CI: 0.208-0.665, P = .001) than no EMRs. Partial EMR implementation was associated with a higher risk of readmission within 14 days than no EMRs (HR 1.387, 95% CI: 1.298-1.485, P < .001).Full EMR adoption improves healthcare quality in medical institutions treating severely ill patients. A prospective study is needed to confirm this finding.

Projection Word Embedding Model With Hybrid Sampling Training for Classifying ICD-10-CM Codes: Longitudinal Observational Study.

BACKGROUND: Most current state-of-the-art models for searching the International Classification of Diseases, Tenth Revision Clinical Modification (ICD-10-CM) codes use word embedding technology to capture useful semantic properties. However, they are limited by the quality of initial word embeddings. Word embedding trained by electronic health records (EHRs) is considered the best, but the vocabulary diversity is limited by previous medical records. Thus, we require a word embedding model that maintains the vocabulary diversity of open internet databases and the medical terminology understanding of EHRs. Moreover, we need to consider the particularity of the disease classification, wherein discharge notes present only positive disease descriptions. OBJECTIVE: We aimed to propose a projection word2vec model and a hybrid sampling method. In addition, we aimed to conduct a series of experiments to validate the effectiveness of these methods. METHODS: We compared the projection word2vec model and traditional word2vec model using two corpora sources: English Wikipedia and PubMed journal abstracts. We used seven published datasets to measure the medical semantic understanding of the word2vec models and used these embeddings to identify the three-character-level ICD-10-CM diagnostic codes in a set of discharge notes. On the basis of embedding technology improvement, we also tried to apply the hybrid sampling method to improve accuracy. The 94,483 labeled discharge notes from the Tri-Service General Hospital of Taipei, Taiwan, from June 1, 2015, to June 30, 2017, were used. To evaluate the model performance, 24,762 discharge notes from July 1, 2017, to December 31, 2017, from the same hospital were used. Moreover, 74,324 additional discharge notes collected from seven other hospitals were tested. The F-measure, which is the major global measure of effectiveness, was adopted. RESULTS: In medical semantic understanding, the original EHR embeddings and PubMed embeddings exhibited superior performance to the original Wikipedia embeddings. After projection training technology was applied, the projection Wikipedia embeddings exhibited an obvious improvement but did not reach the level of original EHR embeddings or PubMed embeddings. In the subsequent ICD-10-CM coding experiment, the model that used both projection PubMed and Wikipedia embeddings had the highest testing mean F-measure (0.7362 and 0.6693 in Tri-Service General Hospital and the seven other hospitals, respectively). Moreover, the hybrid sampling method was found to improve the model performance (F-measure=0.7371/0.6698). CONCLUSIONS: The word embeddings trained using EHR and PubMed could understand medical semantics better, and the proposed projection word2vec model improved the ability of medical semantics extraction in Wikipedia embeddings. Although the improvement from the projection word2vec model in the real ICD-10-CM coding task was not substantial, the models could effectively handle emerging diseases. The proposed hybrid sampling method enables the model to behave like a human expert.

Patterns of perspectives on fall-prevention beliefs by community-dwelling older adults: a Q method investigation.

BACKGROUND: Falling has high incidence and reoccurrence rates and is an essential factor contributing to accidental injury or death for older adults. Enhancing the participation of community-dwelling older adults in fall-prevention programs is crucial. Understanding fall-prevention beliefs will be beneficial for developing a community-based fall-prevention program. The aim of the present study was to identify the distinct types of subjective views on the fall-prevention beliefs of community-dwelling older adults aged 80 years and older by applying the Q method. METHODS: The Q method was adopted to investigate the pattern of perception on fall-prevention beliefs. Forty-two older adults aged 80 - 92 years from a community care center in Northern Taiwan were recruited and requested to complete a Q-sorting. A series of Q-sorts was performed by the participants to rank 30 statements into a normal distribution Q-sort grid. The Q-sorts were subjected to principal component analysis by using PQMethod software Version 2.35. RESULTS: Four statistically independent perspectives were derived from the analysis and reflected distinct viewpoints on beliefs related to fall prevention. Participants in the Considerate perspective believed that health problems caused by falling were serious and fall prevention could decrease the burden they place on their family. Participants in the Promising perspective believed that existing health problems could cause a fall and that fall prevention contributed to their well-being. Participants in the Adaptable perspective perceived low barriers to execute fall prevention and displayed self-confidence and independence in preventing falls. Participants in the Ignorance perspective believed that they could not prevent falls and perceived barriers to fall prevention. CONCLUSIONS: By combining theoretical constructs and the Q methodology approach, this study identified four distinct perspectives on fall prevention among community-dwelling older adults. Critical reflection on older adult personal perspectives and interpretations of the required responsive approach is a key element for appropriating fall-prevention support.