BACKGROUND Shoulder injury related to vaccine administration (SIRVA) accounts for more than half of all claims received by the National Vaccine Injury Compensation Program. However, due to the difficulty of finding SIRVA cases in large health care databases, population-based studies are scarce. OBJECTIVE The goal of the research was to develop a natural language processing (NLP) method to identify SIRVA cases from clinical notes. METHODS We conducted the study among members of a large integrated health care organization who were vaccinated between April 1, 2016, and December 31, 2017, and had subsequent diagnosis codes indicative of shoulder injury. Based on a training data set with a chart review reference standard of 164 cases, we developed an NLP algorithm to extract shoulder disorder information, including prior vaccination, anatomic location, temporality and causality. The algorithm identified 3 groups of positive SIRVA cases (definite, probable, and possible) based on the strength of evidence. We compared NLP results to a chart review reference standard of 100 vaccinated cases. We then applied the final automated NLP algorithm to a broader cohort of vaccinated persons with a shoulder injury diagnosis code and performed manual chart confirmation on a random sample of NLP-identified definite cases and all NLP-identified probable and possible cases. RESULTS In the validation sample, the NLP algorithm had 100% accuracy for identifying 4 SIRVA cases and 96 cases without SIRVA. In the broader cohort of 53,585 vaccinations, the NLP algorithm identified 291 definite, 124 probable, and 52 possible SIRVA cases. The chart-confirmation rates for these groups were 95.5% (278/291), 67.7% (84/124), and 17.3% (9/52), respectively. CONCLUSIONS The algorithm performed with high sensitivity and reasonable specificity in identifying positive SIRVA cases. The NLP algorithm can potentially be used in future population-based studies to identify this rare adverse event, avoiding labor-intensive chart review validation.
INTRODUCTION: There is currently no widely accepted approach to screening for pancreatic cancer (PC). We aimed to develop and validate a risk prediction model for pancreatic ductal adenocarcinoma (PDAC), the most common form of PC, across 2 health systems using electronic health records. METHODS: This retrospective cohort study consisted of patients aged 50–84 years having at least 1 clinic-based visit over a 10-year study period at Kaiser Permanente Southern California (model training, internal validation) and the Veterans Affairs (VA, external testing). Random survival forests models were built to identify the most relevant predictors from >500 variables and to predict risk of PDAC within 18 months of cohort entry. RESULTS: The Kaiser Permanente Southern California cohort consisted of 1.8 million patients (mean age 61.6) with 1,792 PDAC cases. The 18-month incidence rate of PDAC was 0.77 (95% confidence interval 0.73–0.80)/1,000 person-years. The final main model contained age, abdominal pain, weight change, HbA1c, and alanine transaminase change (c-index: mean = 0.77, SD = 0.02; calibration test: P value 0.4, SD 0.3). The final early detection model comprised the same features as those selected by the main model except for abdominal pain (c-index: 0.77 and SD 0.4; calibration test: P value 0.3 and SD 0.3). The VA testing cohort consisted of 2.7 million patients (mean age 66.1) with an 18-month incidence rate of 1.27 (1.23–1.30)/1,000 person-years. The recalibrated main and early detection models based on VA testing data sets achieved a mean c-index of 0.71 (SD 0.002) and 0.68 (SD 0.003), respectively. DISCUSSION: Using widely available parameters in electronic health records, we developed and externally validated parsimonious machine learning-based models for detection of PC. These models may be suitable for real-time clinical application.
Limited guidance exists regarding the optimal approach to management of pain in acute pancreatitis (AP).
Objectives
To investigate sources of variability in opioid use for treatment of acute pain in patients hospitalized for AP and to explore a potential association of opioid prescribing patterns with length of stay.
Design, Setting, and Participants
This retrospective cohort study included 4307 patients 18 years and older hospitalized for AP in a community-based integrated health care system, from January 1, 2008, to June 30, 2015. Analysis began in November 2017.
Exposures
Opioid use was quantified by morphine equivalent dose (MED).
Main Outcomes and Measures
Three analyses were performed: (1) factors associated with increased opioid administration during the initial 12 hours of hospitalization (baseline), (2) association of baseline opioid use with length of stay, and (3) frequency of opioid use 90 days after hospital discharge (persistent use).
Results
The cohort included 4307 patients (median [interquartile range] age, 57.4 [44.0-70.2] years; 2241 women [52.0%]) with AP. At baseline, 3443 patients (79.9%) received opioids, and 388 patients (9.6%) had persistent opioid use after discharge. After adjusting for pain and other clinical factors, women received less MED than men (adjusted event ratio, 0.83; 95% CI, 0.79-0.86;P < .001). Hispanic and Asian patients received less MED than non-Hispanic white patients (adjusted event ratio, 0.85; 95% CI, 0.81-0.90;P < .001; and adjusted event ratio, 0.79; 95% CI, 0.72-0.86;P < .001, respectively). Alcohol-related AP etiology was associated with increased MED vs gallstone disorders (adjusted event ratio, 1.11; 95% CI, 1.05-1.18;P < .001). Two of 13 hospitals administered significantly less opioids compared with the others. Median (interquartile range) length of stay was independently associated with MED at baseline, with 3.0 (2.1-4.5) days among patients not receiving opioids vs 5.0 (3.2-8.7) days among patients in the highest quintile of MED (P < .001).
Conclusions and Relevance
In addition to pain and disease severity, opioid use varied by etiology of AP, sex, race/ethnicity, and institution of treatment. Increased opioid use at baseline was associated with longer hospitalization. These findings suggest opportunities for improved approaches to pain control for patients with AP.
BACKGROUND Asthma-related symptoms are significant predictors of asthma exacerbation. Most of these symptoms are documented in clinical notes in free text format. Methods that can effectively capture the asthma-related symptoms from the unstructured data are lacking. OBJECTIVE The study aims to develop a natural language process (NLP) algorithm and process to identify symptoms associated with asthma from clinical notes within a large integrated healthcare system. METHODS We used unstructured data within two years prior to asthma diagnosis visits in 2013-2018 and 2021-2022 to identify four common asthma-related symptoms. Related terms and phrases were first compiled from publicly available resources and then recursively reviewed and enriched with inputs from clinicians and chart review. A rule-based NLP algorithm was first iteratively developed and refined via multiple rounds of chart review followed by adjudication, and then transformer-based deep learning algorithms were developed and validated using the same manually annotated datasets. Subsequently, a hybrid algorithm was generated by combining the rule-based and the transformer-based algorithms. Finally, the developed algorithms were implemented in all the study notes. RESULTS A total of 11,374,552 eligible study clinical notes with 128,211,793 sentences were retrieved. At least one symptom was identified in 1,663,450 (1.30%) sentences and 858,350 (7.55%) notes, respectively. Cough had the highest frequency at both sentence (1.07%) and note (5.81%) levels while chest tightness had the lowest one at both sentence (0.11%) and note (0.57%) levels. The frequencies of concomitant symptoms ranged from 0.03% to 0.38% at the sentence level and 0.10% to 1.85% at the note level. The validation of the hybrid algorithm against the annotated result of 1,600 clinical notes yielded a positive predictive value ranging from 96.53% (wheezing) to 97.42% (chest tightness) at the sentence level and 96.76% (wheezing) to 97.42% (chest tightness) at the note level, sensitivity ranged from 93.90% (dyspnea) to 95.95% (cough) at the sentence level and 96.00% (chest tightness) to 99.07% (cough) at the note level. The corresponding F1 scores of all four symptoms were > 0.95 at both sentence and note levels regardless of NLP algorithms. CONCLUSIONS The developed NLP algorithms could effectively capture asthma-related symptoms from unstructured notes. These algorithms could be utilized to examine asthma burden and prediction of asthma exacerbation.
To describe the relationships between persistent asthma defined by administrative versus survey data and their stability over time.Longitudinal survey and retrospective administrative database.Administrative data were used to identify patients meeting the Healthcare Effectiveness Data and Information Set (HEDIS) criteria for persistent asthma in year 1 (2006). At the end of year 2 and on 3 occasions during year 3, patients were mailed a survey to define persistent asthma based on symptoms and medication use in the prior month and exacerbations in the prior 12 months. Administrative data were also used to define medical utilization for asthma in year 3.Of 13,833 eligible patients, 2895 (20.9%) returned the survey; 2751 of these respondents reported physician-diagnosed asthma, of whom 2517 (91.5%) had survey-defined persistent asthma. Patients having survey-defined persistent asthma (68.0%) were more likely to requalify as having HEDIS-defined persistent asthma in year 2 than patients not having survey-defined persistent asthma (22.2%). However, 81.6% of survey respondents who did not requalify as having HEDIS-defined persistent asthma in year 2 had survey-defined persistent asthma. Patients with survey-defined persistent asthma in year 2 had significantly more medical utilization for asthma in year 3 than patients without survey-defined persistent asthma. Approximately 82% of the 799 patients completing all 4 surveys had persistent asthma on all surveys.HEDIS-defined persistent asthma is generally consistent with survey-defined persistent asthma. Persistent asthma usually remains persistent over a 3-year period, indicating that it is a stable characteristic of asthma for most patients. The low survey response rate suggests that further population-based studies will be necessary to confirm the validity and generalizability of our study findings regarding persistent asthma.
To examine the association between chorioamnionitis and childhood asthma based on gestational age at birth and race/ethnicity.
Design
A retrospective cohort study using the Kaiser Permanente Southern California (KPSC) Matched Perinatal records.
Setting
Kaiser Permanente Southern California, Pasadena, California.
Participants
All singleton children born in KPSC hospitals between 1991 and 2007 (N = 510 216).
Main Exposure
Clinically diagnosed chorioamnionitis.
Main Outcome Measures
Physician-diagnosed asthma in children aged 8 years or younger.
Results
The incidence rates of asthma among preterm- and full term–born children of pregnancies complicated by chorioamnionitis were 100.7 and 39.6 per 1000 person-years, respectively (incidence rate ratio, 2.9; 95% confidence interval [CI], 2.6-3.3). Children aged 8 years or younger with asthma were more likely to be born to women who were aged 35 years or older, African American, had 13 or more years of education, had maternal asthma, used antibiotics, had chorioamnionitis during the pregnancy, and had a male child. Multivariable Cox regression analysis revealed that children born at 23 to 28, 29 to 33, and 34 to 36 weeks' gestation after pregnancies complicated by chorioamnionitis had a 1.23-fold (95% CI, 1.02-1.49), 1.51-fold (95% CI, 1.26-1.80), and 1.20-fold (95% CI, 1.03-1.47), respectively, increased risk of asthma compared with children of similar gestational age born after pregnancies not complicated by chorioamnionitis. A preterm pregnancy complicated by chorioamnionitis was associated with increased risk of asthma among white (hazard ratio [HR], 1.66; 95% CI, 1.32-2.07), African American (HR, 1.98; 95% CI, 1.60-2.44), and Hispanic (HR, 1.70; 95% CI, 1.45-2.00), but not Asian/Pacific Islander (HR, 1.15; 95% CI, 0.83-1.58) women.
Conclusion
Findings suggest that chorioamnionitis at preterm gestation is independently associated with increased risk of childhood asthma.
