Asymmetry in noise-induced hearing loss: evaluation of two competing theories.

Competing theories exist about why asymmetry is observed in noise-induced hearing loss (NIHL). We evaluated these theories using a cohort of young workers studied over 16 years. The study aim was to describe and evaluate patterns of hearing loss and asymmetry by gender, agricultural exposure and gunfire exposure. This was a secondary analysis of data collected from young adults during follow-up of a randomized controlled trial. This follow-up study evaluated long-term effects of a hearing conservation intervention for rural students. The sample consisted of 392 of 690 participants from the original trial. In total, 355 young adults (aged 29-33 years) completed baseline and follow-up noise exposure surveys and clinical audiometric examinations. Data are displayed graphically as thresholds by frequency and ear and degree of asymmetry between ears (left minus right). In the primary group comparisons, low and high frequency averages and mean high frequency asymmetry were analyzed using mixed linear models. At frequencies >2000 Hz, men showed more hearing loss, with greater asymmetry and a different asymmetry pattern, than women. For men with documented hearing loss, there was a trend toward increasing asymmetry with increasing levels of hearing loss. Asymmetry at high frequencies varied substantially by level of shooting exposure. While "head shadowing" is accepted as the primary explanation for asymmetric hearing loss in the audiologic and related public health literature, our findings are more consistent with physiological differences as the primary cause of asymmetric hearing loss, with greater susceptibility to NIHL in the left ear of men.

Determinants of early-stage hearing loss among a cohort of young workers with 16-year follow-up.

OBJECTIVES: The authors had a unique opportunity to study the early impacts of occupational and recreational exposures on the development of noise-induced hearing loss (NIHL) in a cohort of 392 young workers. The objectives of this study were to estimate strength of associations between occupational and recreational exposures and occurrence of early-stage NIHL and to determine the extent to which relationships between specific noise exposures and early-stage NIHL were mitigated through the use of hearing protection. METHODS: Participants were young adults who agreed to participate in a follow-up of a randomised controlled trial. While the follow-up study was designed to observe long-term effects (up to 16 years) of a hearing conservation intervention for high school students, it also provided opportunity to study the potential aetiology of NIHL in this worker cohort. Study data were collected via exposure history questionnaires and clinical audiometric examinations. RESULTS: Over the 16-year study period, the authors documented changes to hearing acuity that exceeded 15 dB at high frequencies in 42.8% of men and 27.7% of women. Analyses of risk factors for NIHL were limited to men, who comprised 68% of the cohort, and showed that risks increased in association with higher levels of the most common recreational and occupational noise sources, as well as chemical exposures with ototoxic potential. Use of hearing protection and other safety measures, although not universal and sometimes modest, appeared to offer some protection. CONCLUSIONS: Early-stage NIHL can be detected in young workers by measuring high-frequency changes in hearing acuity. Hearing conservation programmes should focus on a broader range of exposures, whether in occupational or non-occupational settings. Priority exposures include gunshots, chainsaws, power tools, smoking and potentially some chemical exposures.

Accuracy of Electronic Medical Record Follow-Up Data for Estimating the Survival Time of Patients With Cancer.

PURPOSE: For real-world evidence, it is convenient to use routinely collected data from the electronic medical record (EMR) to measure survival outcomes. However, patients can become lost to follow-up, causing incomplete data and biased survival time estimates. We quantified this issue for patients with metastatic cancer seen in an academic health system by comparing survival estimates from EMR data only and from EMR data combined with high-quality cancer registry data. MATERIALS AND METHODS: Patients diagnosed with metastatic cancer from 2008 to 2014 were included in this retrospective study. Patients who were diagnosed with cancer or received their initial treatment within our system were included in the institutional cancer registry and this study. Overall survival was calculated using the Kaplan-Meier method. Survival curves were generated in two ways: using EMR follow-up data alone and using EMR data supplemented with data from the Stanford Cancer Registry/California Cancer Registry. RESULTS: Four thousand seventy-seven patients were included. The median follow-up using EMR + Cancer Registry data was 19.9 months, and the median follow-up in surviving patients was 67.6 months. There were 1,301 deaths recorded in the EMR and 3,140 deaths recorded in the Cancer Registry. The median overall survival from the date of cancer diagnosis using EMR data was 58.7 months (95% CI, 54.2 to 63.2); using EMR + Cancer Registry data, it was 20.8 months (95% CI, 19.6 to 22.3). A similar pattern was seen using the date of first systemic therapy or date of first hospital admission as the baseline date. CONCLUSION: Using EMR data alone, survival time was overestimated compared with EMR + Cancer Registry data.

Natural Language Processing to Identify Cancer Treatments With Electronic Medical Records.

PURPOSE: Knowing the treatments administered to patients with cancer is important for treatment planning and correlating treatment patterns with outcomes for personalized medicine study. However, existing methods to identify treatments are often lacking. We develop a natural language processing approach with structured electronic medical records and unstructured clinical notes to identify the initial treatment administered to patients with cancer. METHODS: We used a total number of 4,412 patients with 483,782 clinical notes from the Stanford Cancer Institute Research Database containing patients with nonmetastatic prostate, oropharynx, and esophagus cancer. We trained treatment identification models for each cancer type separately and compared performance of using only structured, only unstructured (bag-of-words, doc2vec, fasttext), and combinations of both (structured + bow, structured + doc2vec, structured + fasttext). We optimized the identification model among five machine learning methods (logistic regression, multilayer perceptrons, random forest, support vector machines, and stochastic gradient boosting). The treatment information recorded in the cancer registry is the gold standard and compares our methods to an identification baseline with billing codes. RESULTS: For prostate cancer, we achieved an f1-score of 0.99 (95% CI, 0.97 to 1.00) for radiation and 1.00 (95% CI, 0.99 to 1.00) for surgery using structured + doc2vec. For oropharynx cancer, we achieved an f1-score of 0.78 (95% CI, 0.58 to 0.93) for chemoradiation and 0.83 (95% CI, 0.69 to 0.95) for surgery using doc2vec. For esophagus cancer, we achieved an f1-score of 1.0 (95% CI, 1.0 to 1.0) for both chemoradiation and surgery using all combinations of structured and unstructured data. We found that employing the free-text clinical notes outperforms using the billing codes or only structured data for all three cancer types. CONCLUSION: Our results show that treatment identification using free-text clinical notes greatly improves upon the performance using billing codes and simple structured data. The approach can be used for treatment cohort identification and adapted for longitudinal cancer treatment identification.

Automated model versus treating physician for predicting survival time of patients with metastatic cancer.

OBJECTIVE: Being able to predict a patient's life expectancy can help doctors and patients prioritize treatments and supportive care. For predicting life expectancy, physicians have been shown to outperform traditional models that use only a few predictor variables. It is possible that a machine learning model that uses many predictor variables and diverse data sources from the electronic medical record can improve on physicians' performance. For patients with metastatic cancer, we compared accuracy of life expectancy predictions by the treating physician, a machine learning model, and a traditional model. MATERIALS AND METHODS: A machine learning model was trained using 14 600 metastatic cancer patients' data to predict each patient's distribution of survival time. Data sources included note text, laboratory values, and vital signs. From 2015-2016, 899 patients receiving radiotherapy for metastatic cancer were enrolled in a study in which their radiation oncologist estimated life expectancy. Survival predictions were also made by the machine learning model and a traditional model using only performance status. Performance was assessed with area under the curve for 1-year survival and calibration plots. RESULTS: The radiotherapy study included 1190 treatment courses in 899 patients. A total of 879 treatment courses in 685 patients were included in this analysis. Median overall survival was 11.7 months. Physicians, machine learning model, and traditional model had area under the curve for 1-year survival of 0.72 (95% CI 0.63-0.81), 0.77 (0.73-0.81), and 0.68 (0.65-0.71), respectively. CONCLUSIONS: The machine learning model's predictions were more accurate than those of the treating physician or a traditional model.

Hearing conservation program for agricultural students: short-term outcomes from a cluster-randomized trial with planned long-term follow-up.

OBJECTIVES: (1) To conduct a contemporary analysis of historical data on short-term efficacy of a 3-year hearing conservation program conducted from 1992 to 1996 in Wisconsin, USA, with 753 high school students actively involved in farm work; (2) to establish procedures for assessment of hearing loss for use in a recently funded follow-up of this same hearing conservation program cohort. METHODS: We analyzed a pragmatic cluster-randomized controlled trial, with schools as the unit of randomization. Thirty-four rural schools were recruited and randomized to intervention or control. The intervention included classroom instruction, distribution of hearing protection devices, direct mailings, noise level assessments, and yearly audiometric testing. The control group received the audiometric testing. RESULTS: Students exposed to the hearing conservation program reported more frequent use of hearing protection devices, but there was no evidence of reduced levels of noise-induced hearing loss (NIHL). CONCLUSION: Our analysis suggests that, since NIHL is cumulative, a 3-year study was likely not long enough to evaluate the efficacy of this intervention. While improvements in reported use of hearing protection devices were noted, the lasting impact of these behaviors is unknown and the finding merits corroboration by longer term objective hearing tests. A follow-up study of the cohort has recently been started.

Automated Survival Prediction in Metastatic Cancer Patients Using High-Dimensional Electronic Medical Record Data.

BACKGROUND: Oncologists use patients' life expectancy to guide decisions and may benefit from a tool that accurately predicts prognosis. Existing prognostic models generally use only a few predictor variables. We used an electronic medical record dataset to train a prognostic model for patients with metastatic cancer. METHODS: The model was trained and tested using 12 588 patients treated for metastatic cancer in the Stanford Health Care system from 2008 to 2017. Data sources included provider note text, labs, vital signs, procedures, medication orders, and diagnosis codes. Patients were divided randomly into a training set used to fit the model coefficients and a test set used to evaluate model performance (80%/20% split). A regularized Cox model with 4126 predictor variables was used. A landmarking approach was used due to the multiple observations per patient, with t0 set to the time of metastatic cancer diagnosis. Performance was also evaluated using 399 palliative radiation courses in test set patients. RESULTS: The C-index for overall survival was 0.786 in the test set (averaged across landmark times). For palliative radiation courses, the C-index was 0.745 (95% confidence interval [CI] = 0.715 to 0.775) compared with 0.635 (95% CI = 0.601 to 0.669) for a published model using performance status, primary tumor site, and treated site (two-sided P < .001). Our model's predictions were well-calibrated. CONCLUSIONS: The model showed high predictive performance, which will need to be validated using external data. Because it is fully automated, the model can be used to examine providers' practice patterns and could be deployed in a decision support tool to help improve quality of care.

New models and online calculator for predicting non-sentinel lymph node status in sentinel lymph node positive breast cancer patients.

BACKGROUND: Current practice is to perform a completion axillary lymph node dissection (ALND) for breast cancer patients with tumor-involved sentinel lymph nodes (SLNs), although fewer than half will have non-sentinel node (NSLN) metastasis. Our goal was to develop new models to quantify the risk of NSLN metastasis in SLN-positive patients and to compare predictive capabilities to another widely used model. METHODS: We constructed three models to predict NSLN status: recursive partitioning with receiver operating characteristic curves (RP-ROC), boosted Classification and Regression Trees (CART), and multivariate logistic regression (MLR) informed by CART. Data were compiled from a multicenter Northern California and Oregon database of 784 patients who prospectively underwent SLN biopsy and completion ALND. We compared the predictive abilities of our best model and the Memorial Sloan-Kettering Breast Cancer Nomogram (Nomogram) in our dataset and an independent dataset from Northwestern University. RESULTS: 285 patients had positive SLNs, of which 213 had known angiolymphatic invasion status and 171 had complete pathologic data including hormone receptor status. 264 (93%) patients had limited SLN disease (micrometastasis, 70%, or isolated tumor cells, 23%). 101 (35%) of all SLN-positive patients had tumor-involved NSLNs. Three variables (tumor size, angiolymphatic invasion, and SLN metastasis size) predicted risk in all our models. RP-ROC and boosted CART stratified patients into four risk levels. MLR informed by CART was most accurate. Using two composite predictors calculated from three variables, MLR informed by CART was more accurate than the Nomogram computed using eight predictors. In our dataset, area under ROC curve (AUC) was 0.83/0.85 for MLR (n = 213/n = 171) and 0.77 for Nomogram (n = 171). When applied to an independent dataset (n = 77), AUC was 0.74 for our model and 0.62 for Nomogram. The composite predictors in our model were the product of angiolymphatic invasion and size of SLN metastasis, and the product of tumor size and square of SLN metastasis size. CONCLUSION: We present a new model developed from a community-based SLN database that uses only three rather than eight variables to achieve higher accuracy than the Nomogram for predicting NSLN status in two different datasets.

Probabilistic Prognostic Estimates of Survival in Metastatic Cancer Patients (PPES-Met) Utilizing Free-Text Clinical Narratives.

We propose a deep learning model - Probabilistic Prognostic Estimates of Survival in Metastatic Cancer Patients (PPES-Met) for estimating short-term life expectancy (>3 months) of the patients by analyzing free-text clinical notes in the electronic medical record, while maintaining the temporal visit sequence. In a single framework, we integrated semantic data mapping and neural embedding technique to produce a text processing method that extracts relevant information from heterogeneous types of clinical notes in an unsupervised manner, and we designed a recurrent neural network to model the temporal dependency of the patient visits. The model was trained on a large dataset (10,293 patients) and validated on a separated dataset (1818 patients). Our method achieved an area under the ROC curve (AUC) of 0.89. To provide explain-ability, we developed an interactive graphical tool that may improve physician understanding of the basis for the model's predictions. The high accuracy and explain-ability of the PPES-Met model may enable our model to be used as a decision support tool to personalize metastatic cancer treatment and provide valuable assistance to the physicians.

Dynamin impacts homology-directed repair and breast cancer response to chemotherapy.

After the initial responsiveness of triple-negative breast cancers (TNBCs) to chemotherapy, they often recur as chemotherapy-resistant tumors, and this has been associated with upregulated homology-directed repair (HDR). Thus, inhibitors of HDR could be a useful adjunct to chemotherapy treatment of these cancers. We performed a high-throughput chemical screen for inhibitors of HDR from which we obtained a number of hits that disrupted microtubule dynamics. We postulated that high levels of the target molecules of our screen in tumors would correlate with poor chemotherapy response. We found that inhibition or knockdown of dynamin 2 (DNM2), known for its role in endocytic cell trafficking and microtubule dynamics, impaired HDR and improved response to chemotherapy of cells and of tumors in mice. In a retrospective analysis, levels of DNM2 at the time of treatment strongly predicted chemotherapy outcome for estrogen receptor-negative and especially for TNBC patients. We propose that DNM2-associated DNA repair enzyme trafficking is important for HDR efficiency and is a powerful predictor of sensitivity to breast cancer chemotherapy and an important target for therapy.

Increased Risk of Cutaneous Squamous Cell Carcinoma After Vismodegib Therapy for Basal Cell Carcinoma.

IMPORTANCE: Smoothened inhibitors (SIs) are a new type of targeted therapy for advanced basal cell carcinoma (BCC), and their long-term effects, such as increased risk of subsequent malignancy, are still being explored. OBJECTIVE: To evaluate the risk of developing a non-BCC malignancy after SI exposure in patients with BCC. DESIGN, SETTING, AND PARTICIPANTS: A case-control study at Stanford Medical Center, an academic hospital. Participants were higher-risk patients with BCC diagnosed from January 1, 1998, to December 31, 2014. The dates of the analysis were January 1 to November 1, 2015. EXPOSURES: The exposed participants (cases) comprised patients who had confirmed prior vismodegib treatment, and the nonexposed participants (controls) comprised patients who had never received any SI. Because vismodegib was the first approved SI, only patients exposed to this SI were included. MAIN OUTCOMES AND MEASURES: Hazard ratio for non-BCC malignancies after vismodegib exposure, adjusting for covariates. RESULTS: The study cohort comprised 180 participants. Their mean (SD) age at BCC diagnosis was 56 (16) years, and 68.9% (n = 124) were male. Fifty-five cases were compared with 125 controls, accounting for age, sex, prior radiation therapy or cisplatin treatment, Charlson Comorbidity Index, clinical follow-up time, immunosuppression, and basal cell nevus syndrome status. Patients exposed to vismodegib had a hazard ratio of 6.37 (95% CI, 3.39-11.96; P < .001), indicating increased risk of developing a non-BCC malignancy. Most non-BCC malignancies were cutaneous squamous cell carcinomas, with a hazard ratio of 8.12 (95% CI, 3.89-16.97; P < .001), accounting for age and basal cell nevus syndrome status. There was no significant increase in other cancers. CONCLUSIONS AND RELEVANCE: Increased risk for cutaneous squamous cell carcinomas after vismodegib therapy highlights the importance of continued skin surveillance after initiation of this therapy.

Randomized trial of a hearing conservation intervention for rural students: long-term outcomes.

OBJECTIVES: We had the rare opportunity to conduct a cluster-randomized controlled trial to observe the long-term (16-year) effects of a well-designed hearing conservation intervention for rural high school students. This trial assessed whether the intervention resulted in (1) reduced prevalence of noise-induced hearing loss (NIHL) assessed clinically and/or (2) sustained use of hearing protection devices. METHODS: In 1992-1996, 34 rural Wisconsin schools were recruited and 17 were assigned randomly to receive a comprehensive, 3-year, hearing conservation intervention. In 2009-2010, extensive efforts were made to find and contact all students who completed the original trial. Participants in the 16-year follow-up study completed an exposure history questionnaire and a clinical audiometric examination. Rates of NIHL and use of hearing protection were compared. RESULTS: We recruited 392 participants from the original trial, 200 (53%) from the intervention group and 192 (51%) from the control group. Among participants with exposure to agricultural noise, the intervention group reported significantly greater use of hearing protection compared with the control group (25.9% vs 19.6%; P = .015). The intervention group also reported significantly greater use of hearing protection for shooting guns (56.2% vs 41.6%; P = .029), but the groups reported similar uses of protection in other contexts. There was no significant difference between groups with respect to objective measures of NIHL. CONCLUSION: This novel trial provides objective evidence that a comprehensive educational intervention by itself may be of limited effectiveness in preventing NIHL in a young rural population.