Kathleen Kan, MD

Abstract

OBJECTIVES: Telemedicine for pediatric lower urinary tract symptoms (pLUTS) is a relatively new mode of delivering bladder health education with scant evidence supporting current practice. We aim to examine the safety of pLUTS-related telemedicine visits surrounding the COVID-19 pandemic.METHODS: We conducted a retrospective cohort study of new pLUTS referral diagnoses to our institution's pediatric urology clinics. Demographics, wait times, and referral diagnoses were captured and compared before and after March 2020 using chi2 /Fisher exact tests and t-tests. A retrospective chart review was performed for an initial telemedicine visit followed by an in-person visit to identify missed radiology, lab, or physical exam findings.RESULTS: Six hundred twelve patients were included from September 2018 to August 2021. Most were 5-10years old (62.3%), female (56.2%), English speaking (86.5%), White (39.4%), and had private insurance (67.2%). Wait times were shorter for telemedicine versus in-person visits (t190 =-3.56, p<.001). After March 2020, patients with a urinary tract infection (UTI) and females utilized in-person visits more often (p<.001). After chart review (11 patients, mean=10.4years), 9 (81.8%) had comorbid conditions and/or family history of lower urinary tract symptoms. None had missed clinical findings that changed management.CONCLUSIONS: pLUTS care can be delivered via telemedicine without a significant change in patient volume and population, though additional investigations will clarify the needs of patients with specific referral diagnoses and comorbid conditions. The in-person exam can be omitted safely with proper clinical history taking, supporting future virtual programs that address delays in care within local communities.

View details for DOI 10.1111/luts.12496

View details for PubMedID 37448183

Abstract

BACKGROUND: Bedwetting, or nocturnal enuresis (NE), is a common childhood disease. Families increasingly turn to free online resources for health education in order to navigate treatment options.OBJECTIVE: We aim to determine the readability and quality of online health information that families may encounter when searching for information on pediatric NE.STUDY DESIGN: The search term 'bedwetting' was queried in commonly used search engines. Included articles were further categorized as institutional/reference, commercial, non-profit/charitable, or personal. An online readability platform calculated 3 commonly used readability assessments as well as a consensus score. Quality was assessed by two independent pediatric urologists using a validated DISCERN instrument. Differences in readability were further assessed by article category type.RESULTS: 36 websites were reviewed, 3 did not include treatment options and were not included in quality assessment. 55.6% of the articles were categorized as institutional/reference, 27.8% as non-profit/charitable, 11.1% as personal, and 5.6% as commercial. The average, standard deviation, and range of readability level were as follows: overall consensus score=9.562.09, FK Grade Level Formula score=9.382.17 (range 5.8-14.1), SMOG Index score=8.891.79 (range 6.4-12.9), and the GF Index score=11.862.34 (range 8.4-16.9). The overall consensus score, 9.56, correlates to a 9th-grade reading level. 16 articles were considered to be of 'good' quality, 12 of 'fair' quality, and 5 of 'poor' quality; according to the DISCERN scoring. There was no statistically significant difference in readability scores across website categories. Two articles included the outdated treatment option of intranasal DDAVP.DISCUSSION: These findings are concordant with the increasing body of literature demonstrating that patient education materials are too difficult to read. Pediatric urologists should be aware of the readability and quality of available online content for common clinical presentations. They can ensure that institutional articles online are useful to patients by participating in the development and design of these materials.CONCLUSION: We demonstrated that online articles that families encounter via a search engine query for bedwetting are written at a higher level than recommended and the majority are of 'fair' quality. There is much room for improvement for institutions to provide high quality, readable content that supports the needs of families seeking information on NE.

View details for DOI 10.1016/j.jpurol.2023.05.001

View details for PubMedID 37248162

View details for DOI 10.1097/JU.0000000000002312.02

View details for PubMedID 34875871

View details for Web of Science ID 000693689000163

View details for DOI 10.1016/j.jbi.2019.103184

View details for Web of Science ID 000525692600015

Abstract

Clinical care guidelines recommend that newly diagnosed prostate cancer patients at high risk for metastatic spread receive a bone scan prior to treatment and that low risk patients not receive it. The objective was to develop an automated pipeline to interrogate heterogeneous data to evaluate the use of bone scans using a two different Natural Language Processing (NLP) approaches.Our cohort was divided into risk groups based on Electronic Health Records (EHR). Information on bone scan utilization was identified in both structured data and free text from clinical notes. Our pipeline annotated sentences with a combination of a rule-based method using the ConText algorithm (a generalization of NegEx) and a Convolutional Neural Network (CNN) method using word2vec to produce word embeddings.A total of 5,500 patients and 369,764 notes were included in the study. A total of 39% of patients were high-risk and 73% of these received a bone scan; of the 18% low risk patients, 10% received one. The accuracy of CNN model outperformed the rule-based model one (F-measure = 0.918 and 0.897 respectively). We demonstrate a combination of both models could maximize precision or recall, based on the study question.Using structured data, we accurately classified patients' cancer risk group, identified bone scan documentation with two NLP methods, and evaluated guideline adherence. Our pipeline can be used to provide concrete feedback to clinicians and guide treatment decisions.

View details for PubMedID 31014980

Abstract

Calcium oxalate (CaOx) is the predominate component within renal calculi and can be divided into two subtypes: CaOx-monohydrate (COM) and CaOx-dihydrate (COD). COM and COD form in differing urinary environments which suggest differential underlying metabolic abnormalities associated with each subtype. We compared clinical and metabolic findings in CaOx stone-formers to delineate factors differentiating COD and COM stone formers and the implication this holds in terms of etiology and treatment.We identified CaOx stone-formers that had passed their stones or had undergone endoscopic extraction between October 2014 and December 2018. Only patients who had a predominant subtype ( 80% COM or COD) and who had a 24-hour urine evaluation prior to medical management were included. Clinical and metabolic factors were compared in the two subgroups.Out of 157 stone-formers, 121 were COM and 36 were COD. COD formers were younger than COM formers with a mean age of 5316 vs 5915, respectively (p=0.038). There were no observable differences in gender, BMI, HTN, DM, or HLD. COM formers exhibited higher rates of hypocitraturia and hyperoxaluria, p= 0.022 and p=0.018, respectively. Conversely, COD formers had significantly higher rates of hypercalciuria (47% vs. 28%, p=0.012). Multivariate analysis found hypercalciuria to independently predict COD (p=0.043) and hyperoxaluria to predict COM stones (p=0.016).COM formers are more likely to have hyperoxaluria, hypocitraturia, and elevated urinary oxalate levels compared to COD formers. COD formers exhibited higher incidence of hypercalciuria. These data suggest that all CaOx stones are not alike, and that distinct metabolic and clinical etiological differences exist that may guide future management and prevention.

View details for DOI 10.1089/end.2019.0245

View details for PubMedID 31154910

Abstract

To develop and test a method for automatic assessment of a quality metric, provider-documented pretreatment digital rectal examination (DRE), using the outputs of a natural language processing (NLP) framework.An electronic health records (EHR)-based prostate cancer data warehouse was used to identify patients and associated clinical notes from 1 January 2005 to 31 December 2017. Using a previously developed natural language processing pipeline, we classified DRE assessment as documented (currently or historically performed), deferred (or suggested as a future examination) and refused.We investigated the quality metric performance, documentation 6months before treatment and identified patient and clinical factors associated with metric performance.The cohort included 7215 patients with prostate cancer and 426227 unique clinical notes associated with pretreatment encounters. DREs of 5958 (82.6%) patients were documented and 1257 (17.4%) of patients did not have a DRE documented in the EHR. A total of 3742 (51.9%) patient DREs were documented within 6 months prior to treatment, meeting the quality metric. Patients with private insurance had a higher rate of DRE 6months prior to starting treatment as compared with Medicaid-based or Medicare-based payors (77.3%vs69.5%, p=0.001). Patients undergoing chemotherapy, radiation therapy or surgery as the first line of treatment were more likely to have a documented DRE 6months prior to treatment.EHRs contain valuable unstructured information and with NLP, it is feasible to accurately and efficiently identify quality metrics with current documentation clinician workflow.

View details for DOI 10.1136/bmjopen-2018-027182

View details for PubMedID 31324681

Abstract

Digital rectal examination (DRE) is considered a quality metric for prostate cancer care. However, much of the DRE related rich information is documented as free-text in clinical narratives. Therefore, we aimed to develop a natural language processing (NLP) pipeline for automatic documentation of DRE in clinical notes using a domain-specific dictionary created by clinical experts and an extended version of the same dictionary learned by clinical notes using distributional semantics algorithms. The proposed pipeline was compared to a baseline NLP algorithm and the results of the proposed pipeline were found superior in terms of precision (0.95) and recall (0.90) for documentation of DRE. We believe the rule-based NLP pipeline enriched with terms learned from the whole corpus can provide accurate and efficient identification of this quality metric.

View details for PubMedID 30815067