Designing Human-Centered AI to Prevent Medication Dispensing Errors: Focus Group Study With Pharmacists.

BACKGROUND: Medication errors, including dispensing errors, represent a substantial worldwide health risk with significant implications in terms of morbidity, mortality, and financial costs. Although pharmacists use methods like barcode scanning and double-checking for dispensing verification, these measures exhibit limitations. The application of artificial intelligence (AI) in pharmacy verification emerges as a potential solution, offering precision, rapid data analysis, and the ability to recognize medications through computer vision. For AI to be embraced, it must be designed with the end user in mind, fostering trust, clear communication, and seamless collaboration between AI and pharmacists. OBJECTIVE: This study aimed to gather pharmacists' feedback in a focus group setting to help inform the initial design of the user interface and iterative designs of the AI prototype. METHODS: A multidisciplinary research team engaged pharmacists in a 3-stage process to develop a human-centered AI system for medication dispensing verification. To design the AI model, we used a Bayesian neural network that predicts the dispensed pills' National Drug Code (NDC). Discussion scripts regarding how to design the system and feedback in focus groups were collected through audio recordings and professionally transcribed, followed by a content analysis guided by the Systems Engineering Initiative for Patient Safety and Human-Machine Teaming theoretical frameworks. RESULTS: A total of 8 pharmacists participated in 3 rounds of focus groups to identify current challenges in medication dispensing verification, brainstorm solutions, and provide feedback on our AI prototype. Participants considered several teaming scenarios, generally favoring a hybrid teaming model where the AI assists in the verification process and a pharmacist intervenes based on medication risk level and the AI's confidence level. Pharmacists highlighted the need for improving the interpretability of AI systems, such as adding stepwise checkmarks, probability scores, and details about drugs the AI model frequently confuses with the target drug. Pharmacists emphasized the need for simplicity and accessibility. They favored displaying only essential information to prevent overwhelming users with excessive data. Specific design features, such as juxtaposing pill images with their packaging for quick comparisons, were requested. Pharmacists preferred accept, reject, or unsure options. The final prototype interface included (1) checkmarks to compare pill characteristics between the AI-predicted NDC and the prescription's expected NDC, (2) a histogram showing predicted probabilities for the AI-identified NDC, (3) an image of an AI-provided "confused" pill, and (4) an NDC match status (ie, match, unmatched, or unsure). CONCLUSIONS: In partnership with pharmacists, we developed a human-centered AI prototype designed to enhance AI interpretability and foster trust. This initiative emphasized human-machine collaboration and positioned AI as an augmentative tool rather than a replacement. This study highlights the process of designing a human-centered AI for dispensing verification, emphasizing its interpretability, confidence visualization, and collaborative human-machine teaming styles.

Facilitators and barriers to blood pressure telemonitoring: A mixed-methods study.

Background: Telemonitoring of blood pressure (BP) may improve BP control. However, many patients are not using BP telemonitoring due to personal, technological, and health system barriers. Individuals are required to have electronic health literacy (e-HL), defined as knowledge and skills to use technology services effectively, such as BP telemonitoring. Objective: The objective was to determine the facilitators and barriers experienced by patients with hypertension in telemonitoring of BP using the e-HL framework (e-HLF). Methods: This study was a prospective mixed-methods study using a convergent design. We recruited a convenience sample of 21 patients with hypertension. The qualitative section was online or phone individual in-depth interviews based on the e-HLF, which has seven domains. The quantitative section was an online survey consisting of demographics, an e-HL questionnaire, and patient-provider communication preferences. A joint display was used in the mixed-methods analysis. Results: Five themes including knowledge, motivation, skills, systems, and behaviors along with 28 subthemes comprising facilitators or barriers of BP telemonitoring were identified. The mixed-methods results showed concordance between the participants' e-HL status and their experiences in the ability to actively engage with BP monitoring and managing digital services (domain 3) of the e-HLF. Other e-HL domains showed discordance. Conclusion: Patients may engage with BP telemonitoring when they feel the usefulness of concurrent access to telemonitoring services that suit their needs.

Implementation outcomes of the Structured and Codified SIG format in electronic prescription directions.

OBJECTIVE: To determine the extent of implementation, completeness, and accuracy of Structured and Codified SIG (S&C SIG) directions on electronic prescriptions (e-prescriptions). MATERIALS AND METHODS: A retrospective analysis of a random sample of 3.8 million e-prescriptions sent from electronic prescribing (e-prescribing) software to outpatient pharmacies in the United States between 2019 and 2021. Natural language processing was used to identify direction components, including action verb, dose, frequency, route, duration, and indication from free-text directions and were compared to the S&C SIG format. Inductive qualitative analysis of S&C direction identified error types and frequencies for each component. RESULTS: Implementation of the S&C SIG format in e-prescribing software resulted in 32.4% of e-prescriptions transmitted with these standardized directions. Directions using the S&C SIG format contained a greater percentage of each direction component compared to free-text directions, except for the indication component. Structured and codified directions contained quality issues in 10.3% of cases. DISCUSSION: Expanding adoption of more diverse direction terminology for the S&C SIG formats can improve the coverage of directions using the S&C SIG format. Building out e-prescribing software interfaces to include more direction components can improve patient medication use and safety. Quality improvement efforts, such as improving the design of e-prescribing software and auditing for discrepancies, are needed to identify and eliminate implementation-related issues with direction information from the S&C SIG format so that e-prescription directions are always accurately represented. CONCLUSION: Although directions using the S&C SIG format may result in more complete directions, greater adoption of the format and best practices for preventing its incorrect use are necessary.

Comparing the variability of ingredient, strength, and dose form information from electronic prescriptions with RxNorm drug product descriptions.

OBJECTIVE: To determine the variability of ingredient, strength, and dose form information from drug product descriptions in real-world electronic prescription (e-prescription) data. MATERIALS AND METHODS: A sample of 10 399 324 e-prescriptions from 2019 to 2021 were obtained. Drug product descriptions were analyzed with a named entity extraction model and National Drug Codes (NDCs) were used to get RxNorm Concept Unique Identifiers (RxCUI) via RxNorm. The number of drug product description variants for each RxCUI was determined. Variants identified were compared to RxNorm to determine the extent of matching terminology used. RESULTS: A total of 353 002 unique pairs of drug product descriptions and NDCs were analyzed. The median (1st-3rd quartile) number of variants extracted for each standardized expression in RxNorm, was 3 (2-7) for ingredients, 4 (2-8) for strength, and 41 (11-122) for dosage forms. Of the pairs, 42.35% of ingredients (n = 328 032), 51.23% of strengths (n = 321 706), and 10.60% of dose forms (n = 326 653) used matching terminology, while 16.31%, 24.85%, and 13.05% contained nonmatching terminology, respectively. DISCUSSION: A wide variety of drug product descriptions makes it difficult to determine whether 2 drug product descriptions describe the same drug product (eg, using abbreviations to describe an active ingredient or using different units to represent a concentration). This results in patient safety risks that lead to incorrect drug products being ordered, dispensed, and used by patients. Implementation and use of standardized terminology may reduce these risks. CONCLUSION: Drug product descriptions on real-world e-prescriptions exhibit large variation resulting in unnecessary ambiguity and potential patient safety risks.

Predictors of Smartphone and Tablet Use Among Patients With Hypertension: Secondary Analysis of Health Information National Trends Survey Data.

BACKGROUND: Uncontrolled hypertension leads to significant morbidity and mortality. The use of mobile health technology, such as smartphones, for remote blood pressure (BP) monitoring has improved BP control. An increase in BP control is more significant when patients can remotely communicate with their health care providers through technologies and receive feedback. Little is known about the predictors of remote BP monitoring among hypertensive populations. OBJECTIVE: The objective of this study is to quantify the predictors of smartphone and tablet use in achieving health goals and communicating with health care providers via SMS text messaging among hypertensive patients in the United States. METHODS: This study was a cross-sectional, secondary analysis of the 2017 and 2018 Health Information National Trends Survey 5, cycles 1 and 2 data. A total of 3045 respondents answered "Yes" to the question "Has a doctor or other healthcare provider ever told you that you had high blood pressure or hypertension?", which defined the subpopulation used in this study. We applied the Health Information National Trends Survey full sample weight to calculate the population estimates and 50 replicate weights to calculate the SEs of the estimates. We used design-adjusted descriptive statistics to describe the characteristics of respondents who are hypertensive based on relevant survey items. Design-adjusted multivariable logistic regression models were fitted to estimate predictors of achieving health goals with the help of smartphone or tablet and sending or receiving an SMS text message to or from a health care provider in the last 12 months. RESULTS: An estimated 36.9%, SE 0.9% (183,285,150/497,278,883) of the weighted adult population in the United States had hypertension. The mean age of the hypertensive population was 58.3 (SE 0.48) years. Electronic communication with the doctor or doctor's office through email or internet (odds ratio 2.93, 95% CI 1.85-4.63; P<.001) and having a wellness app (odds ratio 1.82, 95% CI 1.16-2.86; P=.02) were significant predictors of using SMS text message communication with a health care professional, adjusting for other demographic and technology-related variables. The odds of achieving health-related goals with the help of a tablet or smartphone declined significantly with older age (P<.001) and ownership of basic cellphones (P=.04). However, they increased significantly with being a woman (P=.045) or with being married (P=.03), having a wellness app (P<.001), using devices other than smartphones or tablets to monitor health (P=.008), making health treatment decisions (P=.048), and discussing with a provider (P=.02) with the help of a tablet or smartphone. CONCLUSIONS: Intervention measures accounting for age, gender, marital status, and the patient's technology-related health behaviors are required to increase smartphone and tablet use in self-care and SMS text message communication with health care providers.

Performance evaluation of a prescription medication image classification model: an observational cohort.

Technology assistance of pharmacist verification tasks through the use of machine intelligence has the potential to detect dangerous and costly pharmacy dispensing errors. National Drug Codes (NDC) are unique numeric identifiers of prescription drug products for the United States Food and Drug Administration. The physical form of the medication, often tablets and capsules, captures the unique features of the NDC product to help ensure patients receive the same medication product inside their prescription bottle as is found on the label from a pharmacy. We report and evaluate using an automated check to predict the shape, color, and NDC for images showing a pile of pills inside a prescription bottle. In a test set containing 65,274 images of 345 NDC classes, overall macro-average precision was 98.5%. Patterns of incorrect NDC predictions based on similar colors, shapes, and imprints of pills were identified and recommendations to improve the model are provided.

Fewer patients receive recommendations for pharmacotherapy in primary prevention using the 2018 atherosclerotic cardiovascular disease risk estimator.

Shortly after the introduction of the 2013 original Pooled Cohort Equation (PCE), an overestimation of risk was suggested. As such, the updated 2018 PCE was developed to more accurately assess atherosclerotic cardiovascular disease (ASCVD) risk in the population. Hence, this study aims to compare drug prescribing recommendations in a large, real-world patient population, depending on which PCE is used to estimate 10-year ASCVD risk. This retrospective cohort study identified 20,843 patients aged between 40 and 75 years with no previous ASCVD. The 10-year ASCVD risk score was assessed by using both PCE. Patients were assigned to the four risk categories according to the 2018 ACC/AHA guideline. The percentage of patients qualifying for guideline-recommended primary prevention with statins and/or anti-hypertensives were compared between both PCE. Risk reclassification occurred in 26.7% of patients overall (n = 5571), of which 98.1% (n = 5466) were assigned to lower risk categories with the updated PCE. Non-diabetic (14.0%) patients no longer met the threshold for recommending statins as primary prevention with the updated PCE. Likewise, 13.8% of patients with stage I hypertension no longer met the threshold for recommending antihypertensive drugs with the updated PCE. In conclusion, risk reclassification occurred among 26.7% of patients overall, mostly due to lower risk categories assigned by the updated PCE. Up to 14.0% of patients no longer met the threshold for recommending statin therapy and/or antihypertensive drugs by using the updated PCE. These findings suggest that using the updated PCE could translate into fewer patients receiving pharmacotherapy for ASCVD primary prevention.

Human versus machine editing of electronic prescription directions.

BACKGROUND: Pharmacy staff are responsible for editing poor-quality and difficult-to-read electronic prescription (e-prescription) directions. Machine translation (MT) models are capable of translating free text from 1 sequence into another. However, the quality of MTs of e-prescriptions into pharmacy label directions is unknown. OBJECTIVE: To determine the types and frequencies of e-prescription direction component errors made by an MT model, pharmacy staff, and prescribers. METHODS: A prospective evaluation was conducted on a random sample of 300 patient directions in a test set of e-prescriptions from a mail-order pharmacy. Each row included directions produced by (1) prescribers on e-prescriptions, (2) pharmacy staff on prescription labels, and (3) an open neural MT model. Annotators labeled direction sets for missing direction components, use of abbreviations and medical jargon, and incorrect information (e.g., changing the number of tablets to be taken). The longest common subsequence (LCS) compared the amount of pharmacy staff editing with and without MT. RESULTS: Out of 279 direction sets labeled, the MT model directions contained no quality issues in 196 (70.3%) samples compared with 187 (67.0%) and 83 (29.8%) samples for pharmacy staff directions and prescriber directions, respectively. The MT model directions contained more incorrect components (n = 23). Median LCS was greater without MT (30.0 vs. 18.5, P < 0.01, Wilcoxon signed-rank test), indicating more editing was needed. CONCLUSION: MT could be used to improve the quality of e-prescription directions; however, MT makes high-risk mistakes such as incorrectly predicting the tapering regimen for prednisone. The use of semiautomated MT, where pharmacy staff can review model predictions to detect and resolve quality issues, should be considered to improve safety and decrease total work time compared with current practice. MT has strengths and weaknesses for improving the editing process of the patient directions compared with pharmacy staff alone.

Work effort, readability and quality of pharmacy transcription of patient directions from electronic prescriptions: a retrospective observational cohort analysis.

BACKGROUND: Free-text directions generated by prescribers in electronic prescriptions can be difficult for patients to understand due to their variability, complexity and ambiguity. Pharmacy staff are responsible for transcribing these directions so that patients can take their medication as prescribed. However, little is known about the quality of these transcribed directions received by patients. METHODS: A retrospective observational analysis of 529 990 e-prescription directions processed at a mail-order pharmacy in the USA. We measured pharmacy staff editing of directions using string edit distance and execution time using the Keystroke-Level Model. Using the New Dale-Chall (NDC) readability formula, we calculated NDC cloze scores of the patient directions before and after transcription. We also evaluated the quality of directions (eg, included a dose, dose unit, frequency of administration) before and after transcription with a random sample of 966 patient directions. RESULTS: Pharmacy staff edited 83.8% of all e-prescription directions received with a median edit distance of 18 per e-prescription. We estimated a median of 6.64 s of transcribing each e-prescription. The median NDC score increased by 68.6% after transcription (26.12 vs 44.03, p<0.001), which indicated a significant readability improvement. In our sample, 51.4% of patient directions on e-prescriptions contained at least one pre-defined direction quality issue. Pharmacy staff corrected 79.5% of the quality issues. CONCLUSION: Pharmacy staff put significant effort into transcribing e-prescription directions. Manual transcription removed the majority of quality issues; however, pharmacy staff still miss or introduce following their manual transcription processes. The development of tools and techniques such as a comprehensive set of structured direction components or machine learning-based natural language processing techniques may help produce clear directions.

Quality of Handoffs in Community Pharmacies.

OBJECTIVES: The aims of the study were to characterize handoffs in community pharmacies and to examine factors that contribute to perceived handoff quality. METHODS: A cross-sectional study of community pharmacists in a Midwest State of the United States. Self-administered questionnaires were used to collect information on participant and practice setting characteristics. Data were analyzed using descriptive statistics and multivariate logistic regression. RESULTS: A total of 445 completed surveys were returned (response rate, 82%). In almost half of the time, handoffs that occur in a community pharmacy setting were inaccurate or incomplete. Nearly half of the time handoffs occur in environments full of interruptions and distractions. More than 90% of the respondents indicated that they have undergone no formal training on proper ways of handing off information. Nearly 40% of respondents reported that their pharmacy dispensing technology does not have adequate functionality to support handing off information and that at least 50% of the time, poor handoffs result in additional work to the pharmacist because of the need for complete information before providing patient care. Multivariate analysis showed that being very familiar with patients, lower daily prescription volume, not having a 24-hour operation, and larger percentage of handoffs occurring in a synchronous fashion are all associated with better handoff quality. CONCLUSIONS: Handoffs occur frequently and are problematic in community pharmacies. Current pharmacy environments offer limited support to conduct good handoffs, and as a result, pharmacists report loss of information. This could present as a significant patient safety hazard. Future interventions should target facilitating better communication during shift changes.

Changes in Statin Use Among U.S. Adults With Diabetes: A Population-Based Analysis of NHANES 2011-2018.

OBJECTIVE: To evaluate statin use in the U.S. before and after the 2015 American Diabetes Association position statement, which expanded statin therapy recommendations to include all adults 40-75 years old with diabetes. RESEARCH DESIGN AND METHODS: The National Health and Nutrition Examination Survey (NHANES) was used to obtain a representative sample. The difference-in-differences technique determined the impact of the recommendation on the proportion of people with diabetes for whom statin therapy was newly recommended. RESULTS: Among people with diabetes, the change in statin use in people without atherosclerotic cardiovascular disease (ASCVD) risk factors, controlling for change among people with ASCVD/risk factors, was 6.6% (P = 0.388). In the adjusted analysis, overt ASCVD, age, Black race, health insurance, a place for routine care, and total cholesterol were significantly associated with statin use (P < 0.05). CONCLUSIONS: The most recent change in statin recommendations had minimal impact on the proportion of patients receiving a statin.

Factors that influence patient acceptance of clinical pharmacy services: A nationwide survey.

OBJECTIVE: Examine the factors that influence a patient's likelihood of participating in clinical pharmacy services so that pharmacists can use this knowledge to effectively expand clinical services. METHODS: An online survey was distributed to U.S. citizens 55 years of age or older through a market research company. The survey assessed pharmacy and medication use, general health, interest in clinical pharmacy services, and general demographics. The specific clinical services examined included medication therapy management (MTM) and a collaborative practice agreement (CPA). Logistic regression and best-worst scaling were used to predict the likelihood of participating and determine the motivating factors to participate in clinical pharmacy services, respectively. RESULTS: Two hundred eight (58.45%) respondents reported being likely to participate in MTM services, and 108 (50.6%) reported being likely to participate in the services offered by a pharmacist with a CPA, if offered. The motivations to participate in MTM were driven by pharmacist management of medication interactions and adverse effects (best-worst scores 0.62 and 0.51, respectively). The primary motivator to participate in a CPA was improved physician-pharmacist coordination (best-worst score 0.80). Those with a personal pharmacist were more likely to participate in MTM (odds ratio [OR] 2.43 [95% CI 1.41-4.22], P = 0.002) and a pharmacist CPA (2.08 [1.26-3.44], P = 0.004). Previous experience with MTM increased the likelihood of participating again in MTM (5.98 [95% CI 2.50-14.35], P < 0.001). Patient satisfaction with the pharmacy increased the likelihood of participating in a pharmacist CPA (1.47 [95% CI 1.01-2.13], P = 0.04). CONCLUSION: Patients are interested in clinical pharmacy services for the purposes of medication interaction management, adverse effect management, and improved physician-pharmacist coordination. The factors that influenced the likelihood of participating included having a personal pharmacist, previous experience with MTM, and pharmacy satisfaction. These results suggest a potential impact of the patient-pharmacist relationship on patient participation in clinical services.

A learning pharmacy practice enabled by the pharmacists' patient care process.

To address the Quintuple Aim of health care improvement, the profession of pharmacy is on the verge of a practice transformation that incorporates continuous learning from medication-related data into existing clinical and dispensing roles. The pharmacists' patient care process (PPCP) enables a learning pharmacy practice through the systematic and standardized collection of real-world medication-related data from pharmacists' patient care activities. A learning pharmacy practice continually generates data-powered discoveries as a byproduct of PPCP interactions. In turn, these discoveries improve our medication knowledge while upgrading our predictive powers, thus helping all people achieve optimal health outcomes. Establishing a practice management system connected to the PPCP means that data are generated from every PPCP interaction, combined with existing data, and analyzed by teams of pharmacists and data scientists. The resulting new knowledge is then incorporated into all future PPCP interactions in the form of predictions coupled to actionable advice. The primary purpose of a learning pharmacy practice is to combine the power of predictive modeling with evidence-based best practices to achieve and sustain population-level health improvements. This purpose is achieved by systematically optimizing individual medication use in an equitable manner on a global scale.

Detecting Potential Medication Selection Errors During Outpatient Pharmacy Processing of Electronic Prescriptions With the RxNorm Application Programming Interface: Retrospective Observational Cohort Study.

BACKGROUND: Medication errors are pervasive. Electronic prescriptions (e-prescriptions) convey secure and computer-readable prescriptions from clinics to outpatient pharmacies for dispensing. Once received, pharmacy staff perform a transcription task to select the medications needed to process e-prescriptions within their dispensing software. Later, pharmacists manually double-check medications selected to fulfill e-prescriptions before dispensing to the patient. Although pharmacist double-checks are mostly effective for catching medication selection mistakes, the cognitive process of medication selection in the computer is still prone to error because of heavy workload, inattention, and fatigue. Leveraging health information technology to identify and recover from medication selection errors can improve patient safety. OBJECTIVE: This study aimed to determine the performance of an automated double-check of pharmacy prescription records to identify potential medication selection errors made in outpatient pharmacies with the RxNorm application programming interface (API). METHODS: We conducted a retrospective observational analysis of 537,710 pairs of e-prescription and dispensing records from a mail-order pharmacy for the period January 2017 to October 2018. National Drug Codes (NDCs) for each pair were obtained from the National Library of Medicine's (NLM's) RxNorm API. The API returned RxNorm concept unique identifier (RxCUI) semantic clinical drug (SCD) identifiers associated with every NDC. The SCD identifiers returned for the e-prescription NDC were matched against the corresponding SCD identifiers from the pharmacy dispensing record NDC. An error matrix was created based on the hand-labeling of mismatched SCD pairs. Performance metrics were calculated for the e-prescription-to-dispensing record matching algorithm for both total pairs and unique pairs of NDCs in these data. RESULTS: We analyzed 527,881 e-prescription and pharmacy dispensing record pairs. Four clinically significant cases of mismatched RxCUI identifiers were detected (ie, three different ingredient selections and one different strength selection). A total of 546 less significant cases of mismatched RxCUIs were found. Nearly all of the NDC pairs had matching RxCUIs (28,787/28,817, 99.90%-525,270/527,009, 99.67%). The RxNorm API had a sensitivity of 1, a false-positive rate of 0.00104 to 0.00312, specificity of 0.99896 to 0.99688, precision of 0.00727 to 0.04255, and F1 score of 0.01444 to 0.08163. We found 872 pairs of records without an RxCUI. CONCLUSIONS: The NLM's RxNorm API can perform an independent and automatic double-check of correct medication selection to verify e-prescription processing at outpatient pharmacies. RxNorm has near-comprehensive coverage of prescribed medications and can be used to recover from medication selection errors. In the future, tools such as this may be able to perform automated verification of medication selection accurately enough to free pharmacists from having to perform manual double-checks of the medications selected within pharmacy dispensing software to fulfill e-prescriptions.

Evaluation of Patient Safety Culture in Community Pharmacies.

OBJECTIVE: Medication errors are common in community pharmacies. Safety culture is considered a factor for medication safety but has not been measured in this setting. The objectives of this study were to describe safety culture measured using the Agency for Healthcare Research and Quality (AHRQ) Community Pharmacy Survey on Patient Safety Culture and to assess predictors of overall patient safety. METHODS: This is a cross-sectional survey of community pharmacists practicing in Wisconsin measuring safety culture. Demographic variables collected included pharmacist and pharmacy characteristics. Data were analyzed using descriptive statistics, χ, and multivariate logistic regression analyses. RESULTS: A total of 445 surveys were completed (response rate, 82%). Safety culture was positively associated with the following: an independent pharmacy (adjusted odds ratio [AOR], 1.69; 95% confidence interval [CI], 1.11-2.57), a health maintenance organization or clinic (AOR, 2.25; 95% CI, 1.34-3.78), being somewhat familiar with patients (AOR, 3.35; 95% CI, 1.82-6.19), or very/extremely familiar with patients (AOR, 8.8; 95% CI, 4.68-16.59). Five of the composite scores differed significantly from the results of the AHRQ pilot study (response to mistakes, communication openness, organizational learning-continuous improvement, communication about prescriptions across shifts, and overall patient safety). Consistent with the AHRQ pilot study, the composite describing staffing, work pressure, and pace had the lowest score (37.6%). CONCLUSIONS: Understanding the safety culture of community pharmacies can help identify areas of strength and those that require improvement. Improvement efforts that focus on staffing, work pressure, and pace in community pharmacies may lead to better safety culture.

Effects of Race on Statin Prescribing for Primary Prevention With High Atherosclerotic Cardiovascular Disease Risk in a Large Healthcare System.

Background Although guidelines recommend statins with a high level of evidence for 4 primary prevention benefit groups, prescribing disparities still exist. The objective of this study was to evaluate the effects of race on statin prescribing for primary prevention. Methods and Results A retrospective cohort analysis of patients within a large academic health system was performed to investigate statin prescribing among primary prevention groups. The statin benefits groups were patients diagnosed with diabetes mellitus, with an low-density lipoprotein ≥190 mg/dL, or with an atherosclerotic cardiovascular disease (ASCVD) 10-year risk ≥7.5%. Statin prescribing was 20% in the ASCVD ≥7.5% group, followed by 37.8% in the low-density lipoprotein ≥190 mg/dL group and 40.5% in the diabetes mellitus group. Blacks were less likely to be prescribed a statin compared with whites in the diabetes mellitus (odds ratio, 0.64; 95% CI, 0.49-0.82; P=0.001) and ASCVD ≥7.5% groups (odds ratio, 0.38; 95% CI, 0.26-0.54; P<0.0001). Blacks 60 to 69 years of age (odds ratio, 7.97; 95% CI, 3.14-20.2; P=0.003) and 70 to 79 years of age (odds ratio, 4.21; 95% CI, 1.81-9.79; P=0.008) were more likely to be prescribed a statin compared with blacks <60 years of age in the ASCVD ≥7.5% group. Conclusions Blacks are less likely to be prescribed statins in diabetes mellitus and ASCVD ≥7.5% groups compared with whites. Younger blacks with ASCVD risk ≥7.5% are less likely to be prescribed statins compared with older blacks. Future research should focus on tailored interventions to address statin prescribing disparities in blacks.

Medication Accuracy in Electronic Health Records for Microbial Keratitis.

IMPORTANCE: Electronic health records (EHRs) contain an abundance of health information. However, researchers need to understand data accuracy to ask appropriate research questions. OBJECTIVE: To investigate the concordance of the names of medications for microbial keratitis in the structured, formal EHR medication list and the text of clinicians' progress notes. DESIGN, SETTING, AND PARTICIPANTS: This cross-sectional study, conducted in the cornea section of an ophthalmology department in a tertiary care, referral academic medical center, examined the medications of 53 patients with microbial keratitis treated until disease resolution from July 1, 2015, to August 1, 2018. Documentation of medications was compared between the structured medication list extracted from the EHR server and medications written into the clinical progress note and transcribed by the study team. EXPOSURE: Medication treatment for microbial keratitis. MAIN OUTCOMES AND MEASURES: Medication mismatch frequency. RESULTS: The study sample included 24 men and 29 women, with a mean (SD) age of 51.8 (19.6) years. Of the 247 medications identified, 57 (23.1%) of prescribed medications differed between the progress notes and the formal EHR-based medication list. Reasons included medications not prescribed via the EHR ordering system (25 [43.9%]), outside medications not reconciled in the internal EHR medication list (23 [40.4%]), and medications prescribed via the EHR ordering system and in the formal list, but not described in the clinical note (9 [15.8%]). Fortified antimicrobials represented the largest category for medication mismatch between modalities (17 of 70 [24.3%]). Nearly one-third of patients (17 [32.1%]) had at least 1 medication mismatch in their record. CONCLUSIONS AND RELEVANCE: Almost 1 in 4 medications were mismatched between the progress note and formal medication list in the EHR. These findings suggest that EHR data should be checked for internal consistency before use in research.

Describing the patient experience from Yelp reviews of community pharmacies.

OBJECTIVES: To examine the characteristics of patient experience in community pharmacies through pattern exploration techniques of the unstructured free-text data from an online review website. DESIGN: Retrospective observational study design using structural topic model (STM) and term frequency-inverse document frequency (tf-idf) to categorize free-text data. Tf-idf scores words in terms of importance, and STM extracts latent themes from free-text data based on the co-occurrence of words in a review. Human labels were assigned to STM output, with each topic's prevalence mapped to each level of the 1- to 5-star review ratings. SETTING AND PARTICIPANTS: Data were obtained from the Yelp Academic data set from April 2006 through December 2017. These data were available for analysis from certain cities in the United States, Canada, and Europe. Included reviews were filtered based on the presence of pharmacy-specific character strings (e.g., "prescri"). MAIN OUTCOME MEASURES: Descriptive statistics of Yelp review characteristics, tf-idf scores, and topics produced from STM were used to characterize the content of Yelp reviews at each star-rating level. RESULTS: The filtered data set contained 4463 reviews from 964 pharmacies in 8 U.S. states. The mean (±SD) review rating was 2.97 ± 0.91. The mean number of words in a review was 135 ± 116. STM revealed 9 topics that influenced patient experiences at community pharmacies, including waiting time, service attitude, and physical store characteristics. Friendly and helpful staff accounted for 28.3% of content in 5-star ratings, whereas waiting time accounted for 19.4% of 1-star ratings. CONCLUSION: Yelp reviews provide a public look into patient experience at community pharmacies, and the reviews likely influence other patients' decisions to use the pharmacy. Pharmacies should focus their efforts on enabling pharmacy staff to provide high-quality care and minimizing unnecessary waiting times for patients.