Differences in Self-Reported Stress Versus. Hair and Nail Cortisol Among Adolescent and Young Adult Males.

BACKGROUND: Chronic stress causes harmful physiologic responses that yield increased inflammation and subsequent health conditions. Stress is an important measure among minoritized populations who face social situations that predispose risk to developing mental health problems. Hair and fingernail cortisol have been studied as retrospective measures of chronic stress and to demonstrate biological response to social situations. OBJECTIVES: The objective of this study was to compare the Perceived Stress Scale (PSS) with hair and nail cortisol concentrations and assess the risk factors associated with stress levels among heterosexual and sexual and gender-minoritized adolescent males. METHODS: We recruited a cohort of adolescents who were assigned male sex at birth. Approximately half of our cohort consisted of sexual and gender-minoritized people, and half consisted of heterosexual cisgender males. Participants provided hair and nail samples and completed a survey that included demographic and hair hygiene questions, and the PSS. Hair and nail samples were processed in a lab and survey results were analyzed descriptively. RESULTS: Several samples were not provided or received, and some survey data was missing. Hair and nail cortisol values were significantly correlated. There was no significant relationship between the PSS and hair and nail cortisol values. No significant differences were found between the heterosexual and sexual minoritized groups. Black participants reported lower perceived stress scores compared to White participants. Participants whose gender was non-binary or genderqueer had higher hair cortisol values compared to those who identified as male. Older participants had higher hair cortisol values compared to younger participants. DISCUSSION: Previous researchers have similarly found no correlation between self-report stress scales and cortisol values, increased stress experience among non-binary or genderqueer individuals compared to cisgender individuals, and a positive correlation between aging and stress. Yet, our finding that Black participants reported lower stress levels than White participants is unexpected. Our study demonstrates a high correlation between hair and nail cortisol values, suggeting the potential to interchange these markers as needed.

Hair and nail cortisol levels are associated and affected by glucocorticoid use.

Hair and nail cortisol is increasingly studied as a physiologic proxy for chronic stress response. Glucocorticoid use is an expected confounder for cortisol measurement, yet there remains little evidence of whether external cortisol use should be subject to exclusion in study subjects. In a group of 209 youth (15-22 year-olds), we analyzed hair and fingernail cortisol concentrations. We assessed topical, nasal, oral, and injectable glucocorticoid use via a questionnaire. Extensively validated methods were used for hair and nail cortisol extraction and measurements. The median value of hair cortisol was 10.2 pg/mg (n=200), and the median value of nail cortisol was 7.06 pg/mg (n=203). Topical glucocorticoid use significantly increased hair and nail cortisol concentrations (p<0.005). Hair and nail cortisol concentrations were positively associated (p<0.0001, n=194). Spearman correlation coefficients demonstrated that the positive correlation between hair and nail cortisol values was higher in participants who used external glucocorticoids. Topical glucocorticoids moderated the association between hair and nail cortisol values (p=0.006). Based on these findings, we recommend that the assessment of topical glucocorticoid use must be performed when collecting hair/nail samples and that subjects reporting glucocorticoid use should be excluded from all future hair and nail cortisol studies; also, all outliers must be excluded to account for glucocorticoid medication underreporting and yet-unknown confounders.

Development and initial psychometric properties of the Research Complexity Index.

Objective: Research study complexity refers to variables that contribute to the difficulty of a clinical trial or study. This includes variables such as intervention type, design, sample, and data management. High complexity often requires more resources, advanced planning, and specialized expertise to execute studies effectively. However, there are limited instruments that scale study complexity across research designs. The purpose of this study was to develop and establish initial psychometric properties of an instrument that scales research study complexity. Methods: Technical and grammatical principles were followed to produce clear, concise items using language familiar to researchers. Items underwent face, content, and cognitive validity testing through quantitative surveys and qualitative interviews. Content validity indices were calculated, and iterative scale revision was performed. The instrument underwent pilot testing using 2 exemplar protocols, asking participants (n = 31) to score 25 items (e.g., study arms, data collection procedures). Results: The instrument (Research Complexity Index) demonstrated face, content, and cognitive validity. Item mean and standard deviation ranged from 1.0 to 2.75 (Protocol 1) and 1.31 to 2.86 (Protocol 2). Corrected item-total correlations ranged from .030 to .618. Eight elements appear to be under correlated to other elements. Cronbach's alpha was 0.586 (Protocol 1) and 0.764 (Protocol 2). Inter-rater reliability was fair (kappa = 0.338). Conclusion: Initial pilot testing demonstrates face, content, and cognitive validity, moderate internal consistency reliability and fair inter-rater reliability. Further refinement of the instrument may increase reliability thus providing a comprehensive method to assess study complexity and related resource quantification (e.g., staffing requirements).

Individual and organizational factors influencing well-being and burnout amongst healthcare assistants: A systematic review.

Background: Increasing evidence suggests that clinician well-being influences patient, workforce, and organizational outcomes. Despite increasing attention to well-being among licensed clinicians (e.g., nurses and physicians), collective evidence about well-being among healthcare assistants, such as nursing and medical assistants, is limited. Healthcare assistants make up a substantial portion of the clinical workforce delivering direct patient care. The well-being of healthcare assistants is critical to ensure an ample workforce supply. The objective of this systematic review was to contribute a reproducible search, summary, appraisal, synthesis, and critique of the literature about well-being among healthcare assistants, including factors that induce or inhibit burnout, and to identify gaps in evidence that warrant future research. Methods: We performed a literature search across 4 databases with keywords using BOOLEAN operators. After an initial title and abstract screen, a search of relevant reference lists, and full text review was peformed independently by 2 researchers. Study quality was evaluated using Joanna Briggs Institute Critical Appraisal Checklist for Analytical Cross-Sectional Studies. We extracted study characteristics, results, and deductively analyzed each study's alignment with the United States National Academy of Medicine's Clinician Well-Being Model. Results: We identified 28 articles meeting our inclusion criteria. Our synthesis indicated that most studies investigated personal factors (e.g., financial stressors or physical, emotional, and spiritual health) as opposed to organizational or policy factors (e.g., occupational policies or workplace culture) that may impact well-being. Younger workers and those with fewer years of experience appear to have higher burnout risk. Sleep health, improved unit-based culture (respect and increased decision-making with nurses), shorter shifts, and increased social support appear to be the most protective against burnout. Discussion: There remains a scarcity of evidence about factors impacting well-being among healthcare assistants. Existing literature focuses on individual, as opposed to external or organizational, contributory factors to burnout or well-being risk. Future studies should use specific methods to define and measure healthcare assistant roles, isolate harmful individual and organizational factors, and measure more specific sub-concepts of well-being such as depression. Such studies can contribute greatly to the overall understanding of healthcare assistant health and wellness, which subsequently may promote optimal patient and organizational outcomes. Tweetable abstract: The hidden workforce: Systematic review demonstrates gaps in evidence about wellbeing and burnout among healthcare assistants and aides.

Modifiable work stress factors and psychological health risk among nurses working within 13 countries.

INTRODUCTION: Nurses are identified as having higher work stress and poor mental health risk among health care workforce globally. It remains unclear which modifiable stress factors pose the greatest risk for poor psychological health among nursing workforce and needed to inform targeted practice and policy change. To determine which occupation-related or personal stress factors precipitate higher risk for burnout, depression, anxiety, job satisfaction or intention to leave one's position among nurses globally. DESIGN: A cross-sectional anonymous survey was administered via email using a snowball recruitment strategy. METHODS: Academic researchers and clinical industry leaders across 3 global regions collaborated to generate an email listserv of professional nursing contacts for survey distribution. The survey included valid and reliable measures to scale stress factors (Work Stress Questionnaire), and screen for burnout (single item), depression (Patient Health Questionnaire-2), anxiety (Generalized Anxiety Disorder-2), resilience (Brief Resilience Scale) and intention to leave one's job (single item). We used logistic regression, first unadjusted and then adjusted for personal and professional characteristics, to determine associations between stress factors and psychological health risk. RESULTS: The final sample consisted of responses from 2864 nurses working across 13 countries. Most respondents reported working as a clinical nurse in the Philippines (n = 2275), United States (n = 424) and Saudi Arabia (n = 104). One third of nursing respondents endorsed high burnout and intention to leave their job. Those reporting work conflict had significantly higher odds of burnout (odds ratio 3.18; 95% CI 2.22-4.54) and three times more likely to screen positive for depression (odds ratio 3.02; 95% CI 1.36-6.72) and anxiety (odds ratio 2.92; 95% CI 1.57-5.43). Those endorsing difficulty sleeping were 15 times more likely to screen positive for depression (odds ratio 15.63; 95% CI 2.09-117.06). Lack of social support was significantly associated to higher risk for burnout, job dissatisfaction, depression, anxiety, and intention to leave one's position. CONCLUSIONS: Nurses remain at risk for burnout and poor psychological health stemming from work stress. Factors such as clear workplace goals and assignments, increased engagement, good sleep health and social support may serve as protective factors against suboptimal psychological health, and in-turn poor workforce retention. CLINICAL RELEVANCE: Nurses reporting conflict in the workplace are three times more likely to screen positive for burnout, depression, and anxiety. Nurses reporting difficulty sleeping are 15 times more likely to screen positive for depression. Several modifiable factors can be targeted to reduce poor psychological health and high workforce turnover among nurses across countries.

Construct Validity Testing of the Provider Co-Management Index to Measure Shared Care in Provider Dyads.

BACKGROUND: Co-management encompasses the dyadic process between two healthcare providers. The Provider Co-Management Index (PCMI) was initially developed as a 20-item instrument across three theory-informed subscales. OBJECTIVE: This study aimed to establish construct validity of the PCMI with a sample of primary care providers through exploratory and confirmatory factor analyses. METHODS: We conducted a cross-sectional survey of primary care physicians, nurse practitioners, and physician assistants randomly selected from the IQVIA database across New York State. Mail surveys were used to acquire a minimum of 300 responses for split sample factor analyses. The first subsample (derivation sample) was used to explore factorial structure by conducting an exploratory factor analysis. A second (validation) sample was used to confirm the emerged factorial structure using confirmatory factor analysis. We performed iterative analysis and calculated good fit indices to determine the best-fit model. RESULTS: There were 333 responses included in the analysis. Cronbach's alpha was high for a three-item per dimension scale within a one-factor model. The instrument was named PCMI-9 to indicate the shorter version length. DISCUSSION: This study established the construct validity of an instrument that scales the co-management of patients by two providers. The final instrument includes nine items on a single factor using a 4-point, Likert-type scale. Additional research is needed to establish discriminant validity.

Organizational trust breaches among nurses and aides: A qualitative study.

BACKGROUND: Healthcare worker retention and burnout are confounding issues. Trust among workers and their employer, that is, organization, is an important yet underexplored concept in research. RESEARCH AIM: The aim of this qualitative study is to explore organizational actions and systems that promote or denigrate trust among registered nurses and patient care aides (aides). RESEARCH DESIGN: The study uses the Model of Psychological Contract as a theoretical framework. Focus groups were conducted to explore the concept of organizational trust and the consequences of broken trust. PARTICIPANTS: Registered nurses (RNs) (n=6) and aides (n=6) participated in the study. Six focus groups (three RN and three aide) were conducted, with two participants per group. Focus groups were conducted online. ETHICAL CONSIDERATIONS: The study's methods were reviewed by the University of Pittsburgh Institutional Review Board. FINDINGS: Among RNs and aides, a sense of trust and feeling valued were important to their sense of relationship with their employers. Trust was breached when resources were scarce, employees did not feel validated and listened to, and problems were not addressed. RNs and aides described feeling devalued when compensation practices were unjust or inequitable, they had limited autonomy, and the employer created an organizational climate where business needs superceded human caring. Consequences of trust breach included burnout, dejection, and feelings of non-belonging. DISCUSSION: Tangible organizational resources (compensation and staffing) and intangible resources (value, respect, autonomy) are important to RNs and aides alike. Inability to provide these resources diminishes trust and even causes a sense of betrayal. CONCLUSION: Future research can explore the concepts of organizational justice and interventions to restore lost trust and improve healthcare worker well-being. This is one of only a few identified studies to explore organizational factors and well-being among aides and more research among this healthcare worker population is warranted.

Sleep Disturbance and Burnout in Emergency Department Health Care Workers.

This cross-sectional study examines the association of sleep disturbances with burnout among emergency medicine health care workers.

The Impact of a Nursing-Led Intervention Bundle With a Bedside Checklist to Reduce Mortality During the Initial COVID-19 Pandemic and Implications for Future Emergencies.

OBJECTIVE: The study objective was to examine the effectiveness of a bedside checklist enforcing nursing-led interventions in hospitalized COVID-19 patients early in the pandemic. BACKGROUND: The absence of treatment guidelines for COVID-19 presented challenges to reducing mortality rates early in the pandemic. A bedside checklist and a bundle of nursing-led interventions named "Nursing Back to Basics (NB2B)" were assembled for patient care after a scoping review of evidence. METHODS: A retrospective analysis was conducted to investigate the impact of the evidence-based interventions randomly implemented based on patient bed assignment. Electronic data (patient demographics, bed assignment and ICU transfers, length of stay, and patient discharge disposition) were extracted and calculated using descriptive statistics, t tests, and linear regression. RESULTS: Patients receiving the NB2B intervention enforced with a bedside checklist had significantly lower mortality rates (12.3%) compared with those receiving standard nursing care (26.9%). CONCLUSIONS: Bedside checklists enforcing evidence-based nursing-led interventions may be beneficial as a 1st-line public health emergency response.

Forum Theater: An Innovative Strategy to Enhance Clinician Communication With Marginalized Trauma Patients.

BACKGROUND: Marginalized groups experience a higher frequency of traumatic injury and are more likely to report negative experiences in the health care setting. Trauma center staff are prone to compassion fatigue, which impairs patient and clinician interactions for these groups. Forum theater (a form of interactive theater designed for addressing social issues) is proposed as an innovative method of exploring bias and has never been applied in the trauma setting. OBJECTIVE: This article aims to determine the feasibility of implementing forum theater as an adjunct to enhance clinician understanding of bias and its influence on communication between clinicians and trauma populations. METHODS: This is a descriptive qualitative analysis of adopting forum theater at a Level I trauma center in a New York City borough with a racially and ethnically diverse population. The implementation of a forum theater workshop was described, including our work with a theater company to address bias in the health care setting. Volunteer staff members and theater facilitators participated in an 8-hr workshop leading to a 2-hr multipart performance. Participant experiences were collected in a postsession debrief to understand the utility of forum theater. RESULTS: Debriefing sessions after forum theater performances demonstrated that forum theater is a more engaging and effective method for dialogue surrounding bias than personal past experiences with other educational models. CONCLUSION: Forum theater was feasible as a tool to enhance cultural competency and bias training. Future research will examine the impact it has on levels of staff empathy and its impact on participants' level of comfort communicating with diverse trauma populations.

Ethics and frontline nursing during COVID-19: A qualitative analysis.

BACKGROUND: Nurses experienced intense ethical and moral challenges during the COVID-19 pandemic. Our 2020 qualitative parent study of frontline nurses' experiences during the COVID-19 pandemic identified ethics as a cross-cutting theme with six subthemes: moral dilemmas, moral uncertainty, moral distress, moral injury, moral outrage, and moral courage. We re-analyzed ethics-related findings in light of refined definitions of ethics concepts. RESEARCH AIM: To analyze frontline U.S. nurses' experiences of ethics during the COVID-19 pandemic. RESEARCH DESIGN: Qualitative analysis using a directed content methodology. PARTICIPANTS AND RESEARCH CONTEXT: The study included 43 nurses from three major metropolitan academic medical centers and one community hospital in the northeastern, mid-Atlantic, midwestern, and western United States. ETHICAL CONSIDERATIONS: Participant privacy and data confidentiality were addressed. FINDINGS: Moral dilemmas arose from many situations, most frequently related to balancing safety and patient care. Moral uncertainty commonly arose from lacking health information or evidence about options. Moral distress occurred when nurses knew the right thing to do, but were prevented from doing so, including with end-of-life issues. Moral injury (accompanied by suffering, shame, or guilt) occurred after doing, seeing, or experiencing wrongdoing, often involving authority figures. Nurses expressed moral outrage at events and people within and outside healthcare. Despite difficult ethical situations, some nurses exemplified moral courage, sometimes by resisting policies they perceived as preventing compassionate care, guided by thinking about what was best for patients. DISCUSSION: This content analysis of ethics-related subthemes revealed conceptual characteristics and clarified distinctions with corresponding exemplars. Conceptual clarity may inform responses and interventions to address ethical quandaries in nursing practice. CONCLUSIONS: Ethics education in nursing must address the moral dilemmas of pandemics, disasters, and other crises. Nurses need time and resources to heal from trying to provide the best care when no ideal option was available.

Assessing sleep health dimensions in frontline registered nurses during the COVID-19 pandemic: implications for psychological health and wellbeing.

The COVID-19 pandemic altered work environments of nurses, yielding high rates of stress and burnout. Potential protective factors, including effective sleep, may influence psychological health and wellbeing. Evidence about sleep in nurses may help develop interventions that mitigate burnout and poor psychological outcomes. A cross sectional survey was distributed across three hospitals to nurses in New York City (NYC). During the first wave of the pandemic (March-April 2020), NYC had the highest incidence of laboratory-confirmed COVID-19 cases (915/100 000) and half of all COVID-related deaths nationwide. Multivariable logistic regression was used to determine associations between Pittsburgh Sleep Quality Index (PSQI) global sleep score, PSQI sleep dimensions, and psychological health (burnout, depression, anxiety, and compassion fatigue), unadjusted and then controlling for individual and professional characteristics. More than half of the participants reported burnout (64%), depression, (67%), and anxiety (77%). Eighty percent of participants had PSQI global scores >5 (poor sleep) (mean 9.27, SD 4.14). Respondents reporting good sleep (PSQI ≤ 5) had over five times the odds of no burnout (OR: 5.65, 95% CI: 2.60, 12.27); increased odds of screening negative for depression (OR: 6.91, 95% CI: 3.24, 14.72), anxiety (OR: 10.75, 95% CI: 4.22, 27.42), and compassion fatigue (OR: 7.88, 95% CI: 1.97, 31.51). Poor subjective sleep quality PSQI subcomponent was associated with burnout (OR: 2.21, 95% CI: 1.41, 3.48) but sleep duration subcomponent was not (OR: 0.84, 95% CI: 0.59, 1.19). Daytime dysfunction was significantly associated with all psychological outcomes. Sleep disturbances and medications yielded higher anxiety odds. Overall, sleep quality appears more strongly related to burnout than sleep duration in nurses working during the COVID-19 pandemic. Sleep interventions should target individual sleep dimensions in nurses.

Emergency Nursing Workforce, Burnout, and Job Turnover in the United States: A National Sample Survey Analysis.

INTRODUCTION: Few studies have examined emergency nurses who have left their job to better understand the reason behind job turnover. It also remains unclear whether emergency nurses differ from other nurses regarding burnout and job turnover reasons. Our study aimed to test differences in reasons for turnover or not currently working between emergency nurses and other nurses; and ascertain factors associated with burnout as a reason for turnover among emergency nurses. METHODS: We conducted a secondary analysis of 2018 National Sample Survey for Registered Nurses data (weighted N = 3,004,589) from Health Resources and Services Administration. Data were analyzed using descriptive statistics, chi-square and t-test, and unadjusted and adjusted logistic regression applying design sampling weights. RESULTS: There were no significant differences in burnout comparing emergency nurses with other nurses. Seven job turnover reasons were endorsed by emergency nurses and were significantly higher than other nurses: insufficient staffing (11.1%, 95% confidence interval [CI] 8.6-14.2, P = .01), physical demands (5.1%, 95% CI 3.4-7.6, P = .44), patient population (4.3%, 95% CI 2.9-6.3, P < .001), better pay elsewhere (11.5%, 95% CI 9-14.7, P < .001), career advancement/promotion (9.6%, 95% CI 7.0-13.2, P = .01), length of commute (5.1%, 95% CI 3.4-7.5, P = .01), and relocation (5%, 95% CI 3.6-7.0, P = .01). Increasing age and increased years since nursing licensure was associated with decreased odds of burnout. DISCUSSION: Several modifiable factors appear associated with job turnover. Interventions and future research should account for unit-specific factors that may precipitate nursing job turnover.

An Intensive Longitudinal Assessment Approach to Surveilling Trajectories of Burnout over the First Year of the COVID Pandemic.

Frontline clinicians responding to the COVID-19 pandemic are at increased risk of burnout, but less is known about the trajectory of clinician burnout as caseloads increase and decrease. Personal and professional resources, including self-efficacy and hospital support, can attenuate the risk of burnout. Yet, empirical data documenting how burnout and resources changed as the pandemic waxed and waned are limited. This intensive longitudinal prospective study employed ecological momentary assessment methods to examine trajectories of burnout and resources over the pandemic's first year in a New York City hospital. A 10-item survey was emailed every 5 days to frontline clinicians (physicians, nurses, and physician assistants). The primary outcome was a single-item validated measure of burnout; predictors included daily hospital COVID-19-related caseloads and personal and professional resources. Clinicians (n = 398) completed the initial survey and an average of 12 surveys over the year. Initially, 45.3% of staff reported burnout; over the year, 58.7% reported burnout. Following the initial COVID peak, caseloads declined, and burnout levels declined. During the second wave of COVID, as caseloads increased and remained elevated and personal and professional resource levels decreased, burnout increased. This novel application of intensive longitudinal assessment enabled ongoing surveillance of burnout and permitted us to evaluate how fluctuations in caseload intensity and personal and professional resources related to burnout over time. The surveillance data support the need for intensified resource allocation during prolonged pandemics.

Initial psychometric properties of the provider-co-management index-RN to scale registered nurse-physician co-management: Implications for burnout, job satisfaction, and intention to leave current position.

Team-based care has become a cornerstone of care delivery to meet the demands of high-quality patient care. Yet, there is a lack of valid and reliable instruments to measure the effectiveness of co-management between clinician dyads, particularly physicians and registered nurses (RNs). The purpose of this study was to adapt an existing instrument, Provider Co-Management Index (PCMI), previously used among primary care providers into a new version to scale RN-physician co-management (called PCMI-RN). We also aimed to explore preliminary associations between RN-physician co-management and burnout, job satisfaction, and intention to leave current job. Face, cognitive, and content validity testing, using mixed methods approaches, were preceded by initial pilot testing (n = 122 physicians and nurses) in an acute care facility. The internal consistency reliability (α=.83) was high. One-quarter of participants reported burnout, 27% were dissatisfied with their job, and 20% reported intention to leave their job. There was a weak significant correlation between co-management and burnout (p = .010), and co-management and job satisfaction (p = .009), but not intention to leave current position. Construct validity testing is recommended. Future research using PCMI-RN may help to isolate factors that support or inhibit effective physician-nurse co-management.

Nursing perspectives about the critical gaps in public health emergency response during the COVID-19 pandemic.

INTRODUCTION: The purpose of this qualitative study was to synthesize frontline U.S. nursing perspectives about the current state of U.S. public health emergency preparedness and response. The study findings may inform public health policy change and improve future national pandemic planning and responses. DESIGN: We conducted a secondary thematic qualitative analysis using grounded theory methodology. METHODS: Data collection occurred through semi-structured, in-depth focus groups between July and December 2020, from 43 frontline nurses working in hospitals in four states (Ohio, California, Pennsylvania, and New York). Data were analyzed deductively, aligned with Khan et al.'s Public Health Emergency Preparedness Framework and inductively for emergent themes. RESULTS: Three themes emerged: (1) Validation of the presence of health disparities and inequities across populations; (2) Perceived lack of consistency and coordination of messaging about pandemic policies and plans across all levels; and (3) challenges securing and allocating nursing workforce resources to areas of need. CONCLUSION: From a frontline nursing perspective, this study demonstrates the critical need to address health inequities and inequalities across populations, a consistent national vehicle for communication, and national plan for securing and allocating nursing workforce resources.

Speech recognition can help evaluate shared decision making and predict medication adherence in primary care setting.

OBJECTIVE: Asthma is a common chronic illness affecting 19 million US adults. Inhaled corticosteroids are a safe and effective treatment for asthma, yet, medication adherence among patients remains poor. Shared decision-making, a patient activation strategy, can improve patient adherence to inhaled corticosteroids. This study aimed to explore whether audio-recorded patient-primary care provider encounters can be used to: 1. Evaluate the level of patient-perceived shared decision-making during the encounter, and 2. Predict levels of patient's inhaled corticosteroid adherence. MATERIALS AND METHODS: Shared decision-making and inhaled corticosteroid adherence were assessed using the SDM Questionnaire-9 and the Medication Adherence Report Scale for Asthma (MARS-A). Speech-to-text algorithms were used to automatically transcribe 80 audio-recorded encounters between primary care providers and asthmatic patients. Machine learning algorithms (Naive Bayes, Support Vector Machines, Decision Tree) were applied to achieve the study's predictive goals. RESULTS: The accuracy of automated speech-to-text transcription was relatively high (ROUGE F-score = .9). Machine learning algorithms achieved good predictive performance for shared decision-making (the highest F-score = .88 for the Naive Bayes) and inhaled corticosteroid adherence (the highest F-score = .87 for the Support Vector Machines). DISCUSSION: This was the first study that trained machine learning algorithms on a dataset of audio-recorded patient-primary care provider encounters to successfully evaluate the quality of SDM and predict patient inhaled corticosteroid adherence. CONCLUSION: Machine learning approaches can help primary care providers identify patients at risk for poor medication adherence and evaluate the quality of care by measuring levels of shared decision-making. Further work should explore the replicability of our results in larger samples and additional health domains.