Can incorrect artificial intelligence (AI) results impact radiologists, and if so, what can we do about it? A multi-reader pilot study of lung cancer detection with chest radiography.

OBJECTIVE: To examine whether incorrect AI results impact radiologist performance, and if so, whether human factors can be optimized to reduce error. METHODS: Multi-reader design, 6 radiologists interpreted 90 identical chest radiographs (follow-up CT needed: yes/no) on four occasions (09/20-01/22). No AI result was provided for session 1. Sham AI results were provided for sessions 2-4, and AI for 12 cases were manipulated to be incorrect (8 false positives (FP), 4 false negatives (FN)) (0.87 ROC-AUC). In the Delete AI (No Box) condition, radiologists were told AI results would not be saved for the evaluation. In Keep AI (No Box) and Keep AI (Box), radiologists were told results would be saved. In Keep AI (Box), the ostensible AI program visually outlined the region of suspicion. AI results were constant between conditions. RESULTS: Relative to the No AI condition (FN = 2.7%, FP = 51.4%), FN and FPs were higher in the Keep AI (No Box) (FN = 33.0%, FP = 86.0%), Delete AI (No Box) (FN = 26.7%, FP = 80.5%), and Keep AI (Box) (FN = to 20.7%, FP = 80.5%) conditions (all ps < 0.05). FNs were higher in the Keep AI (No Box) condition (33.0%) than in the Keep AI (Box) condition (20.7%) (p = 0.04). FPs were higher in the Keep AI (No Box) (86.0%) condition than in the Delete AI (No Box) condition (80.5%) (p = 0.03). CONCLUSION: Incorrect AI causes radiologists to make incorrect follow-up decisions when they were correct without AI. This effect is mitigated when radiologists believe AI will be deleted from the patient's file or a box is provided around the region of interest. CLINICAL RELEVANCE STATEMENT: When AI is wrong, radiologists make more errors than they would have without AI. Based on human factors psychology, our manuscript provides evidence for two AI implementation strategies that reduce the deleterious effects of incorrect AI. KEY POINTS: • When AI provided incorrect results, false negative and false positive rates among the radiologists increased. • False positives decreased when AI results were deleted, versus kept, in the patient's record. • False negatives and false positives decreased when AI visually outlined the region of suspicion.

A Pilot Study of Accelerometer-Based Biometric Data Collection among Patients Undergoing Locoregional Therapies.

This study evaluated the feasibility of measuring patient recovery after locoregional therapies (LRTs) using a wearable activity tracker (WAT). Twenty adult patients with cancer were provided with a WAT device to wear for a minimum of 7 days prior to their procedure (baseline) and for up to 30 days after their procedure (recovery). Daily step counts were continuously recorded. Patient responses to the Short Form 36-Item Health Survey (SF-36) were also collected before and after LRT. Analysis of WAT data demonstrated a mean of 4,850 daily steps taken at baseline, which decreased to 2,000 immediately after LRT and then rapidly increased to approximately 4,300 daily steps over an average of 10 days (P < .001). No significant changes were observed in SF-36 responses between baseline and follow-up assessments (P > .10). These results suggest that WAT devices capture dynamic periprocedural data not reflected in survey-based assessments and may be used to monitor patient recovery after interventional oncologic procedures.

Digital Breast Tomosynthesis and Digital Mammography Recall and False-Positive Rates by Time of Day and Reader Experience.

Background Digital breast tomosynthesis (DBT) image interpretation might be more cognitively demanding than interpretation of digital mammography (DM) images. The time of day of interpretation might affect recall and false-positive (FP) rates, especially for DBT. Purpose To determine whether recall and FP rates vary by time of day of interpretation for screening mammography for breast cancer performed with DM and DBT. Materials and Methods This is a retrospective study examining 97 671 screening mammograms interpreted by 18 radiologists between January 2018 and December 2019 at one of 12 community radiology sites. The association between the time of day of interpretation, the type of image interpreted (DM vs DBT), and radiologist experience (≤5 posttraining years vs >5 posttraining years) and the likelihood of a patient being recalled from screening mammography and the likelihood of whether the interpretation was FP or true positive were analyzed. Analyses were conducted using generalized linear mixed modeling with a binary distribution and sandwich estimation where observations were nested by radiologist. Results Screening mammograms interpreted by 18 radiologists were reviewed (40 220 DBTs, 57 451 DMs). Nine radiologists had 5 or fewer posttraining years of experience, and nine had more than 5 posttraining years of experience. The overall recall rates for DM (10.2%) and DBT (9.0%) were different (P = .006); FP rate also differed (9.8% DM, 8.6% DBT; P = .004). For radiologists with 5 or fewer posttraining years of experience, odds of recall increased 11.5% (odds ratio [OR] = 1.12, P = .01) with every hour when using DBT, but this was not found for DM (OR = 1.09, P = .06); DBT and DM were different (OR = 1.12 vs 1.09, P = .02). For radiologists with more than 5 posttraining years of experience, no evidence of increase in recall was observed for DBT (OR = 1.02, P = .27) or DM (OR = 1.0, P = .80), and there was no evidence that these were different (OR = 1.02 vs 1.0, P = .13). Conclusion Patients were more likely to be recalled when their screening digital breast tomosynthesis images were interpreted later in the day by less-experienced radiologists. © RSNA, 2022 Online supplemental material is available for this article.

How do US orthopaedic surgeons view placebo-controlled surgical trials? A pilot online survey study.

Randomised placebo-controlled trials (RPCTs) are the gold standard for evaluating novel treatments. However, this design is rarely used in the context of orthopaedic interventions where participants are assigned to a real or placebo surgery. The present study examines attitudes towards RPCTs for orthopaedic surgery among 687 orthopaedic surgeons across the USA. When presented with a vignette describing an RPCT for orthopaedic surgery, 52.3% of participants viewed it as 'completely' or 'mostly' unethical. Participants were also asked to rank-order the value of five different types of evidence supporting the efficacy of a surgery, ranging from RPCT to an anecdotal report. Responses regarding RPCTs were polarised with 26.4% viewing it as the least valuable (even less valuable than an anecdote) and 35.7 .% viewing it as the most valuable. Where equipoise exists, if we want to subject orthopaedic surgeries to the highest standard of evidence (RPCTs) before they are implemented in clinical practice, it will be necessary to educate physicians on the value and ethics of placebo surgery control conditions. Otherwise, invasive procedures may be performed without any benefits beyond possible placebo effects.

Artificial Intelligence and the Future of Primary Care: Exploratory Qualitative Study of UK General Practitioners' Views.

BACKGROUND: The potential for machine learning to disrupt the medical profession is the subject of ongoing debate within biomedical informatics and related fields. OBJECTIVE: This study aimed to explore general practitioners' (GPs') opinions about the potential impact of future technology on key tasks in primary care. METHODS: In June 2018, we conducted a Web-based survey of 720 UK GPs' opinions about the likelihood of future technology to fully replace GPs in performing 6 key primary care tasks, and, if respondents considered replacement for a particular task likely, to estimate how soon the technological capacity might emerge. This study involved qualitative descriptive analysis of written responses ("comments") to an open-ended question in the survey. RESULTS: Comments were classified into 3 major categories in relation to primary care: (1) limitations of future technology, (2) potential benefits of future technology, and (3) social and ethical concerns. Perceived limitations included the beliefs that communication and empathy are exclusively human competencies; many GPs also considered clinical reasoning and the ability to provide value-based care as necessitating physicians' judgments. Perceived benefits of technology included expectations about improved efficiencies, in particular with respect to the reduction of administrative burdens on physicians. Social and ethical concerns encompassed multiple, divergent themes including the need to train more doctors to overcome workforce shortfalls and misgivings about the acceptability of future technology to patients. However, some GPs believed that the failure to adopt technological innovations could incur harms to both patients and physicians. CONCLUSIONS: This study presents timely information on physicians' views about the scope of artificial intelligence (AI) in primary care. Overwhelmingly, GPs considered the potential of AI to be limited. These views differ from the predictions of biomedical informaticians. More extensive, stand-alone qualitative work would provide a more in-depth understanding of GPs' views.

Do what you say and say what you are going to do: A preliminary meta-analysis of client change and sustain talk subtypes in motivational interviewing.

Objective: This meta-analysis examines the predictive validity of client change language subtypes in motivational interviewing (MI) sessions addressing addictive behavior change. Method: A systematic review identified k = 13 primary studies, contributing 16 MI conditions (N = 1556). The pooled correlation coefficient was used to assess the significance, direction, and strength of seven language subtypes (i.e., reason, desire, need, ability, commitment, taking steps, and other) by three valences (i.e., frequency positive or change talk, frequency negative or sustain talk, and proportion change talk) and their relationship to subsequent engagement in addictive behavior. Results: For frequency measures, more sustain talk related to reason, desire, ability, and other were associated with more addictive behavior at follow up. Other change talk was associated with MI outcomes but in an unexpected direction (i.e., more addictive behavior). Proportion measures showed more proportion change talk-reason and -other statements were associated with less addictive behavior at follow up. Sensitivity analyses indicated some heterogeneity and instability of effect sizes, but no evidence of publication bias. Conclusions: This preliminary meta-analysis suggests that aggregate measures of change and sustain talk are comprised of statement subtypes that are not equally meaningful in predicting outcome following MI for addictive behavior change.

Effect of anticipatory stress on placebo alcohol consumption in a bar laboratory.

BACKGROUND: Only one laboratory study has examined the relation between stress and alcohol use in an ecologically valid drinking context. In that prior study, drinking was measured after the stressful situation. OBJECTIVE: To examine the effect of an anticipatory stressor, and trait social anxiety on "alcohol" consumption in a bar laboratory. METHODS: College students aged 18 and older (N = 127) in same-sex groups of two or three participants took part in a study that ostensibly examined alcohol's effect on language fluency. Using a between-subjects design, participants were randomly assigned to a stress or control condition. Participants in the stress condition anticipated giving a stressful speech for the fluency procedure, whereas those in the control group anticipated a nonstressful activity. Before the alleged fluency task, groups could order and consume mixed drinks ad lib in a bar laboratory. No beverages actually contained alcohol, but we used a validated procedure to ensure that participants included in these analyses were deceived. Primary analyses were performed with a hierarchical linear model (HLM) due to a substantial group/modeling effect. RESULTS: Counter to expectations, participants in the control group consumed more placebo alcohol than those in the stress condition. This main effect was moderated by past 3 months' drinks per week, such that the effect was attenuated (or reversed) among heavy drinkers. No main or interaction effects were observed for trait social anxiety. CONCLUSIONS: Some stressors (i.e., those invoking performance anxiety) may decrease consumption. People with higher levels of alcohol involvement might be especially motivated to drink for tension reduction purposes.

A Novel Approach for Streamlining Delivery of Brief Motivational Interventions to Mandated College Students: Using Group and Individual Sessions Matched to Level of Risk.

BACKGROUND: Mandated college students (those in violation of a campus alcohol policy) are heterogenous with respect to alcohol consumption. Thus, when universities consider required treatment for mandated students, one promising option is to match treatment intensity according to level of alcohol involvement. The present study evaluates such an approach with minimal resources. METHODS: Mandated students (N = 285) were required to complete a baseline assessment. Participants identified as high-risk (5+ past month alcohol problems AND 2+ past month heavy drinking episodes) received a one session individual Brief Motivational Intervention (I-BMI) whereas those identified as low-risk (all others) received a one session group Brief Motivational Intervention (G-BMI). I-BMI and G-BMI sessions were delivered by doctoral students in Clinical Psychology. Follow-up assessments were collected 1 month post-intervention (response rate = 73%). RESULTS: The vast majority of students complied with their requirement. Participants assigned to I-BMI reported an 82% reduction in drinks per week, a 58% reduction in heavy episodic drinking, and a 74% reduction in alcohol-related problems at a 1 month follow-up. Participants assigned to G-BMI reported a 61% reduction in drinks per week and a 42% reduction in alcohol-related problems at follow-up relative to baseline. Conclusion/Importance: We demonstrate that matching intervention intensity on baseline alcohol involvement with mandated students is feasible and associated with short-term reductions in alcohol use and consequences. Universities may wish to consider the procedure outlined here as a way of allocating more resources to those who drink at problematic levels.

The Effectiveness of Message Framing and Temporal Context on College Student Alcohol Use and Problems: A Selective E-Mail Intervention.

AIMS: Only one study has examined message framing on college drinking, but did so in a laboratory setting among a general sample of college students. The current study was designed to: (a) compare the efficacy of emailed interventions differing by message framing and temporal context on alcohol involvement among heavy drinking college students and (b) examine need for cognition (NFC), consideration of future consequences (CFC) and self-efficacy as putative moderators. METHODS: Hazardous drinking college students (N = 220) were randomly assigned to conditions in a 2 (Frame: gain vs. loss) × 2 (Temporal Context: long-term vs. short-term consequences) factorial design. Participants received four emails on heavy drinking consequences phrased in a manner consistent with their condition. After each message, participants were given a manipulation check. Participants were sent a 1-month follow-up assessment. Primary outcome measures were heavy episodic drinking (HED) and alcohol-related problems. We hypothesized two main effects (less alcohol consumption in the gain-frame and short-term condition), qualified by a Frame × Temporal Context interaction with substantially less alcohol involvement in the gain-frame/short-term condition. RESULTS: There was very little study attrition (96.4% completed follow-up survey, 93.2-99.5% completed manipulation checks), and strong effects were observed for the manipulations. A 2 × 2 ANCOVA, controlling for baseline alcohol involvement, revealed no consistent main effects or interactions on either outcome. No moderation was observed for any putative moderator. CONCLUSIONS: These results do not replicate prior laboratory-based research. The null findings may be attributed to the heavy drinking sample or electronic means of message delivery.

Do bisexual girls report higher rates of substance use than heterosexual girls? A failure to replicate with incarcerated and detained youth.

Prior research suggests that sexual minority females, particularly bisexuals, report greater rates of substance use than heterosexuals. However, to our knowledge, no study has compared alcohol/drug use between bisexual and heterosexual incarcerated or detained female youth. The current study pools data from three prior treatment studies with incarcerated or detained adolescent girls that self-identify as bisexual or heterosexual (N=86). Hierarchical regression models were conducted to determine whether 12-month prevalence of alcohol, cigarettes, marijuana, or other drug use differed between bisexual and heterosexual participants. In contrast to most prior work, no differences were observed. Findings are considered in light of the recruitment setting, which drew a sample with high levels of substance use prevalence.

Hostility and cigarette use: a comparison between smokers and nonsmokers in a matched sample of adolescents.

INTRODUCTION: We examined the association between hostility-a personality trait reflective of negativity and cynicism toward others-and smoking in adolescents by measuring (a) several subcomponents of hostility, and (b) facial emotion processing ability, which has been previously linked to hostility. METHODS: Participants (N = 241 aged 14-19) were 95 smokers and 95 demographically matched nonsmokers as well as 51 nonmatched smokers. All participants completed the Cook-Medley (C-M) hostility scale, which provides a general hostility score and 3 component scores (cynicism, hypersensitivity, and aggressive responding), and a facial emotion processing task. This task, designed to assess emotion recognition, requires quickly identifying the emotion of faces that gradually morph from neutral to high-intensity happy, angry, or fearful. RESULTS: Independent sample t tests indicated that matched smokers scored significantly higher in cynicism and aggressive responding than nonsmokers. Among smokers, age of smoking onset was negatively correlated with general hostility and aggressive responding. All hostility scales were positively correlated with the intensity needed to recognize happy faces. Counterintuitively, smokers required a greater intensity to recognize angry faces than nonsmokers. No other relations between hostility/smoking status and facial emotion processing were observed. CONCLUSIONS: Aspects of hostility, particularly aggressive responding, may be a risk factor for early onset smoking. Although hostile participants exhibited a deficiency in their ability to recognize happiness in facial pictures, these results did not translate to differences in smoking status. This study elucidates some of the complex interrelations between hostility, emotion processing, and adolescent smoking, which may have implications for teen smoking prevention.

Should I Stay, or Should I Go? Early Phase Instrument Development of Workforce Movement-A Pilot Study with Breast Radiologists.

OBJECTIVE: The goal of this study was to develop a psychometrically valid survey on workplace satisfaction and examine predictors of workforce movement among breast radiologists. METHODS: Actively practicing members of the Society of Breast Imaging were invited to complete a survey on workplace satisfaction. Radiologists also indicated whether they had recently left their practice or were thinking of leaving their practice. RESULTS: In total, 228 breast radiologists provided valid responses (8.7% response rate); 45% were thinking of leaving or had left their practice. Factor analysis yielded five factors, and discriminant function analysis found six main aspects associated with workforce movement in breast radiologists: (1) not enough work-life balance; (2) salary too low; (3) not feeling valued; (4) wanting a different challenge and/or more growth opportunity; (5) safety concerns; and (6) not feeling respected by physician leadership. CONCLUSIONS: Pending further validation in larger and different cohorts, the survey created here can be administered by radiology practices to predict when breast radiologists are vulnerable to quitting. Atlhough this measure was designed for breast radiologists specifically, it could be adapted for other subspecialties.

How beliefs about coronavirus disease (COVID) influence COVID-like symptoms? - A longitudinal study.

OBJECTIVE: Evidence from psychosomatic and nocebo research has indicated that believing one will develop symptoms makes the experience of such symptoms more likely. We applied this idea in the context of coronavirus disease 2019 (COVID-19). Specifically, we assessed whether beliefs regarding COVID-19 predict COVID-like symptoms 3-4 weeks later, and what specific belief has the greatest influence on symptom experience. METHOD: We conducted two studies with over 300 participants, approached at two successive timepoints, 3-4 weeks apart. Participants reported their experienced symptoms, COVID-19-related beliefs, demographics, and state anxiety. To target COVID-like symptoms, participants who reported having contracted COVID-19 or attributed their symptoms to another known cause were excluded. Regression analyses were conducted to test the predictive value of beliefs regarding COVID-19 on experienced symptoms. RESULTS: A particular belief regarding one's estimated symptom severity if infected with coronavirus predicted the experience of symptoms 3-4 weeks later (ß = .17, p = .011). This result persisted after controlling for potential confounds, including state anxiety (ß = .22, p = .002). Findings were preregistered and replicated in a separate cohort. A novel scale for perception of the body's ability to fight diseases contributed to mediating the effect of estimated symptom severity on later experienced symptoms. CONCLUSIONS: A particular belief about estimated symptom severity if infected with COVID predicted the experience of COVID-like physical symptoms several weeks later. These findings contribute to the understanding of the development of unexplained physical symptoms. Furthermore, identification of a particular belief that increases the likelihood of symptoms informs intervention that may mitigate its effects. (PsycInfo Database Record (c) 2022 APA, all rights reserved).

Heavy Smoking Patients Receiving a Lung Cancer Screen Want to Quit: A Call for Tailored Cessation Interventions.

Background: Lung cancer screening for current or former heavy smokers is now recommended among all asymptomatic adults 50-80 years old with a 20 pack-year history of smoking. However, little is known about the smoking-related attitudes of this population. Method: An assessment was conducted among 1,472 current smokers who presented for an annual lung cancer screen at one of 12 diagnostic imaging sites in Rhode Island between April 2019 and May 2020. Patients were asked about their use of smoking products, interest in quitting, and smoking-related attitudes. Results: Patients smoked a median of 16 cigarettes per day; 86.6% were daily cigarette smokers and 30.1% were daily cigar smokers. In total, 91.4% of patients were, to some degree, interested in quitting smoking and 71.4% were seriously thinking about quitting in the next 6 months or sooner. Patients planned on smoking less regardless of whether their lung screen was positive or negative for cancer, though they were more likely to plan on smoking less if negative (on 0-3 pt Likert scale: 0.31, 95% CI [0.27, 0.34] vs. 0.77, 95% CI [0.72, 0.81]). Confidence in quitting and belief in one's inherent ability to quit smoking varied substantially within the sample. Conclusion: Nearly all current smokers receiving a lung cancer screen have some interest in smoking cessation. Due to the heterogeneity in some smoking-related attitudes, tailored interventions for this population should be tested.

Electronic Cigarettes for Smoking Cessation: The Gap Between Behavior in Smokers and Medical Education.

INTRODUCTION: E-cigarettes have engendered a great deal of controversy within the public health and medical communities.  Methods: Two cross-sectional surveys were administered. First, patients at an annual lung cancer screening appointment who self-identified as former smokers were asked about strategies for achieving and maintaining smoking cessation with open-ended questions. Second, medical students at a single university reported their opinion and knowledge of combustible cigarettes and e-cigarettes. RESULTS: Among the n=102 in the patient survey indicating that they used e-cigarettes or over-the-counter (OTC) nicotine replacement products for smoking cessation, 34.3% (35/102) vaped e-cigarettes, making it the second most common next to patches (47.1% {48/102}). By comparison, n=48 reported using medication. Medical student participants (n=168) were mixed regarding whether a patient should switch from traditional to electronic cigarettes (56.0% yes; 44.0% no) and reported receiving education about traditional cigarettes (92.3%) at a much higher rate than for e-cigarettes (46.4%), p<.001. CONCLUSION: Many former heavy smokers undergoing a lung cancer screen used e-cigarettes to achieve smoking cessation. However, nearly half of medical students surveyed do not think patients should switch from traditional to e-cigarettes.