An Ethically Supported Framework for Determining Patient Notification and Informed Consent Practices when Using Artificial Intelligence in Healthcare.

Artificial intelligence (AI) is increasingly being used in healthcare, and without an ethically supportable, standard approach to knowing when patients should be informed about AI, hospital systems and clinicians run the risk of fostering mistrust among their patients and the public. Therefore, hospital leaders need guidance on when to tell patients about the use of AI in their care. In this paper we provide such guidance. To determine which AI technologies fall into each of the identified categories (no notification/no informed consent (IC), notification-only, and formal IC), we propose that AI use-cases should be evaluated using the following criteria: 1) AI model autonomy; 2) departure from standards of practice; 3) whether or not the AI model is patient facing; 4) clinical risk introduced by the model; and 5) administrative burdens. We take each of these in turn, using a case example of AI in healthcare to illustrate our proposed framework. As AI becomes more commonplace in healthcare, our proposal may serve as a starting point for creating consensus on standards for notification and IC for the use of AI in patient care.

Virtual prenatal visits associated with high measures of patient experience and satisfaction among average-risk patients: a prospective cohort study.

BACKGROUND: Virtual visits have the potential to decrease barriers to prenatal care stemming from transportation, work, and childcare concerns. However, data regarding patient experience and satisfaction with virtual visits remain limited in obstetrics. To address this gap, we explore average-risk pregnant women's experiences with virtual visits and compare satisfaction with virtual vs. in-person visits as a secondary aim. METHODS: In this IRB-approved, prospective cohort study, we surveyed pregnant women after their first virtual visit between October 7, 2019 and March 20, 2020. Using heterogeneous purposive sampling, we identified a subset of respondents with diverse experiences and opinions for interviews. For comparison, Consumer Assessment of Healthcare Providers and Systems (CAHPS) satisfaction data were collected after in-person visits during the study timeframe from a control cohort with the same prenatal providers. Logistic regression controlling for age, previous pregnancies, and prior live births compared satisfaction data between virtual and in-person visits. Other quantitative survey data were analyzed through descriptive statistics. Free text survey responses and interview data were analyzed using content analysis. RESULTS: Ninety five percent (n = 165/174) of surveys and 90% (n = 18/20) of interviews were completed. Most participants were Caucasian, married, and of middle to high income. 69% (114/165) agreed that their virtual appointment was as good as in-person; only 13% (21/165) disagreed. Almost all (148/165, 90%) would make another virtual appointment. Qualitative data highlighted ease of access, comparable provider-patient communication, confidence in care quality, and positive remote monitoring experiences. Recognizing these advantages but also inherent limitations, interviews emphasized interspersing telemedicine with in-person prenatal encounters. CAHPS responses after in-person visits were available for 60 patients. Logistic regression revealed no significant difference in three measures of satisfaction (p = 0.16, 0.09, 0.13) between virtual and in-person visits. CONCLUSIONS: In an average-risk population, virtual prenatal visits provide a patient-centered alternative to traditional in-person encounters with high measures of patient experience and no significant difference in satisfaction. Obstetric providers should explore telemedicine to improve access - and, during the ongoing pandemic, to minimize exposures - using patients' experiences for guidance. More research is needed regarding virtual visits' medical quality, integration into prenatal schedules, and provision of equitable care for diverse populations.

Informed or anxious: patient preferences for release of test results of increasing sensitivity on electronic patient portals.

Worldwide, governments and healthcare systems are moving towards increased transparency to improve care quality, increase patient engagement, and decrease costs. For example, the American 21st Century Cures Act Final Rule requires providers to grant patients access to their electronic medical record. Unfortunately, limited research guides release of test results to online patient portals, especially concerning emotionally sensitive information. To address this gap, we surveyed the largest patient sample published to date. This cross-sectional survey project was conducted by the Market Research & Insights and Office of Patient Experience departments at a large academic medical center. Data were analyzed in SPSS using descriptive statistics and Z-tests. Of 8030 respondents, 74% and 57% accepted first learning their results online for cholesterol and strep throat tests, respectively. Most prefer in-person appointments for more serious tests detecting cancer (54%) and fetal miscarriage (53%). Excluding sexually transmitted disease (STD) testing, there are no clinically significant differences in preference between respondents previously diagnosed with the condition in question and respondents without such experience. When weighing the possibility of a 3-week wait to hear from their provider, most patients want automatic release of cholesterol (94%), strep throat (90%), genetic (68%), and STD (60%) test results, but the majority say it is unacceptable to receive Alzheimer's (52%), fetal miscarriage (51%), and cancer (59%) test results this way. Electronic results release is acceptable for less serious tests, but not for more consequential tests. Providers should consider patient preferences when developing policies to increase healthcare transparency. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1007/s12553-021-00628-5.

Ophthalmology Provider Ratings and Patient, Disease, and Appointment Factors.

The purpose of the current study is to examine how nonmodifiable sociodemographic, disease, appointment, management, and survey factors correlate with provider rating. This was a retrospective cross-sectional study conducted on 29 857 patient Clinician and Group Consumer Assessment of Healthcare Providers and Systems surveys collected from January 2017 to January 2019 at a tertiary eye center. We included surveys of patients aged 18 years or older, who answered at least 4 of 6 subfield questions, and completed the survey within 90 days of the appointment. The main outcome was the odds of receiving top box score (TBS) of 10/10 on the survey question regarding overall provider rating. The results showed that the variables with higher odds of TBS included higher overall appointment attendance (odds ratio [OR]: 2.66 [95% CI: 1.23-5.75], P = .013); older patient age (OR 2.44 [95% CI: 2.08-2.87], P < .001]; higher percentage of survey questions completed (OR: 2.02 [95% CI: 1.79-2.27], P < .001); better best corrected visual acuity (OR: 1.85 [95% CI: 1.3-2.64], P = .001); optometry clinic visit (OR: 1.25 [95% CI: 1.15-1.36], P < .001); having procedures (OR: 1.19 [95% CI: 1.04-1.36], P = .013), surgery scheduled (OR: 1.18 [95% CI: 1.03-1.36], P = .020], or refraction done (OR: 1.16 [95% CI: 1.08-1.25], P < .001); being seen by male providers (OR: 1.11 [95% CI: 1.04-1.17], P = .001); and having additional eye testing performed (OR: 1.06 [95% CI: 1.00-1.13], P = .048). Variables associated with lower odds of TBS included longer time to complete survey (OR: 0.42 [95% CI: 0.3-0.58], P = .001); new patient encounter (OR: 0.62 [95% CI: 0.58-0.65], P < .001); and glaucoma (OR: 0.66 [95% CI: 0.59-0.75], P < .001), cornea (OR: 0.79 [95% CI: 0.71-0.87], P < .001), or comprehensive clinic visits (OR: 0.86 [95% CI: 0.79-0.94], P < .001). Thus, nonmodifiable factors may affect the provider rating, and these factors should be studied further and accounted for when interpreting the results of patient experience surveys.

Categorizing and Rating Patient Complaints: An Innovative Approach to Improve Patient Experience.

The Ombudsman Office at a large academic medical center created a standardized approach to manage and measure unsolicited patient complaints, including methods to identify longitudinal improvements, accounting for volume variances, as well as incident severity to prioritize response needs. Data on patient complaints and grievances are collected and categorized by type of issue, unit location, severity, and individual employee involved. In addition to granular data, results are collated into meaningful monthly leadership reports to identify opportunities for improvement. An overall benchmark for improvement is also applied based on the number of complaints and grievances received for every 1000 patient encounters. Results are utilized in conjunction with satisfaction survey results to drive patient experience strategies. By applying benchmarks to patient grievances, targets can be created based on historical performance. The utilization of grievance and complaint benchmarking helps prioritize resources to improve patient experiences.

Incorporating Stakeholder Perspectives on Scarce Resource Allocation: Lessons Learned from Policymaking in a Time of Crisis.

The coronavirus disease (COVID-19) crisis provoked an organizational ethics dilemma: how to develop ethical pandemic policy while upholding our organizational mission to deliver relationship- and patient-centered care. Tasked with producing a recommendation about whether healthcare workers and essential personnel should receive priority access to limited medical resources during the pandemic, the bioethics department and survey and interview methodologists at our institution implemented a deliberative approach that included the perspectives of healthcare professionals and patient stakeholders in the policy development process. Involving the community more, not less, during a crisis required balancing the need to act quickly to garner stakeholder perspectives, uncertainty about the extent and duration of the pandemic, and disagreement among ethicists about the most ethically supportable way to allocate scarce resources. This article explains the process undertaken to garner stakeholder input as it relates to organizational ethics, recounts the stakeholder perspectives shared and how they informed the triage policy developed, and offers suggestions for how other organizations may integrate stakeholder involvement in ethical decision-making as well as directions for future research and public health work.

Making the most of the first prenatal visit: The challenge of expanding prenatal genetic testing options and limited clinical encounter time.

OBJECTIVE: Advances in prenatal genetics place additional challenges as patients must receive information about a growing array of screening and testing options. This raises concerns about how to achieve a shared decision-making process that prepares patients to make an informed decision about their choices about prenatal genetic screening and testing options, calling for a reconsideration of how healthcare providers approach the first prenatal visit. METHODS: We conducted interviews with 40 pregnant women to identify components of decision-making regarding prenatal genetic screens and tests at this visit. Analysis was approached using grounded theory. RESULTS: Participants brought distinct notions of risk to the visit, including skewed perceptions of baseline risk for a fetal genetic condition and the implications of screening and testing. Participants were very concerned about financial considerations associated with these options, ranking out-of-pocket costs on par with medical considerations. Participants noted diverging priorities at the first visit from those of their healthcare provider, leading to barriers to shared decision-making regarding screening and testing during this visit. CONCLUSION: Research is needed to determine how to restructure the initiation of prenatal care in a way that best positions patients to make informed decisions about prenatal genetic screens and tests.

Association Between Public Trust and Provider Specialty Among Physicians With Financial Conflicts of Interest.

OBJECTIVES: To characterize public perception of physicians' conflicts of interest (COIs) across medical and surgical specialties. PATIENTS AND METHODS: A cross-sectional 6-arm randomized survey of a nonprobability sample from Amazon's Mechanical Turk occurred on December 11 to 16, 2018. Survey respondents were randomly assigned to vignettes that varied the physician specialty with COI. The primary outcome was mean difference in Mayer Trust, and the secondary outcome included the proportion who desire to discontinue care. RESULTS: There were 1729 of 1920 respondents who completed the experiment (90.1% completion rate). Respondents were male (52.5%; n=907), white (71.4%; n=1234), and between the ages of 25 and 44 years (70.9%; n=1227). Mean ± SD Mayer Trust across the 6 specialties was 3.7±.60, with the only between-specialty differences observed for psychiatry compared with the other specialties (F=5.4; P<.001). The median dollar amount that would affect respondents' trust in a physician was $5000 (interquartile range, $100-$100,000). A total of 75.1% (n=1298) of respondents desired COI information, with 41.6% (n=720) discontinuing care. Age older than 34 years (adjusted odds ratio [aOR], 0.7; 95%, CI, 0.49-0.99; P=.047), nonwhite race (aOR, 1.3; 95% CI, 1.02-1.6; P=.03), educational attainment of 4 or more years of college (aOR, 1.31; 95% CI, 1.05-1.6; P=.016), and physician specialty as a psychiatrist (aOR, 1.5; 95% CI, 1.03-2.2; P=.034) were predictors for discontinuing care. CONCLUSION: Public COI disclosure is a common method for managing financial conflicts. Although survey respondents were more likely to discontinue care with a physician with COI, they will act on this knowledge of COI differently depending on the specialty of the physician. The finding that psychiatry is an outlier may be a chance finding that warrants further confirmation. Continued efforts to ensure best practices for disclosure are required.

Poverty: Not a Justification for Banning Physician-Assisted Death.

Many critics of the legalization of physician-assisted death oppose it in part because they fear it will further disadvantage those who are already economically disadvantaged. This argument points to a serious problem of how economic considerations can influence medical decisions, but in the context of PAD, the concern is not borne out. We will provide empirical evidence suggesting that concerns about money influence medical decisions throughout the full course of illness, but at the end of life, financial pressure is much more likely to influence a decision to pursue or reject aggressive life-extending care than it is to influence a request for PAD. We will also address the question of whether financial pressure as a result of being poor-particularly in the context of an inadequate social safety net-robs people of their autonomy, rendering their informed request and consent invalid. We argue that it does not. We will emphasize the impracticality and injustice of rejecting the role of poverty as a legitimate factor in decision-making, the irrational distinction between PAD and withdrawal of life-sustaining care, and the more appropriate focus on the great flaws in the American health care system.

Use of Visual Decision Aids in Physician-Patient Communication: A Pilot Investigation.

INTRODUCTION: A risk calculator paired with a personalized decision aid (RC&DA) may foster shared decision-making in primary care. We assessed the feasibility of using an RC&DA with patients in a primary care outpatient clinic and patients' experiences regarding communication and decision-making. METHODS: This pilot study was conducted with 15 patients of 3 primary care physicians at a clinic within a tertiary medical center. An atherosclerotic cardiovascular disease (ASCVD) risk calculator was used to generate a personalized RC&DA that displayed absolute 10-year risk information as an icon array graphic. Patient perceptions of utility of the RC&DA, preferences for decision-making, and uncertainty with risk reduction decisions were measured with a semi-structured interview. RESULTS: Patients reported that the RC&DA was easy to understand and knowledge gained was useful to modify their ASCVD risk. Patients used the RC&DA to make decisions and reported low uncertainty with those decisions. CONCLUSIONS: Our findings demonstrate the feasibility of, and positive patient experiences related to using, an RC&DA to facilitate shared decision-making between physicians and patients in an outpatient primary care setting.

Decision-Making for an Incapacitated Pregnant Patient.

Decisions about continuing or terminating a pregnancy touch on profound, individualized questions about bodily integrity, reproductive autonomy, deeply held values regarding one's capacity for parenthood, and, in the case of a high-risk pregnancy, the risks one is willing to take to have a baby. So far as possible, reproductive decisions are made between a patient, in some cases her partner, and her medical provider. However, this standard framework cannot be applied if the patient lacks decision-making capacity. In this essay, we discuss one such case that came before our clinical ethics team. We describe the challenges of respecting a patient's reproductive preferences when the patient cannot share what those preferences are, and we argue that decisions regarding reproductive health care should not be treated with exceptionalism. Rather, they should proceed under the normal processes of surrogate decision-making, including the application of substituted judgment. This approach enables us to take the patient's values into account when considering the questions implicated in reproductive health care, just as we do for other kinds of health care decisions in which a patient's deeply held values are salient.

A new survey to evaluate conflict of interest policies at academic medical centers.

BACKGROUND: A majority of academic medical centers (AMCs) have now adopted conflict of interest policies (COI) to address relationships with pharmaceutical and device industries that can increase the risk of bias in patient care, education and research. However, AMCs may have little information on the impact of their policies. This paper provides a new method, which is a free, publicly-available survey, to fill this information gap and improve COI programs at AMCs. METHODS & FINDINGS: The survey, piloted in three AMCs and designed in collaboration with national conflicts of interest policy experts, covers a range of universal compliance-related concerns, which allows institutions to tailor questions to align with their own policies and culture. The survey was low-burden, and provided important data for these AMCs to evaluate their policies. A descriptive analysis of the pooled pilot site data (n = 1578) was performed, which found that a majority of respondents did not have financial ties with industry and a majority was satisfied with specific COI policies at their institutions. The analysis also showed that the survey is sensitive to differences that AMCs will find meaningful. For instance, individuals with industry ties were significantly more likely than individuals without ties to think that COI policies unnecessarily hindered interactions with industry (p = .004), were ineffective at reducing harm to patients (p < .001), and were ineffective in reducing bias in medical education (p>.001). CONCLUSION: The survey is now free and publicly available for use by any institution. AMCs can use the results to update and refine policies, and to provide ongoing education regarding existing policies.

Patient Advocacy Organizations, Industry Funding, and Conflicts of Interest.

Importance: Patient advocacy organizations (PAOs) are influential health care stakeholders that provide direct counseling and education for patients, engage in policy advocacy, and shape research agendas. Many PAOs report having financial relationships with for-profit industry, yet little is known about the nature of these relationships. Objective: To describe the nature of industry funding and partnerships between PAOs and for-profit companies in the United States. Design, Setting, and Participants: A survey was conducted from September 1, 2013, to June 30, 2014, of a nationally representative random sample of 439 PAO leaders, representing 5.6% of 7865 PAOs identified in the United States. Survey questions addressed the nature of their activities, their financial relationships with industry, and the perceived effectiveness of their conflict of interest policies. Main Outcomes and Measures: Amount and sources of revenue as well as organizational experiences with and policies regarding financial conflict of interest. Results: Of the 439 surveys mailed to PAO leaders, 289 (65.8%) were returned with at least 80% of the questions answered. The PAOs varied widely in terms of size, funding, activities, and disease focus. The median total revenue among responding organizations was $299 140 (interquartile range, $70 000-$1 200 000). A total of 165 of 245 PAOs (67.3%) reported receiving industry funding, with 19 of 160 PAOs (11.9%) receiving more than half of their funding from industry. Among the subset of PAOs that received industry funding, the median amount was $50 000 (interquartile range, $15 000-$200 000); the median proportion of industry support derived from the pharmaceutical, device, and/or biotechnology sectors was 45% (interquartile range, 0%-100%). A total of 220 of 269 respondents (81.8%) indicated that conflicts of interest are very or moderately relevant to PAOs, and 94 of 171 (55.0%) believed that their organizations' conflict of interest policies were very good. A total of 22 of 285 PAO leaders (7.7%) perceived pressure to conform their positions to the interests of corporate donors. Conclusions and Relevance: Patient advocacy organizations engage in wide-ranging health activities. Although most PAOs receive modest funding from industry, a minority receive substantial industry support, raising added concerns about independence. Many respondents report a need to improve their conflict of interest policies to help maintain public trust.

Content and Outcomes of Social Work Consultation for Patients with Diabetes in Primary Care.

PURPOSE: Social workers are positioned to address social determinants of health (SDHs), but their specific roles in outpatient primary care practice have not been well described. We aimed to describe needs of patients with diabetes addressed during social work (SW) consultations and their impact on disease control. METHODS: This study was a retrospective review of electronic medical records of 977 patients with diabetes with a SW consultation at 3 primary care internal medicine sites in 2014. Diabetes and cardiovascular (CV) risk factor control were assessed before and after the SW encounter. Patient subgroups with uncontrolled diabetes or CV risk factors were compared with propensity-matched patients without a SW encounter. Of the 977 records, 300 were randomly selected for abstraction of needs addressed at the SW consultation using SDH categories established by Wilkinson and Marmot. RESULTS: Patient insurance status included 52% Medicare and 32% Medicaid. The SDHs most often addressed were social gradient (67%; obtaining medications or health insurance) and social support (25%). Among our total population, there were no significant improvements in glycosylated hemoglobin (HbA1c), low-density lipoprotein (LDL) cholesterol, systolic blood pressure, or body mass index at least 3 months after the first SW consultation. For patients with uncontrolled diabetes (HbA1c >9% or LDL cholesterol >130 mg/dl), HbA1c improved by 1.5 versus 1.1% for matched controls (P = .03) and LDL improved by 37.7 versus 21.3 mg/dl for matched controls (P = .002). CONCLUSIONS: In this sample with a preponderance of Medicare and Medicaid patients, social workers most often assisted patients with diabetes in obtaining medications or health insurance. For patients with uncontrolled diabetes or cholesterol, a temporal association between SW consultation and improved disease control was noted.

Gender Differences in Physicians' Financial Ties to Industry: A Study of National Disclosure Data.

BACKGROUND: Academic literature extensively documents gender disparities in the medical profession with regard to salary, promotion, and government funded research. However, gender differences in the value of financial ties to industry have not been adequately studied despite industry's increasing contribution to income and research funding to physicians in the U.S. METHODS & FINDINGS: We analyzed publicly reported financial relationships among 747,603 physicians and 432 pharmaceutical, device and biomaterials companies. Demographic and payment information were analyzed using hierarchical regression models to determine if statistically significant gender differences exist in physician-industry interactions regarding financial ties, controlling for key covariates. In 2011, 432 biomedical companies made an excess of $17,991,000 in payments to 220,908 physicians. Of these physicians, 75.1% were male. Female physicians, on average, received fewer total dollars (-$3,598.63, p<0.001) per person than men. Additionally, female physicians received significantly lower amounts for meals (-$41.80, p<0.001), education (-$1,893.14, p<0.001), speaker fees (-$2,898.44, p<0.001), and sponsored research (-$15,049.62, p=0.05). For total dollars, an interaction between gender and institutional reputation was statistically significant, implying that the differences between women and men differed based on industry's preference for an institution, with larger differences at higher reputation institutions. CONCLUSIONS: Female physicians receive significantly lower compensation for similarly described activities than their male counterparts after controlling for key covariates. As regulations lead to increased transparency regarding these relationships, efforts to standardize compensation should be considered to promote equitable opportunities for all physicians.

Developing and testing a checklist to enhance quality in clinical ethics consultation.

Checklists have been used to improve quality in many industries, including healthcare. The use of checklists, however, has not been extensively evaluated in clinical ethics consultation. This article seeks to fill this gap by exploring the efficacy of using a checklist in ethics consultation, as tested by an empirical investigation of the use of the checklist at a large academic medical system (Cleveland Clinic). The specific aims of this project are as follows: (1) to improve the quality of ethics consultations by providing reminders to ethics consultants about process steps that are important for most patient-centered ethics consultations, (2) to create consistency in the ethics consultation process across the medical system, and (3) to establish an effective educational tool for trainers and trainees in clinical ethics consultation. The checklist was developed after a thorough literature review and an iterative process of revising and testing by a group of experienced ethics consultants. To pilot test the checklist, it was distributed to 46 ethics consultants. After a six-month pilot period in which ethics professionals used the checklist during their clinical activities, a survey was distributed to all of those who used the checklist. The 10-item survey examined consultants' perceptions regarding the three aims listed above. Of the 25 survey respondents, 11 self-reported as experts in ethics consultation, nine perceived themselves to have mid-level expertise, and five self-reported as novices. The majority (68 percent) of all respondents, regardless of expertise, believed that the checklist could be a "helpful" or "very helpful" tool in the consultation process generally. Novices were more likely than experts to believe that the checklist would be useful in conducting consultations. The limitations of this study include: reduced generalizability given that this project was conducted at one medical system, utilized a small sample size, and used self-reported quality outcome measures. Despite these limitations, to the authors' knowledge this is the first investigation of the use of a checklist systematically to improve quality in ethics consultation. Importantly, our findings shed light on ways this checklist can be used to improve ethics consultation, including its use as an educational tool. The authors hope to test the checklist with consultants in other healthcare systems to explore its usefulness in different healthcare environments.