Barriers to Conservative Management Conversations: Perceptions of Nephrologists and Fellows-in-Training

Abstract

Background: Not all treatments are appropriate for all individuals with kidney failure (KF). Studies suggest that conversations surrounding end-of-life decisions occur too late or not at all.

Objectives: The aim of this research was to identify perceived barriers to such discussions among nephrologists and nephrology fellows to determine if barriers differ by experience level.

Design: Phase I consisted of semistructured telephone interviews with nephrologists and fellows. Phase II included focus groups with nominal group technique in which providers ranked barriers to discussions about not initiating/withholding dialysis (NIWD) or discontinuing dialysis (DD).

Setting/Subjects: U.S. community-based nephrologists and nephrology fellows.

Results: Seven interviews were conducted with each group (n = 14) in phase I. Many barriers cited were similar among providers, however, differences were related to fellows' position as trainees citing the “reaction of their attending/supervising physician or other providers” as a barrier to NIWD and “lacking their attending physician's support” as a barrier to DD. Six focus groups were conducted, nephrologists (n = 22) and fellows (n = 18), in phase II. The highest ranked barrier to NIWD for nephrologists was “discordant opinions among patient and family”; fellows ranked “time to hold conversation” highest. Nephrologists' highest barrier to DD was the “finality of the decision (death)”; fellows ranked the “inertia of the clinical encounter” highest.

Conclusions: Capturing the perspectives of nephrologists and fellows concerning the barriers to conservative management of patients with KF may inform the development of targeted education/training interventions by experience level focused on communication skills, conflict resolution, and negotiation.

Introduction

Since the 1972 enactment of the Social Security amendments established Medicare as the primary payer for dialysis services in the United States, initiation of dialysis among individuals with kidney failure (KF) remains common place.^1,2 The prevalence of dialysis use continues to be impacted by: (1) the volume (∼37 million) of Americans diagnosed with chronic kidney disease (CKD); (2) high prevalence of primary contributors to CKD—diabetes and hypertension; and (3) recent correlations identified with acute kidney injury caused by the coronavirus (SARS-CoV-2) that may cause lasting kidney damage.^3–6 Although dialysis has been described as “a remarkable, life-prolonging therapy,” clinical practice guidelines suggest that not all patients with KF should receive dialysis.^7,8 The demands of treatment can outweigh the benefits and confer no survival benefit, particularly among subgroups of geriatric patients.^9,10 Dialysis may result in high symptom burden, including nausea, pruritus, pain, and reduced quality of life (QOL), and contribute to increased hospitalizations.⁸ While many patients find the life-sustaining benefit dialysis offers to be worth the symptom burden, overuse of the treatment can be discerned when the care delivered is out of alignment with patient goals. Medical management without dialysis (i.e., conservative management [CM]) is an option for all patients as it aligns with their goals of care and preferences for treatment.

Prior research has aimed to assess nephrologist perceptions, practice approaches, barriers to and rationale for decision making in the use of CM among patients with KF. Four distinct approaches to decision making have been identified, which include the following: paternalist, informative (patient led), interpretive (navigator), and institutionalist.¹¹ Differences between these approaches include how a nephrologist considers his or her role, patient autonomy, engagement/deliberation, influence of institutional norms, and importance of clinical outcomes.¹¹ Paternalists and institutionalists tend to view dialysis initiation as a measure of success, whereas interpretive and informative nephrologists focus on patient engagement and aligning patient values with treatment.¹¹ Decision making should incorporate person-centered practices in which treatment decisions are based on individual patients' priorities while remaining flexible to changing needs.¹² However, international research has suggested an absence of formal training of nephrologists in how to communicate with patients and experience in CM contributes to poor advance care planning (ACP).¹³ Therefore, the purpose of this study was to identify perceived barriers to discussions about not initiating/withholding dialysis (NIWD) or discontinuing dialysis (DD) among nephrologists and nephrology fellows (providers) to determine if barriers differ by experience level for potential educational interventions.

Methods

Study design

The study consisted of two phases conducted over two years. Phase I incorporated a qualitative approach, using semistructured interviews with providers. Phase II consisted of mixed-methods with quantitative data embedded in a qualitative design, using focus groups with nominal group technique (NGT).¹⁴ Phase I was approved by the West Virginia University Institution Review Board (IRB) (1508784632A001) and phase II was approved by Duke University IRB (Pro00075729).

Participant selection

Physicians were eligible for recruitment if they specialized in adult nephrology and were fluent in English. Nephrologists must have been practicing in a clinical setting (50% of time) and fellows must have finished their first year of fellowship.

Recruitment

Participants for phase I of the study were initially recruited using colleague nominations. Kidney community stakeholders including executive directors of Medicare's 18 ESRD Network Organizations, national nonprofit organizations (i.e., American Kidney Fund, National Kidney Foundation, and American Association of Kidney Patients), members of medical review boards, and renowned researchers were e-mailed requesting provider nominations. The snowball sampling technique was used to recruit additional providers from study participants. Nephrologists for phase II focus groups were recruited using the registration list for the 2017 National Kidney Foundation's Spring Clinical Meeting in Tampa, Florida. Fellows were recruited from a convenience sample in one southern state local to an academic setting in the spring of 2018. Additional fellows were recruited through virtual recruitment in the fall of 2018 to expand the pool of participants and eliminate geographic restrictions. Consent was obtained from all participants before initiating phone interviews or focus groups.

Data collection

Cognitive interviews with two nephrologists and two fellows revealed providers approach conversations about NIWD and DD differently. Therefore, NIWD and DD discussions were discussed as separate questions. Participants were asked to focus on their own experiences caring for patients 65 years or older requiring chronic dialysis. In phase I, semistructured telephone interviews were conducted with the selected providers. Phase II employed NGT to rank barriers based on highest influence on both types of conversations, as identified in phase I. Qualitative data were collected from NGT rankings and qualitative data from transcribed narrative text.

Procedure

Phase I

Interviews were conducted by the principal investigator (PI) lasting ∼60 minutes. See Table 1 for sample interview questions. Participants received a $50 gift card upon interview completion.

Table 1.

Phase I Semistructured Interview Sample Questions

• In your experience, what makes it most difficult for you to hold a conversation about discontinuing dialysis with a patient, 65 years of age or older?

• Now I would like you to think of the last patient with whom you held a conversation about not initiating or withholding dialysis or discontinuing dialysis in the outpatient setting.

– Tell me a little bit about this patient. And what was the topic of the conversation—not initiating or discontinuing?

– Who brought up the topic?

– Overall, how would you say the conversation went?

– Was there a particular part, or parts, of that discussion that was especially difficult for you? Can you describe that part and why you found it to be difficult?

– How did you feel during this discussion with this patient?

Phase II

Six focus groups were facilitated by a group facilitator or the PI, both trained facilitators lasting between 120 and 180 minutes. See Table 2 for sample discussion and ranking questions. Lunch was provided during in-person focus groups and participants received a $150 gift card upon completion.

Table 2.

Phase II Focus Group with Nominal Group Technique Sample Questions

• General discussion sample questions

– How have you explained not initiating or withholding dialysis to a patient and/or family?

PROBE: How have you explained the symptoms they might experience?

– Are there patients you are more likely to discuss discontinuing dialysis with than others and how do you decide that?

PROBES: Do you use the “surprise question?”—“Would I be surprised if this patient died in the next year?” which identifies patients who are at high risk for early mortality.

• Nominal group technique (ranking) questions

– In your experience, what makes it most difficult for [nephrologist/nephrology fellows] to hold a conversation about not initiating or withholding dialysis with a patient, 65 years of age and old, in the outpatient setting?

– In your experience, what makes it most difficult for [nephrologist/nephrology fellows] to hold a conversation about discontinuing dialysis with a patient, 65 years of age and older, in the outpatient setting?

All participants were asked to complete a demographic survey. Telephone interviews and focus groups were recorded with permission and transcribed for analysis.

Analysis

Data were analyzed, as collected, using the constant comparative method to determine if new information was emerging. Once data saturation was reached, no additional interviews or focus groups were held. The research team (i.e., PI, a nephrology fellow, and a nephrology nurse) reviewed and coded phase I interview transcripts using Excel spreadsheets for tracking, comparison, and discussion. The team met via conference call to discuss discrepancies until consensus was reached. Related codes were grouped into themes and organized by domain (i.e., personal, interpersonal, and system). A list of barriers for NIWD and DD was generated from the interviews and used for ranking during phase II focus groups. Phase II focus group rankings were combined and compared.

Results

A total of 54 physicians participated in this study. Provider characteristics are listed in Table 3.

Table 3.

Characteristics of Study Participants

	Phase I		Phase II
	Nephrologists (n = 7), n (%)	Fellows (n = 7), n (%)	Nephrologists (n = 22), n (%)	Fellows (n = 18), n (%)
Age	47 (SD = 11)	33 (SD = 4)	49 (SD = 11)	32 (SD = 2)
Gender
Female	3 (43)	3 (43)	11 (50)	11 (61)
Male	4 (57)	4 (57)	11 (50)	7 (39)
Race
African American	0 (0)	1 (14)	1 (5)	1 (6)
Asian	1 (14)	5 (71)	11 (50)	8 (44)
Caucasian	6 (86)	0 (0)	8 (36)	8 (44)
Middle Eastern	0 (0)	1 (14)	0 (0)	1 (6)
Mixed race	0 (0)	0 (0)	2 (9)	0 (0)
Ethnicity
Hispanic or Latino(a)	0 (0)	0 (0)	2 (9)	0 (0)
Non-Hispanic	7 (100)	7 (100)	20 (91)	18 (100)
Medical school
United States	6 (86)	4 (57)	11 (50)	10 (56)
Non-United States	1 (14)	3 (43)	10 (45)	6 (33)
Missing	0 (0)	0 (0)	1 (5)	2 (11)
Years practicing	16 (SD = 11)	n/a	16 (SD = 11)	n/a
Communication skills training
Yes	0 (0)	1 (14)	5 (23)	2 (11)
No	7 (100)	6 (86)	17 (77)	16 (89)
Living will
Yes	3 (43)	0 (0)	7 (32)	2 (11)
No	4 (57)	7 (100)	15 (68)	16 (89)
Medical power of attorney
Yes	5 (71)	1 (14)	8 (36)	2 (11)
No	2 (29)	6 (86)	13 (59)	16 (89)
Missing	0 (0)	0 (0)	1 (5)	0 (0)

SD, standard deviation.

Phase I (n = 14)

Seven nephrologists and seven fellows participated in semistructured telephone interviews. Nephrologists were predominately male (57%), Caucasian (86% vs. 14% Asian), with a mean age of 47 (SD = 11), and practicing an average of 16 years (SD = 11). A majority attended a U.S. medical school (86% vs. 14% outside of the United States) and none had participated in communication skills training. Five (71%) had a medical power of attorney while three (43%) had a living will. A majority of fellows were male (57%), Asian (71% vs. 14% African American, 14% Middle Eastern), with a mean age of 33 (SD = 4). A majority attended a U.S. medical school (57% vs. 43% outside of the United States) and one fellow (14%) had participated in communication skills training. The majority of fellows (86%) did not have a medical power of attorney and none had a living will.

Thematic analysis

Barriers spanned multiple domains (i.e., personal, interpersonal, and system) with consistencies observed between NIWD and DD among providers, regardless of the experience level. Personal barriers related to NIWD included a “fear of how the patient/family will respond” and the inability to “predict how it will go for the patient.” A participant commented:

“I think even just saying, ‘well, you may become nauseated or you may get increasing swelling or things like that,’ is a very different than saying, ‘and eventually you will die from your kidney disease.’ Just saying those words is a lot harder.”—Nephrologist

Personal barriers related to DD were similar in that both providers cited the inability to “predict how it will go for the patient.” Both suggested talking about DD is an “uncomfortable discussion” and can be “a difficult conversation” to have with patients/families (Table 4).

Table 4.

Personal Barriers to Holding Palliative Care Conversations

NIWD		DD
Nephrologists	Fellows	Nephrologists	Fellows
• Fear of how patient or family will respond • Unable to predict how it will go for patient; patient experience • Don't want to be seen as giving up on patient • Fear of letting patient down • Difficult topic to bring up • Don't want to be the bearer of bad news • Lack of a script—unsure of what to say	• Fear of how patient or family will respond • Unable to predict how it will go for patient; patient experience	• Unable to predict how it will go for patient; patient experience • Uncomfortable discussion • Difficult conversation • Emotionally draining • Finality of decision (death) • Don't want to be seen as giving up on patient	• Unable to predict how it will go for patient; patient experience • Uncomfortable discussion • Difficult conversation • Finding the “right” words because death is eminent and in a short time • Fear of how the conversation will go

DD, discontinuing dialysis; NIWD, not initiating/withholding dialysis.

Interpersonal barriers focused on characteristics of and relationships with patients/families. “Emotional patients” were seen as barriers to both types of discussions. The “length of provider/patient relationship,” either short term (nephrologists and fellows) or long term (nephrologists), was cited as a barrier for both types of discussions. “Denial” and “discordant opinions between patient/family” were barriers to NIWD conversations by both providers (Table 5). A participant commented:

Table 5.

Interpersonal Barriers to Holding Palliative Care Conversations

NIWD		DD
Nephrologists	Fellows	Nephrologists	Fellows
Patient • Emotional • In denial • Wants “everything” • Lack of patient understanding − Asymptomatic; doesn't know why they are seeing a nephrologist − Thinks palliative care is no care • Impaired − Cognitive impairment Unable to speak for him/herself	Patient • Emotional • In denial • Cultural beliefs • Language barriers • “Younger” patient (e.g., 65 years old vs. 85 years old) • Lacks social support • Doesn't realize how sick s/he is Family • Feelings of failure if treatment isn't attempted	Patient • Emotional • Doesn't bring it up • Doesn't know what “palliative care” means • Fears dying • In denial • Lacks social support • Impairment − Cognitive impairment − Unable to speak for him/herself	Patient • Emotional • Language barriers • Lack of trust • Not expecting conversation; thinks things are fine Family • Wants to continue no matter what
Relationships  • Discordant opinions between patient and family • Length of relationship with patient—either long or short • Strong relationship with patient • Lack rapport with patient	Relationships • Discordant opinions between patient and family • Lack of relationship with patient/short relationship with patient • Family members disagree with each other • Discordant opinions between patient and nephrologist	Relationships • Length of relationship with patient—either long or short • Discordant opinions between patient and family • Patient/family could misinterpret nephrologist's intent	Relationships • Lack of relationship with patient • Discordant opinions between patient and nephrologist • Lack of contact with family/care partners

“It's [the patient's] decision. They cannot expect me to force a treatment on their family member. That's what I find sometimes challenging.”—Nephrologist

Lack of time was cited as a system barrier to both types of discussions by both providers. In addition, both suggested reimbursement as a barrier to NIWD conversations. The current reimbursement system pays physicians more for patients on dialysis rather than receiving CM, regardless that CM is more time-intensive for physicians (Table 6). A participant commented:

Table 6.

System Barriers to Holding Palliative Care Conversations

NIWD		DD
Nephrologists	Fellows	Nephrologists	Fellows
• Time to hold conversation • Reimbursement • Medical mistrust • Emergent starts—no time to discuss or educate	• Time to hold conversation • Reimbursement • Reaction of attending or other providers • Uncertainty of what other providers have told patient • Expectations set by other providers	• Time to hold conversation • Lack of privacy in dialysis facility Medical mistrust	• Time to hold conversation • Lack of privacy in dialysis facility • Lack attending's support • Uncertainty of what other providers have told patient • Inertia of clinical encounter; easier to continue doing what you have been doing

“Right now, the way it's reimbursed a nephrologist is actually acting against their best financial interest to do what is best for the patient's interest. It's never good when you create that kind of misalignment of incentive in the system.”—Fellow

A “lack of privacy within dialysis facilities” was suggested as system barrier to DD conversations by both providers. Nephrologists suggested “medical mistrust” as a barrier to both types of discussions. Fellows identified patient communication with other medical specialties as a barrier to both types of discussions, specifically “expectations set by other providers” as a barrier to discussions about NIWD, while “uncertainty of what other providers have told the patient” as a barrier to DD. While similarities exist between providers, differences seemed to relate to fellows' unique position as trainees. Fellows expressed the “reaction of their supervising physician/other providers” as a barrier to discussions about NIWD, citing “lacking their attending physician's support” when asked about barriers to DD discussions.

Phase II (n = 40)

A total of 22 nephrologists and 18 fellows participated in 1 of 6 focus groups. Groups included six to eight physicians. Nephrologists were 50% female, predominately Asian (50% vs. 36% Caucasian, 9% mixed race, 5% African American), with a mean age of 49 (SD = 11), and practicing for an average of 16 years (SD = 11). Five (23%) had participated in a communication skills training. A majority of fellows were female (61%), Caucasian or Asian (44% vs. 6% African American, 6% Middle Eastern), with a mean age of 32 (SD = 2), and 2 (11%) had participated in communication skills training.

NGT rankings

Barriers among providers were similar, while rankings differed. Nephrologists' top ranked barriers were primarily interpersonal, focusing on patients/families and the complex relationships that exist among them (Table 7). Nephrologists ranked “discordant opinions among patient and family” as the greatest barrier to NIWD. Many described scenarios of patients refusing dialysis but pressured by family to initiate dialysis. Nephrologists described feeling “in the middle,” between patients/family members, and feeling pressured by family members to convince patients “not to give up.” However, nephrologists described their duty to respect patient wishes, assisting family members to understand the patient's perspective. Nephrologists ranked the “finality of decision” as the most influential barrier to DD describing how quickly death occurs, often within 7 to 10 days of dialysis withdrawal. Specifically, the finality of the decision, there are no “do overs,” and challenges associated with shifting goals of care from preserving life to comfort care.

Table 7.

Rankings of Barriers to Holding Palliative Care Conversations

Rank	NIWD		DD
	Nephrologists	Fellows	Nephrologists	Fellows
1	Discordant opinions among patient and family (I)	Time to hold conversation (S)	Finality of decision (death) (P)	Inertia of clinical encounter; easier to just keep doing what we have been doing (S)
2	Patient in denial (I)	Patient doesn't understand need for treatment (asymptomatic) (I)	Cognitively impaired patient (e.g., dementia, memory issues, delirium) (I)	Lack of relationship with patient (I)
3	Patient doesn't understand need for treatment (asymptomatic) (I)	Discordant opinions between nephrologist and patient (I)	Discordant opinions among patient and family (I)	No privacy in dialysis facility setting (S)

Domain: I, interpersonal; P, personal; S, system.

Fellows' barriers were divided between the interpersonal and system domains, ranking “time to hold conversations” as the greatest barrier to NIWD. Fellows described NIWD conversations as time-intensive, requiring multiple interactions, and found them difficult to hold given the competing demands fellows experience as trainees. Fellows ranked “inertia of clinical encounter; it is easier to just keep doing what we have been doing” as the most influential barrier to DD. Fellows described scenarios of stable patients in overall poor condition where status quo prevails, and little discussion is had to discontinue dialysis.

Discussion

Both providers identified personal barriers to NIWD due to the inability to predict a patient's response clinically and emotionally. Dialysis initiation decision making in older adults can create an emotional burden for nephrologists. Research identified several contributing factors for this burden, including uncertainty about how patients will do on dialysis and realization that the alternative to dialysis is death in cases where CM will not substantially prolong life.¹⁵ The implications of barriers for NIWD and DD can contribute to moral distress among nephrology providers occurring when providers are aware of a problem, perceive a responsibility to act, but cannot do so in a way consistent with their moral judgment.¹⁰ Providers who routinely offer CM relieve their moral distress, whereas those who do not continue to experience significant distress.¹⁰ Among fellows, perceived barriers related to NIWD perhaps stem from limited experience in CM. A 2013 national survey of nephrology fellows found less than half (46.9%) of study participants had formal didactic palliative care experience.¹⁶ The majority of fellows (90%) felt comfortable prescribing dialysis, only 35% felt comfortable with NIWD among patients in the ICU with multiorgan failure.¹⁶ Nearly one in five fellows surveyed felt obligated to offer dialysis to every patient regardless of the benefit.¹⁶

Both providers cited discordant opinions between patients/families as barriers for NIWD and/or DD. Literature suggests early initiation of these conversations to understand and mediate patient, family, clinician, and organizational phenomena contributing to disparities in the acceptance and use of palliative care resources.¹⁷ Furthermore, ACP discussions develop trust in providers and clarify values/beliefs about their condition.¹⁸ This leads to greater congruence between patients/families and increased quality of communication between patients and providers.¹⁸ Fellows identified the need for longitudinal conversations with patients as patients may misunderstand CM initially. Literature suggests that although most nephrologists recognize the importance of ongoing conversations with patients about goals, they acknowledge the burden of these conversations and exhibit restraint in assessing treatment preferences unless patients express ambivalence.¹² The development of palliative skills, including holding goals-of-care conversations and initiating ACP among nephrology providers, is needed to guide shared decision making. This includes incorporating the use of prognostic tools for identifying patients who may benefit from CM, disclosing prognostic information to patients who may not do well with dialysis therapy, incorporating patient goals/values in outlining the treatment plan, and preparing patients/families for end-of-life (EOL) transitions.¹⁹

Both providers cited reimbursement as a barrier to NIWD. Patients with advanced CKD, nearing KF, are complex and need a multidisciplinary care strategy. However, physicians may not feel incentivized to offer CM when appropriate as continuing dialysis treatment yields higher reimbursement.^20,21 Kurella et al. cite the need to fully integrate palliative care into KF management by improving EOL care training, eliminating structural and financial barriers to hospice use, and identifying optimal methods to palliative care delivery.²² These changes could lower health care costs at the EOL, commonly associated with intensive care patterns, including multiple hospital admissions, surgical procedures, and resuscitation attempts.²³

Fellows cited discordant perspectives among medical specialties as a barrier to NIWD. Medical ethics require physicians not only to refrain from carrying out unwanted interventions but also to promote a patient's ability to make informed decisions.²⁴ Many nephrologists have reported expectations to dialyze patients whom they believe may receive little benefit from dialysis treatment.²⁴ Advocating for CM when clinically relevant could be supported by advanced training and improved communication among nephrologists, primary care, and specialists. Prior research found a communication workshop for nephrology fellows (NephroTalk), focused on delivering bad news, defining goals of care, and EOL preferences, an effective way to improve preparedness for difficult conversations.²⁵ System barriers to NIWD align with an institutionalist perspective in which barriers to CM are attributed to system-level influences such as financial incentives, time constraints, and lack of a common approach within the care team, congruent with barriers acknowledged in the literature.¹⁰

Nephrologists' perspectives of DD were riddled with barriers associated with the finality of the decision. Providing patients the opportunity to express their EOL wishes early on is a mitigation strategy. A 2016 study had nephrologists identify patients who may benefit from ACP by asking, “Would I be surprised if this patient died in the next year?”²⁶ Of 50 patients identified, 82% were full codes and 18% were do-not-resuscitate order (DNR).²⁶ Following ACP, 42 percent of those patients expressed preference for DNR status.²⁶ The “surprise question” provides a practical way for providers to prioritize patients for DD discussions. Furthermore, fellows identified maintenance of the dialysis care plan, once initiated, to be easier than DD. Chen et al., found most withdrawal decisions were made by patients (60%) or a family member (33%) often due to an acute medical complication (51%) or failure to thrive/frailty (22%).²⁷ Only a third (34%) of patients opting to discontinue dialysis received palliative care services suggesting that initiating a palliative consult can change the inertia of a patient's care plan and improve QOL.

Limitations

Extrapolation from these findings should be done cautiously due to the small sample size. Participant perspectives may not be generalizable to all nephrology providers. Inherent biases may exist as it is possible that providers who participated in this study may tend to communicate about palliative care more than others and therefore have a higher awareness of barriers; conversely, some providers may believe these discussions are unnecessary and chose not to participate in this study.

Conclusions

These findings establish a framework for future studies with larger, more geographically distributed samples. Capturing the perspectives of additional providers concerning the barriers identified as well as prioritization can inform the development of targeted education/training interventions by experience level to address barriers within the personal and interpersonal domains. Barriers related to communication, knowing what to say or having a script, may be addressed through communication skills training. However, other barriers, mediating discordant opinions (i.e., patients, family members, or medical colleagues), may be best addressed through conflict resolution/negotiation training. System barriers, including reimbursement issues, require collaboration among multiple stakeholders within the kidney community, including patients, families, providers, and dialysis organizations. There has never been a more opportune time than now to incorporate resources, such as NephroTalk, to address the way we communicate with patients/families and offer treatment options to patients with KF. The Advancing American Kidney Health executive order (EO) established in 2019 seeks to reform the way we treat kidney disease in the United States. However, CM as a viable treatment option was not incentivized in the EO. Efforts to continue to illuminate the equality of CM as a treatment option for select patients in alignment with their desired QOL could improve the quality of EOL care for patients with KF.

Authors' Contributions

J.S.R. designed and implemented the study. J.S.R., A.O., and A.P. contributed to the interpretation of the results. A.P. led writing the article with support from J.S.R. and A.O. All authors provided critical feedback and helped shape the research, analysis, and article.

Funding Information

This project was supported by grant number R03HS024317 from the Agency for Healthcare Research and Quality (AHRQ). The content is solely the responsibility of the authors and does not necessarily represent the official views of the AHRQ.

Author Disclosure Statement

No competing financial interests exist.