Patient and Physician Perspectives on Use of Patient-Reported Outcome Measures in Non-Dialysis-Dependent CKD Care: A Qualitative Study

Abstract

Rationale & Objective

People with non-dialysis-dependent chronic kidney disease (NDD-CKD) can experience substantial symptoms related to physical and mental health. Patient-reported outcome measures (PROMs) assess these aspects of health-related quality of life (QOL). Using the Kidney Disease Symptom Survey as an example to guide discussions, we sought to understand perspectives of patients and nephrology physicians about using PROMs in routine NDD-CKD care.

Study Design

Individual semi-structured interviews informed by the Theoretical Domains Framework and associated Capability, Opportunity, Motivation model of behavior change.

Setting & Participants

Adults with NDD-CKD and nephrology physicians in the Greater Baltimore-Maryland region.

Analytical Approach

Hybrid inductive-deductive thematic analysis.

Results

We analyzed interview transcripts from 15 nephrology physicians and 21 people with NDD-CKD. Capability to use PROMs was hindered by limited knowledge and skills of nephrology care teams to act on PROM results and manage symptoms. Opportunity to use PROMs was influenced by concerns surrounding environmental resources, including limited accessibility of PROMs to select patient users and limited time available to physicians to review PROMs data. Participants believed that PROMs could guide person-centered care, but motivation to use them was impeded by the uncertain role of nephrology physicians in assessing and managing symptoms and QOL, as well as concerns that PROMs may capture symptoms unrelated to CKD while also lacking sensitivity for key issues relevant to CKD patient care.

Limitations

Impact of clinical practice setting (including the availability of multidisciplinary support) and the views of individual PROM domains was not assessed.

Conclusions

Patient and clinician participants recognized the importance of symptom and QOL assessment in nephrology care. However, behavior changes required to integrate PROMs into routine CKD care may be limited by barriers related to the capability, opportunity, and motivation of users. These findings can inform the development of programs which support PROM implementation.

Plain-Language Summary

People with kidney disease often have symptoms, which worsen quality of life. Patient-reported outcome measures (PROMs) are surveys that patients can complete to share how they are feeling with health care teams. We spoke with people with kidney disease and kidney doctors to explore perspectives on using PROMs in kidney care. Participants agreed that PROMs could help focus care on patient needs. Some were uncertain if kidney doctors should be responsible for managing symptoms, especially because some symptoms might not be caused by kidney disease. Some doctors felt they did not have enough time and training to review and act on PROM results. These findings can help guide the creation of programs that enable the use of PROMs in kidney care.

Patient-reported outcome measures (PROMs) capture patients’ experience of disease.¹ PROMs have been shown in numerous medical fields to enable person-centered care, such that patients not only have an opportunity to report their symptoms and quality of life (QOL) to clinical care teams, but also so that care teams incorporate patients’ needs into care planning.2, 3, 4 People with chronic kidney disease (CKD) are at risk of high symptom burden and impairments to QOL,^5,6 and international guideline bodies have highlighted the need to incorporate symptom and QOL assessment into routine CKD care.⁷ However, at this time, PROMs are not a routine component of care for people with non-dialysis-dependent CKD (NDD-CKD).

To effectively implement PROMs, perspectives of users must be elucidated and incorporated into implementation planning. Relevant users not only include patients with NDD-CKD who would complete PROMs, but also clinical care teams who would review and act on PROMs data. Furthermore, because the needs of patients and physicians can vary significantly based on their access to resources (eg, socioeconomic status of patients, clinical resources, and care structures for physicians), implementation should be tailored to patient populations and care settings.

In this study, we sought to identify perspectives of patients with NDD-CKD and nephrology physicians in the Greater Baltimore-Maryland region of the United States surrounding use of PROMs. We hypothesized that behavioral determinants would influence users’ perceptions of PROM implementation. We therefore designed our study around the determinant Theoretical Domains Framework (TDF)⁸ and the associated Capability, Opportunity, Motivation (COM-B) model of behavior change,^9,10 which outlines concepts pertaining to capability (the psychological or physical ability to enact a behavior), opportunity (the physical and social environment which enables the behavior), and motivation (reflective and adaptive mechanisms that activate or inhibit the behavior) to use PROMs. Our goal was to use data to develop a framework to guide PROM implementation in routine NDD-CKD care.

Methods

We conducted a qualitative study in which thematic analysis was used to identify determinants of PROM implementation in routine NDD-CKD care. Research was conducted in accordance with consolidated criteria for reporting qualitative research (COREQ) guidelines.¹¹

Setting and Participants

This study was conducted at Johns Hopkins (JH) University, located in the Greater Baltimore-Maryland region. Clinical practices located in this region care for patients who are diverse in race, ethnicity, and socioeconomic status. Patients have variable insurance coverage and access to multidisciplinary/interdisciplinary providers within or external to JH. Multidisciplinary/interdisciplinary care is not a routine component of nephrology care including for patients with advanced kidney disease.

JH Medicine outpatient adult nephrology clinics (located in 4 sites in the Greater Baltimore-Maryland region) distributed an electronic PROM, termed the Kidney Disease Symptom Survey (KDSS), to adult patients with NDD-CKD (stages G3-5) who presented for a follow-up visit from 2022-2024, as previously described.¹² Briefly, patients received a request to complete the PROM 7 days before their clinic visit via the electronic patient health portal, with a reminder sent one day before the visit if not completed. Electronic tablets were available for patients to complete the PROM in-person on clinic check in. PROM results were immediately uploaded into the electronic health record (EHR) and were available for physicians to review during clinic visits. Additional reminders or assistance for patients to complete the PROM, or for physicians to review and discuss results, were not provided. Quantitative data surrounding utilization of the KDSS PROM are reported separately.¹³

We interviewed JH patients with NDD-CKD who had received a request to complete the KDSS (including patients who had not completed it), and nephrology physicians with at least 2 outpatient NDD-CKD clinics per month in the Greater Baltimore-Maryland region (internal and external to JH). Although all JH physicians had received information about KDSS dissemination, utilization of the KDSS was not a requirement for interview participation. Physicians external to JH were not involved in the KDSS intervention.

Sampling Strategy

Nephrology physicians were recruited via an opt-in email distributed by the study team. An effort was made to balance characteristics of physician participants, with purposive sampling used to recruit (1) physicians from JH and non-JH academic centers, and independent practice, and (2) physicians who self-identified as men and women.

Patient participants were selected from a group of JH NDD-CKD patients who participated in a separate questionnaire study,¹³ which asked for perspectives on the KDSS PROM described above. Purposive sampling was used to recruit participants diverse in age, sex, race, and CKD stage. Individuals who were noted to have significant cognitive limitations or a non-English primary language in their electronic medical record were excluded.

We anticipated some variability in perspectives of physicians working in different practice settings, and of patients with different sociodemographic backgrounds and different degrees of kidney function impairment. We chose initial sample sizes of n = 15 physicians and n = 20 people with NDD-CKD to capture perspectives across these groups,¹⁴ with plans to extend recruitment if thematic saturation had not been reached and/or if subsets of participant groups had been underrepresented.

Measures

We identified topics of interest informed by relevant domains of the determinant TDF⁸ and the associated COM-B model of behavior change^9,10 (Table S1). Open-ended semi-structured interview questions were designed to capture perspectives on these topics of interest, as well as others that arose naturally (Item S1-S2). Interviews began by asking participants their perspectives on assessment of patient-reported outcomes broadly and then asked more specific questions on use of PROMs using the KDSS as a reference. Interview questions were pilot-tested within the study team to ensure clarity, and select follow-up probing questions were planned for use as indicated. After the study PI (D.M.P.) reviewed an initial set of 3-5 transcripts for each participant group, the interview guides were revised as needed to ensure clarity of questions and alignment with research objectives.

Procedures

Participants were contacted by a member of the research team (D.S. or J.G.) to obtain consent and schedule an interview. Each participant was emailed or mailed (per participant preference) an informational packet, which included an overview of the KDSS PROM to refer to during the interview.

Interviews were conducted by a female research program coordinator (J.G.) with experience in conducting qualitative interviews. The interviewer discussed the research subject of interest with D.M.P. before conducting interviews, but otherwise had no predetermined biases or assumptions surrounding use of PROMs in CKD care. The interviewer had no established relationship with interview participants. Interviewees were not provided with information about the interviewer. The interviewer ensured that each participant had reviewed and had access to the informational packet, referring to the packet as needed. Field notes were not taken, and repeat interviews were not conducted.

Interviews occurred remotely and were conducted in English. Interviews were audio recorded using the Zoom teleconferencing platform or Virtual Observer software, with recording of audio files only. Audio recordings were submitted to Acolad (Ubiqus) Transcription Services, which provided redacted transcripts for analysis.

Patient participants were remunerated $25 on completion of semi-structured interviews. Physicians were not remunerated. Although transcripts were not returned to participants, a summary of qualitative data was distributed to interview participants for review.

All study activities were approved by the JH Institutional Review Board (IRB00311975).

Analytic approach

We thematically analyzed transcripts and used MaxQDA software to manage data. Three authors (D.M.P., A.P., and S.F.) independently reviewed an initial set of transcripts and collectively refined a codebook. We used a hybrid inductive-deductive strategy to derive codes, which arose naturally as well as those which were guided by interview questions. Codes were organized by relevant TDF and COM-B domains. The final codebook (Table S2) was then applied to all transcripts. Intercoder agreement was calculated for 17 transcripts which were coded by 2 authors, ensuring a kappa statistic > 0.6 (calculated using qualitative software) to represent substantial agreement.¹⁵ The remaining transcripts were coded by one coder.

Coded segments were reviewed in aggregate by authors to derive themes. Thematic saturation was defined as the point at which review of additional coded segments did not identify new concepts or relationships between data.¹⁶

Results

Participant and Interview Characteristics

We interviewed 15 nephrology physicians and 21 NDD-CKD patients (Table 1). The median interview time was 18 minutes (interquartile range 16-24 minutes).

Table 1.

Characteristics of Interview Participants

Physician Interviewees (n = 15)
Type of practice
Academic (JH)	5 (33%)
Academic (non-JH)	5 (33%)
Independent practice	5 (33%)
Number of years in practice
< 1 y	0 (0%)
1 to <5 y	2 (13%)
5 to <10 y	4 (27%)
10 to <20 y	4 (27%)
20+ y	5 (33%)
Frequency of outpatient clinic
2-3 per mo	2 (13%)
1-2 per wk	7 (47%)
3+ per wk	6 (40%)

Patient interviewees (n = 21)
Age, y (mean ± S.D.)	60 ± 17
Male sex (n %)	9 (43%)
African American race (n %)	8 (38%)
CKD stage
G3a	3 (14%)
G3b	7 (33%)
G4	8 (38%)
G5	3 (14%)
Completed KDSS PROM before clinic visit (n %)	12 (57%)

Abbreviations: JH, Johns Hopkins; KDSS, kidney disease symptom survey; PROM, patient-reported outcome measure.

Perspectives on Capability

In general, patient participants did not express concern surrounding their capability to complete the PROM. However, nephrology physicians had concerns surrounding capability to act on PROM results. These concerns stemmed from a self-perceived lack of knowledge and skills required to manage symptoms. Nephrology physician participants noted that they are not traditionally trained in symptom assessment and management. Given these limitations, physician participants cited a need for multidisciplinary/interdisciplinary support to help manage symptoms. The potential benefit of a care team member dedicated to patient wellness was supported by a patient participant (Table 2).

Table 2.

Perspectives on Capability to use PROMs in Routine CKD Care

Theme	Subthemes	Representative Quotations
Nephrology care teams have limited knowledge and skills to act on PROM results and manage symptoms	Nephrology physicians are not trained in assessment and management of symptoms and QOL.	•“I think the barrier for me is how do I make them better with those symptoms. And I may not have all the tools necessary to improve those symptoms. And that’s where I think we, as nephrologists, may fall short because we are not trained to manage those symptoms necessarily or even to make a good diagnosis of why they have those symptoms.” (physician 1, JH academic practice) •“I feel like we’re pretty good at diagnosing. But, as far as managing the symptoms, in terms of if I need to prescribe any antidepressants or anything like that, I don’t think I’m the right person to do that. In that case, it would be definitely helpful if I have a psychiatrist on board who's obviously more in tune with the treatment options than I am.” (physician 10, independent practice)
	Multidisciplinary/-interdisciplinary support is needed to assist both patients and physicians in managing all aspects of PROMs.	•“I do feel that patients with chronic kidney disease, they are complex. They have multiple competing symptoms and demands. And having a multidisciplinary team or people who would be well-equipped in assessing the symptoms and, as you said, not only physical but quality of life and mental health, I think that would be potentially beneficial for patients with CKD to address their symptoms, and help them navigate the system, and have a higher quality of life.” (physician 13, non-JH academic) •“I feel, in terms of quality of life, easier access to allied health professional support, social work support, nutrition, definitely, palliative. So having a team to address multiple issues that sometimes it’s a little bit hard to be handled by the physician alone, I think, helps. Even simple nursing support. Because every contact the patient has with the health care system is usually an additional support for the patient…” (physician 15, JH academic practice) •“…if they had a specialist, where their whole job was to check in on all the kidney care patients. Here’s what it is. Here’s what you answered in your symptom survey. How are those going for you? Let’s find some solutions. You can do either changes in your daily life, in your routine, or changes in over-the-counter medications can be very helpful. Like, do you need Tylenol. Do you need laxatives? Do you need anti-diarrheal medications? How do we connect you with the supplies that you need and to make sure that they have the resources to get those supplies and all that kind of thing, like just somebody whose whole job is to make somebody feel welcomed and comfortable, would be a game-changer. It would be fantastic.” (patient 1, CKD G5)

Note: Participants are identified by participant type (physician or patient), participant ID, and CKD stage for patients.

Abbreviations: CKD, chronic kidney disease; PROMs, patient-reported outcome measures.

Perspectives on Opportunity

Physician and patient participants felt that the opportunity to use PROMs in routine CKD care would be facilitated by their ease of their completion. Using the KDSS as an example to reference, most participants viewed the PROM as being nonburdensome to patients while capturing many relevant symptoms. One physician participant suggested that screening questions could be used first, which could help avoid asking additional questions to patients who were not experiencing symptoms (Table 3).

Table 3.

Perspectives on Opportunity to use PROMs in Routine CKD Care

Theme	Subthemes	Representative Quotations
PROMs are relatively easy to complete.	For most patients, completion of PROMs is not burdensome.	•“I think the survey seems to address relevant symptoms and would be not particularly burdensome for the vast majority of patients.” (physician 5, non-JH academic practice) •“I think you’ve kind of hit the high points on potential illness. You do a general quality of life assessment, which is good. You check for mental health issues a little bit, which is good. I like it. I think it covers what it needs to cover without being too intrusive.” (patient 9, CKD G3b)
	PROMs may need optimization to be more user friendly.	•“I would probably try to make this a little bit more user friendly and also shorter, so if we can—if they can start with maybe three questions, and if one of those is a yes and that’s possibly for depression or for any other symptom, then more questions come up after that. And probably that would help screen patients that would benefit more from a more detailed survey.” (physician 7, JH academic practice) •“Just make certain that it's really easy to complete, that it doesn’t involve long and drawn-out thought on a subject because you’re already there at the doctor’s, with a certain amount of anxiety and concern. So, to have something that's difficult and hard to get you along with what you're already experiencing, that would pose problems…Just a simple, maybe a yes or no, questions that are yes or no.” (patient 12, CKD G4)
PROMs should allow multiple potential methods of data collection.	PROMs could be collected in-person on clinic check in.	•“I think the tablet in the clinic is a good place because you're really focused on your kidneys at that point, where getting to [EHR patient portal] requires a bit of a discipline.” (patient 17, CKD G4)
	Clinic staff could assist patients in completing PROMs.	•“I can imagine subgroups of patients, the very elderly patients with cognitive impairment, patients with visual impairment or other physical impairment for whom completion of an electronic questionnaire would be more challenging. In that case, another option I think would be just to have it administered by support staff as part of say, for example, the initial vital sign measurements.” (physician 5, non-JH academic practice)
Review of PROMs data is limited by time constrains of clinic.	Physicians require additional time to review and discuss PROMs data with patients.	•“…to me, the challenge with these conversations, it doesn’t have anything to do with whether I’m willing to bring it up and talk about even things like hospice care or other ways to manage end-stage type disease because I do that pretty regularly and I do feel comfortable doing that. But sometimes I think that it requires a lot more time than what we build into our day.” (physician 4, independent practice) •“I think that they’re important questions to get from a patient. But my downside concern is they’re going to answer a lot of them positive. And then you’re obliged to address that. And I wouldn’t ignore them because that’s not my personality. But now a 30-minute visit becomes a 45-minute visit. And I’m running late for everyone else.” (physician 8, JH academic practice) •“There’s a real drive to get to the stuff that we need to discuss before we run out of time in clinic: kidney function, hypertension control, anemia, bone disease, and, often, quality of life, mental health, these things often take a back seat to those issues, simply because there's so much to go through…It's almost always, okay, we've got to check this box, or we’ve got to check that box…Often, because of that, these issues take a back seat and are not frequently addressed.” (physician 9, non-JH academic practice)
	PROMs could be collected before clinic visits via the EHR.	•“I think if the survey is given to the patient beforehand, like we said, like they get a reminder text to fill out the survey before their appointment, then we can just review the results on the day of instead of having them fill it out during the appointment time.” (physician 14, independent practice)
	Clinic staff could help ensure that PROMs data are reviewed.	•“I don’t know if it would help if, for example, when they’re checked in, if they answered the questionnaire through the iPad if you’re alerted as well that they, if the [medical assistant] who’s checking in the patient, adds that, in addition to vital signs, it said they answered the [PROM]…” (physician 15, JH academic practice)
	PROM results should be easily understandable.	•“…if I have the three scores, the global summary score, the physical, and the mental, specifically, and these are the scores, and it will appear green, yellow, or red, based on the score. And if it’s red or yellow, I know it’s not great, but if it’s red I know it’s bad, then I can click on that tab, and then get the detail of what’s giving that score. And I can look into the detail of it and focus on it more.” (physician 1, JH academic practice) •“Well, one would be if there was some kind of automatic way to, like let’s say somebody has a high or a low score or something stands out that signals that they’re having some substantial symptoms. If there’s any way that if the score surpasses a certain number, you know, you get like an alert.” (physician 12, independent practice)

Participants are identified by participant type (physician or patient), participant ID, and CKD stage for patients.

Abbreviations: CKD, chronic kidney disease; EHR, electronic health record; PROMs, patient-reported outcome measures.

Physician and patient participants noted a need to provide multiple methods of PROM data collection. Although many patients complete the PROM before clinic visits, one patient participant stated a preference to complete it on a tablet in clinic, in which they were more focused on their kidney disease. Electronic PROMs were viewed as being challenging for patients with visual or physical impairments, in which case alternate means of completion (such as a staff-assisted completion) would be beneficial.

Physician participants also noted several concerns surrounding time constraints of clinic. PROM data would add to an already long list of items to review during nephrology clinic visits. Having patients complete PROMs before clinic visits so that results are available during pre-charting, having clinic staff alert physicians to the availability of PROMs data, and using color-coded graphs or summary tables to quickly highlight critical data points (eg, severe symptoms) could help facilitate PROM utilization.

Perspectives on Motivation

Physician and patient participants expressed motivation to use PROMs as a tool to facilitate person-centered care (Table 4). As voiced by participants, PROMs could provide a mechanism for patients to update their care teams on the development of symptoms and QOL impairments, which would have otherwise gone unrecognized. When PROMs provide quantitative scores rather than binary yes/no responses, they could help identify the most significant issues that patients are experiencing. One patient commented that PROMs could therefore help focus clinic visits on the issues that are most important to the patient.

Table 4.

Perspectives on Motivation to use PROMs in Routine CKD Care

Theme	Subthemes	Representative Quotations
PROMs can enhance person-centered care.	Symptoms and QOL are pertinent to CKD care.	•“…as the level of kidney disease progresses, and they become more and more advanced CKD, then I think those symptoms can become more and more prominent and more important for the patients, not able to sleep at night, waking up several times at night to make urine because of lack of concentration ability or basically decreased appetite, leg cramps, a lot of complaints about leg cramps. And they wake up at night because of that. So all the symptoms start becoming more and more important. And we, as nephrologists, I believe need to pay more attention to them.” (physician 1, JH academic practice)
	PROMs may identify symptoms and impairments which would not have been otherwise recognized.	•“I could see if these are important symptoms that really crop up for some people…let’s say itchy skin, a person may not think that that has anything to do with kidneys, and if suddenly became a big, a real problem, it probably would be good - by doing the survey and telling the physician, it would point to those symptoms that otherwise they would not have thought of as kidney related.” (patient 5, CKD G4) •“People have different personalities, so some people aren’t necessarily as forthcoming. This might help them be a little more open because they're like, well, I’ve already answered the question on paper. It’s not like they feel weirded out or anything if their kidney doctor is asking and confirming information that they’ve already provided already.” (patient 20, CKD G5)
	PROMs can help identify the most significant issues that patients are experiencing.	•“…part of the benefit of having a survey that’s done beforehand—and I certainly see some benefit to that—is that it at least helps frame the discussion. To me, it’s sort of like the questionnaire that still gets handed out in paper to new patients that it’s certainly not the only thing that we go on, but it at least helps frame a discussion. It gives me some sense of how the patient has been feeling the past few weeks or months. And we can use that as a jumping off point.” (physician 2, JH academic practice) •“…I know that they have so much to do during the visits, this is something that might help them focus on things that are important to both of us.” (patient 9, CKD G3b)
	PROMs could inform medication management and care planning.	•“…to me, it would motivate my provider and myself to sort of say, is there anything I can do about this? Or is there anything I should be doing differently? Should I be, I don't know, being more active? Should I be losing weight? Should my medication be changed? I think to me, it just changes care because it makes you not just accept feeling crummy just because you have kidney disease. And I think that's kind of for me, how I would see it. Changing care. It’s just that sometimes you just accept feeling terrible because you think maybe there’s nothing you can do about it. But there probably is something you can do.” (patient 8, CKD G3a)
	PROMs can be used to monitor changes in symptoms and QOL over time.	•“From all of this, if we have five important symptoms, that patient would feel like physicians can say, I can see that we are having five important symptoms. So that will help me to be more targeted in reviewing medications, clarifying the symptoms…So that will be more, first of all, patient centered; second, targeted treatment, and measurable results, because next time, they could say that last time you had, for example, shortness of breath. It was 8 out of 10, and we made the changes in medication. How would you measure? I can see you measure it today 7, so we made some progress. Let’s see if we have room for improvement.” (physician 13, non-JH academic practice)
	PROMs can increase patient knowledge of the impact of CKD.	•“I follow a lot of CKD4 or 5 patients. I maybe could identify one or two patients who particularly said that with the symptom survey they felt that they didn’t realize how bad they were feeling, and because of that, it helped them come to terms with, okay, yeah, I think I do need to start planning dialysis, or we probably should start planning dialysis soon. And those are patients who are highly functional…because they can maintain their function, so they’re not realizing that these other symptoms are actually signs of uremia. So I think for those patients it helped them concretize or have acceptance of how bad they’re feeling, and it led towards better preparation.” (physician 15, JH academic practice) •“I think it would get less educated people more in tune with what's going on in their body and why. I think maybe they should even say why. Like, a lot of itchy skin for me was only when my phosphorous was really high in my blood…People should know what's going on.” (patient 4, CKD G4) •“…there might have been things—there—as I was filling it out. I remember thinking, gosh, I didn't know that this was—could be a kidney thing. I thought it was just me or something, or just, you know, how people live, or—so I-I just I even with fatigue I never would have thought maybe I'm more fatigued than anyone else…And so if I wasn't really asked about it, I could see myself not sort of—sort of discrediting it and sort of saying, oh, that probably has nothing to do with this.” (patient 8, CKD G3a)
	PROMs could enhance patient-physician relationships.	•“It could go either way. Some would say, oh gee, they’re making sure they get everything. Others would say, oh gee, I’m just a checklist.” (physician 3, independent practice) •“I feel like that allows them to feel more seen and for their symptoms to be, or what they’re experiencing to be validated externally and be acknowledged by the provider even if there’s nothing that we particularly do to resolve or address these symptoms. For some patients, just the acknowledgement of how bad they’re feeling makes a difference.” (physician 15, JH academic practice) •“…I've met so many people where they have such a hard time really connecting with their new care team, because it's all just so alien to them. I think this would be a great way to bridge that gap, make them feel heard, make them feel seen, allow them to really connect with their doctor, like, hey, we're here to help you. That's the whole point.” (patient 1, CKD G5)
There is uncertainty surrounding the role of nephrology physicians in assessing symptoms and QOL.	CKD care has historically focused on numbers.	•“I think I have witnessed, or overheard, or seen other visits where the visits are very factual, in the sense that we’re a specialty that relies heavily on numbers and trends. And it’s very easy to just sort of say well, six months ago your GFR was this and now it’s this. And the protein in your urine was this, and now it’s this. And this is the way you’re headed and just make it very factual.” (physician 4, independent practice) •“It's because when I see the doctor I always have these blood tests that are looking at my creatinine and so on, which is probably much more important in terms of evaluating the status of my kidneys. We haven't—I'm trying to remember if we ever discussed symptoms in any depth, and I think the answer is no.” (patient 5, CKD G4)
	Mental health is not routinely addressed in CKD care.	•“I don’t screen them for mental health issues. I am very focused on CKD part of things, not that I discount the mental health part…The question is how much time do you have with the patient, and what is important to them when they bring it up to you? And I can tell you, in my practice it’s been rare that a patient will bring up the mental health issue spontaneously because I don’t always ask them directly about their mental health. I don’t screen for depression or things like that.” (physician 1, JH academic practice) •“I can do the best I can. But I get to a point where they may need to see a professional. I’m not that professional. I don’t have the expertise. I don’t have the time to get the expertise. And I don’t want to get the expertise, to the degree that a psychiatrist would. Obviously, I want to be able to help, but there is a limit as a nephrologist. And I’m happy with that limit.” (physician 8, JH academic practice)
There is hesitation surrounding the role of PROMs in clinical care.	PROM results may not be actionable.	•“You may not be able to do anything about them or you may give the patient the impression that it's from the kidney disease and so you're sort of expected to resolve that, which may not be possible.” (physician 12, independent practice) •“…if they’re starting to feel more symptomatic, what does that mean? It’s not like we’re going to start them on dialysis just yet. So I feel like it's also having realistic expectations of what we can do for these symptoms. Are these surveys just to guide us on when we should time them for dialysis? Or, for example, if they report that their muscles are more sore than before, what do I do at that point in time? Is that something I can address? Probably not. Is that something I can reverse? Probably not as well. Will I start them on dialysis at that time if they report worsening? Maybe not as well.” (physician 15, JH academic practice)
	Some users prefer to discuss symptoms and QOL through direct conversation.	•“…I feel like for most patients it's been working like just asking verbally face-to-face. I think some patients sometimes like for those symptoms, like mental health symptoms, maybe they may feel more comfortable filling out a questionnaire because maybe they don't feel comfortable opening up about certain symptoms to everyone, but I think maybe a combination of the two is probably best.” (physician 14, independent practice) •“I prefer to report them in person. I prefer in-person physician visits, but if that wasn't available and I was having specific symptoms, then I wouldn't have a problem using that to get the information to my doctor, using electronic report to get the information for my doctor.” (patient 12, CKD G4)
	Patient-physician relationships may impact the shared experience of discussing symptoms and QOL.	•“Doctors are rushed often, and they focus more on the creatinine and the phosphorous and the hemoglobin and the PTH than sometimes they do on the patient. For those types of physicians, I think this could be very helpful. For those of us who also focus as much on the patient and who have longstanding relationships with the patients, it may not be as important.” (physician 6, non-JH academic practice) •“Most of the kidney doctors I've talked to, they're interested in, in whole health and interested in - interested in kind of the whole body and not just the kidneys. And I think they also get that this is kind of a systemic disease that affects a lot of things, so I think they would be willing to, to talk to me about it.” (patient 9, CKD G3b)
There is uncertainty surrounding the sensitivity and specificity of PROMs.	Data captured by PROMs can be non-specific.	•“…so many of the symptoms of kidney disease are multi-system or spread across various systems that we really have to try to be comprehensive, in terms of asking about these things for our patients, although it's hard because so many of the symptoms are not specific to kidney disease necessarily, so it's hard to distinguish what's causing what sometimes. But I think it's pretty important. I think, you know, nephrology being a branch of internal medicine and we deal with so many interrelated problems, I think we have to do a pretty thorough job of that as opposed to more specialty, not so inclusive disciplines.” (physician 12, independent practice) •“…I guess the only thing that I thought when answering them was that some of these things for sure could be attributed to almost anything else. And I guess that's not for me to worry about. But I did think to myself, well what, you know, what if I, I don't know had a loss of appetite because I don't know I was just stressed. And I tend to just not eat because I'm you know, I'm stressed about, like my kids' school or something? I don't know. And I just kind of felt like, how—at times I was like, how do I respond to this? Am I just keeping in mind my—what I think are kidney related symptoms? But that's hard to do, you know? Because you're just like a whole person.” (patient 8, CKD G3a)
	PROMs may not capture all relevant aspects of patient care.	•“It's important for me to know if that patient has problems with transportation, if that patient has problems with mobilization, if they have to use a wheelchair, scooter, or walker, or if they're going to be able to come to their appointments. And, the financial part plays an important role, if they're going to be able to come up to their clinic appointments and be able to afford the medications that are prescribed to them.” (physician 11, non-JH academic practice) •“I think, one thing that might be worth adding or expanding the survey into is sort of asking about medication burden and just asking patients do they feel they, you know, do they feel burdened by the number of medications that they take or do they feel like they take too much medication?” (physician 12, independent practice)

Note: Participants are identified by participant type (physician or patient), participant ID, and CKD stage for patients.

Abbreviations: CKD, chronic kidney disease; PROMs, patient-reported outcome measures; QOL, quality of lift.

Physician and patient participants also identified several mechanisms by which PROMs could inform clinical care. PROMs could identify symptoms, which may be treatable with medications or lifestyle modifications and could also identify opportunities to de-prescribe medications, which could be causing symptoms. PROMs could inform discussions of advanced care and kidney replacement therapy planning, especially because PROMs would facilitate longitudinal monitoring of symptoms and QOL. One physician commented that PROM completion helped highly functional patients recognize the increasing burden of their symptoms and alerted them to the need to plan for kidney replacement therapy. PROMs could also increase patient knowledge/awareness surrounding the impact of their kidney disease. Multiple participants viewed PROMs as a mechanism to enhance patient-physician relationships, enabling patients to feel heard and feel seen, although one physician participant commented that PROMs could make some patients feel that they are just a checklist.

Motivation to use PROMs was limited by uncertainty surrounding the role of nephrology physicians in assessing symptoms, QOL, and mental health. Physician and patient participants cited a historical focus of kidney care on numbers (eg, trends in laboratory assessments). Assessment of symptoms is less traditionally incorporated into nephrology care. One physician participant alluded to not wanting to gain additional expertise in issues such as mental health.

Motivation to use PROMs was further dampened by concerns surrounding their use in clinical care. Physician participants were concerned that if symptoms were not actionable, PROMs would generate unrealistic expectations from patients. Some patient and physician participants expressed a preference to discuss symptoms and QOL through direct conversation, rather than a survey. Patient-physician relationships may also impact the utility of PROMs: one physician participant noted that PROMs might be less necessary for physicians who already have a longstanding relationship with patients, which facilitates an ability to focus on the patient as a whole.

Additionally, physician and patient participants noted that PROMs may capture data which is not specific to kidney disease. Symptoms may be multifactorial and related to patients’ comorbid conditions or other psychosocial factors. Physician and patient participants observed that PROMs may not capture all aspects of the patient experience which are relevant to CKD care, such as issues surrounding medication adherence and burden, social interactions, finances, and social determinants of health.

Data Synthesis

Guided by these data aligned with the COM-B model of behavior change,^9,10 we synthesized key factors which participants identified as mechanisms to facilitate use of PROMs in routine NDD-CKD care (Fig 1).

Figure 1.

Synthesis of participant perspectives on mechanisms to facilitate use of PROMs in routine NDD-CKD care. NDD-CKD, non-dialysis-dependent chronic kidney disease; PROMs, patient-reported outcome measures.

Discussion

PROMs capture key aspects of patients’ self-perceived health-related QOL and have been shown to enable person-centered care in other fields of medicine.^1,17 PROMs are not routinely used in NDD-CKD care, and their implementation will necessitate behavioral change by users. We interviewed nephrology physicians and people with NDD-CKD to elicit perspectives surrounding the use of PROMs in routine NDD-CKD care, specifically considering the impact of behavioral determinants. We identified key barriers related to the limited capability, opportunity, and motivation for patients and physicians to effectively use PROMs. We found consistency in patient and physician views of the need to support utilization by patients (through multiple methods of PROM collection, including the option of staff-assisted PROM completion) and physicians (through multidisciplinary/interdisciplinary approaches to manage symptoms), and a need to clarify the clinical utility of PROMs, especially if symptoms identified were unrelated to kidney disease. Perspectives were consistent for physicians working in different clinical practice settings, and for patients with different CKD stages. Our findings indicate that addressing these barriers and building the evidence base supporting the benefits of PROMs would support PROM adoption and implementation.

Broadly, our results parallel key issues highlighted in the Patient-Reported Outcomes Tools: Engaging Users & Stakeholders (PROTEUS) consortium guidelines,⁴ including the need to ensure that all patients have access to PROMs. Our study participants noted a need for multiple methods of PROM completion, outside of electronic PROMs integrated into electronic health records. Electronic PROMs can reduce time and costs required for data collection and analysis, but can exacerbate disparities if unable to be used by patients with limited literacy or access to electronic systems.¹⁸ Making electronic equipment available in clinic spaces, as well as providing paper versions of PROMs and ensuring availability in multiple languages, could increase opportunity for patients to use PROMs. In a study of PROMs collected from people on hemodialysis,¹⁹ over 60% of kidney care centers reported that patients required carer and/or nurse assistance to complete PROMs. Staff assistance has been shown to improve PROM completion rates in oncology care as well.²⁰ Resources to support staff-assisted PROM completion may therefore be needed during PROM implementation in CKD care.

Implied in these suggestions and PROTEUS guidelines is the need to integrate PROMs throughout clinic workflows, including processes by which care teams review PROMs data. Even after patients complete PROMs, participants identified several barriers to them being integrated into physician workflows and used as part of clinical care. Potential methods to promote physician opportunity to use PROMs included strategies to inform physicians of available results and use of user-friendly visualizations of PROM results (such as traffic-light systems) to help physicians quickly identify and prioritize relevant issues to address with patients.

Our data reflected another issue highlighted in PROTEUS guidelines,⁴ which is the need for users to agree on the main goals of using PROMs. Many participants highlighted a key motivation of using PROMs to be their potential to inform person-centered care and to serve as an adjunct to traditional management strategies, which parallels findings from Aiyegbusi et al.² However, other users may lack motivation to implement tools which are not shown to improve traditional patient outcomes such as mortality. Such evidence for PROMs exists in other fields of medicine. For instance, in a randomized clinical trial of patients with metastatic cancer, use of PROM-based systems was associated with increased survival,^21,22 and improvements in physical function, symptoms, and QOL.²³ Ongoing trials, such as the RePROM study for people with NDD-CKD²⁴ and the SWIFT trial for people treated with dialysis,²⁵ will provide valuable insight into the utility of PROMs for kidney patients. Electronic PROMs applied in pragmatic trials²⁶ may also facilitate data collection surrounding their effectiveness in NDD-CKD patient outcomes. A robust evidence base could generate additional buy-in from PROM users whose motivations are more aligned with changes in clinical outcomes.

In conjunction with integrating PROMs into workflows and ensuring that all users are motivated to use them, resources and education are needed to increase the capability of care teams to act on PROMs data. As noted by our participants, a lack of knowledge and skills to discern the etiology of symptoms and to manage them discourages physicians from reviewing and acting on the results. PROTEUS guidelines state that clinical teams must be equipped to respond to alerts or refer to other support services before integrating PROMs into clinical care.⁴ Although some guidelines to manage symptoms in people with CKD exist,^27,28 related training and educational materials may need to be developed and disseminated to support the capability of care teams to act on PROMs. Such strategies may contribute to a larger culture shift needed to include symptom assessment and management in the routine scope of nephrology practice.²⁹ As noted by our participants, the availability of multidisciplinary/interdisciplinary support to manage symptoms should also be incorporated into implementation planning within each care environment. Multidisciplinary care is commonly available in oncology care settings, which may, in part, explain the more common uptake of PROMs for symptom assessment and monitoring in routine oncology care.³⁰ Specifically, palliative care services have an established role in oncology^31,32 and are much less commonly used in nephrology.³³

Our study had some limitations. Some interviews were short, but given the narrow scope of the research question, we still achieved saturation of themes presented. Given an interest in implementing PROMs in our local care environment, we exclusively interviewed potential PROM users (patients and nephrology physicians) who received or provided care in the Greater Baltimore-Maryland region. Patient participants were recruited from a group of patients who had consented to participate in a separate questionnaire study,¹³ although we attempted to mitigate selection bias by selecting interviewee participants who had either completed or not completed the KDSS PROM as part of routine clinical care. Although we would expect our participant characteristics to be translatable to other regions caring for similarly diverse populations, further studies are needed to confirm generalizability. We did not consider the potential impact of multidisciplinary clinics and/or value-based care models, which were not employed for NDD-CKD in our region of practice at the time of this study. We did not include perspectives of other relevant parties, which include care partners of people with NDD-CKD, administrators, support staff, and multidisciplinary care team members. We did not explore the potential impact of incentives and reimbursement policies, which have been previously shown to result in an increase in PROM collection by kidney care teams.¹⁹ The KDSS was used as an example PROM to guide discussions, and some perspectives may have applied to the KDSS specifically rather than PROMs generally.

Notable strengths of our study include it being among very few studies reporting perspectives of users on use of PROMs or other assessments of QOL in non-dialysis nephrology care settings.^2,34, 35, 36, 37 To our knowledge, it is the first qualitative study to report perspectives on use of PROMs among US patients with NDD-CKD. We framed interview guides with the well-evidenced frameworks pertaining to the use of behavioral change theory, which will help translate concepts identified here to interventions, which can help support the desired behavior of implementing PROMs in routine clinical care.⁸

In conclusion, although PROMs represent a potential method to deliver person-centered care and to augment traditional clinical management strategies, we identified several barriers that limit the capability, opportunity, and motivation of physicians and patients to use PROMs in routine NDD-CKD care. Interventions guided by the behavior change wheel can be developed and tested for effectiveness in promoting use of PROMs in kidney care settings. Implementation science frameworks can then be used by teams of nephrology physicians, people with NDD-CKD, and multidisciplinary care partners to collaboratively co-design a PROM system, which is tailored to the needs and preferences of users in each clinical setting.

Supplementary Material

Supplementary File (PDF)

Items S1-S2; Tables S1-S2.