How the Routine Use of Patient-Reported Outcome Measures for Hemodialysis Care Influences Patient-Clinician Communication

Visual Abstract

Abstract

Background and objectives

Patient-reported outcome measures invite patients to self-report aspects of their quality of life and have been reported to enhance communication with clinicians. We aimed to examine how routine use of patient-reported outcome measures in in-center hemodialysis units influenced patient-clinician communication.

Design, setting, participants, & measurements

A concurrent, longitudinal, mixed-methods approach was used. We used data from a cluster randomized controlled trial of 17 hemodialysis units in northern Alberta that introduced a patient-reported outcome measures intervention. Patient-clinician communication was assessed using a modified Communication Assessment Tool. Using interpretive description, we explored patients’ and nurses’ perceptions of communication pertaining to routine patient-reported outcome measure use. Through purposeful sampling, we interviewed ten patients and eight nurses and conducted six observations in the dialysis units, which were documented in field notes. We reviewed 779 patient responses to open-ended survey questions from randomized controlled trial data. Qualitative data were thematically analyzed.

Results

Overall, patient-reported outcome measure use did not substantively improve patient-clinician communication. There was a small positive change in mean total Communication Assessment Tool scores (range, 1–5) from baseline to 12 months in patient-reported outcome measure use units (0.25) but little difference from control group units that did not use patient-reported outcome measures (0.21). The qualitative findings provide in-depth insights into why patient-reported outcome measure use did not improve patient-clinician communication. The purpose of patient-reported outcome measure use was not always understood by patients and clinicians; patient-reported outcome measures were not implemented as originally intended in the trial, despite clinician training; there were challenges using patient-reported outcome measures as a means to communicate; and patient-reported outcome measure use was perceived to have limited value.

Conclusions

While patient-reported outcome measures use did not improve patient-clinician communication, qualitative data suggest implementation challenges, including limited clarity of purpose and perceived limited value, that may have limited the effectiveness of the intervention.

Introduction

In 2019, there were >40,000 Canadians with kidney failure requiring dialysis, a 33% increase from 2010 (1). Dialysis is associated with high symptom burden and severity that negatively affects quality of life. Symptoms are often under-recognized and under-treated (2,3), possibly because patients are unable to share their symptoms/needs (4,5) and clinicians have difficulty identifying the spectrum and severity of patients’ symptoms/needs (2,4). Further, symptoms can change considerably over time, emphasizing the need for regular self-reporting of symptoms (6). Incorporating routine use of patient-reported outcome measures (PROMs) may address this problem. PROMs are standardized instruments for people to appraise and self-report health outcomes pertaining to their quality of life (e.g., physical, mental, functional status, and social well-being) (7). PROMs are increasingly used in a variety of patient care settings, including dialysis (8).

Theoretical and conceptual frameworks (9–11) and systematic reviews (12–14) addressing PROM use for clinical care propose enhanced communication between patients, families, and clinicians and improved processes of care and shared decision making; however, the effect on health outcomes is uncertain (10,12,13,15–19). In a realist synthesis of PROM use in dialysis care (8), two empirical studies found that PROM use supported patient-clinician communication (20,21), which may play a key role in symptom identification, patient-clinician relationships, shared decision making, and improved health outcomes. Our objective was to examine how routine PROM use influences patient-clinician communication in in-center hemodialysis units in northern Alberta.

Materials and Methods

We used a concurrent, longitudinal, mixed-methods research design (22). This study used data from a multicenter, cluster randomized controlled trial: Evaluation of Routinely Measured Patient-Reported Outcomes in Hemodialysis Care (EMPATHY) trial (23). Each cluster (i.e., in-center hemodialysis unit) was randomized to one of four study arms: (1) patients complete the Edmonton Symptom Assessment System, Revised: Renal (ESAS-r: Renal) (24), (2) patients complete EQ-5D-5L (25), (3) patients complete both ESAS-r: Renal and EQ-5D-5L, and (4) usual care (i.e., control group). Nurses were trained on and delivered the intervention of routine PROM use (described in detail elsewhere) (23,26), which encompassed (1) self-reporting with allocated PROM(s) repeated every 2 months, (2) reviewing and discussing PROM(s) scores, and (3) using decision supports and patient handouts (27) to manage symptoms at the clinicians’ discretion.

Quantitative and qualitative data came from EMPATHY trial participants from 17 in-center hemodialysis units in northern Alberta, with over 900 patients. Eligible patients were undergoing long-term hemodialysis, ≥18 years, and willing/able to complete PROMs. For the qualitative components, we purposefully sampled patients and nurses across settings, unit size, and models of care. Patients who spoke English and all nurses were eligible to participate in interviews and/or observations; participants provided informed consent for interviews and could opt out of observations. The University of Alberta Health Ethics Research Board (HREB) approved the EMPATHY trial (HREB reference #: Pro00077850) and qualitative study (HREB reference #: Pro00085021).

We used repeated cross-sectional data from a modified Communication Assessment Tool (28) (September 2018 to October 2019) to assess how routine PROM use changed patient-clinician communication (Supplemental Material). Total Communication Assessment Tool scores were computed by taking the average of all 14 items, each rated from one to five (poor to excellent), with listwise deletion in case of missing items (28). Descriptive statistics were computed for demographic and communication variables. Chi-squared tests and t tests were computed to test for differences between the intervention and control groups. Quantitative analyses were performed using STATA 14.2 (29).

We used an interpretive description approach (30) to examine how routine PROM use in in-center hemodialysis units influenced patient-clinician communication. Data sources included interviews with patients and nurses (May to November 2019), field notes from dialysis unit observations (March to July 2019), and patients’ open-ended survey responses (September 2018 to October 2019). For interviews, we recruited from all 17 EMPATHY trial units via posters with the assistance of site leads. Nurses and/or patients from eight of the sites contacted the research team to volunteer. For unit observations, we purposefully selected diverse units by EMPATHY trial study arm, setting (urban/rural), and size. People were notified of unit observations through a summary of the study posted at the unit desk and distributed.

Two trained qualitative researchers (H.S. and L.A.W.) with no prior relationships with participants conducted interviews and observations 6–12 months into the trial to ensure sufficient exposure to routine PROM use. Interviews were conducted using a semistructured guide (Supplemental Material) by telephone (n=12) or in person (n=6) according to participant choice, and they ranged from 15 to 50 minutes. Three in-person interviews were conducted in private rooms; the other three were conducted in dialysis units with patients’ consent and understanding that they could answer according to their comfort. Interviews were digitally recorded, transcribed verbatim, and verified for accuracy (H.S.). Observations ranged from 0.75 to 4.75 hours and were recorded using standardized forms and field notes (Supplemental Material). No identifiable information was collected. Finally, we compiled patients’ responses to two open-ended survey questions from the EMPATHY trial: (1) If you could make one change in the care you received, what would it be? (2) Any other suggestions to improve the quality of your care?

Three researchers (K.S.-M., H.S., and L.A.W.) used reflexive thematic analysis (31–34), including negative case analysis for meaning saturation (a more robust parameter compared with coding saturation), to understand the diversity of experiences (31,35) in three stages: read the data, generated and iteratively refined codes/code definitions related to the research aim, and met regularly to develop themes using memos and testing their accuracy within the data. Analysis of qualitative data focused on examining how routine PROM use in in-center hemodialysis units influenced patient-clinician communication. Qualitative data were managed using ATLAS.ti 8. We followed the Consolidated Criteria for Reporting Qualitative Research framework (36) (Supplemental Material). Following a concurrent, longitudinal, mixed-methods design (22), qualitative data analysis occurred at the same time as quantitative data analysis, and then findings were compared and contrasted.

Results

Quantitative Results

There were 417 (of 904) patients who completed the Communication Assessment Tool and demographic survey at baseline (response rate 46%). The average age of patients was 64 years (SD 15), 57% were men, and 87% were not working. Differences in age and education level were statistically significant between intervention and control groups; the control group was older and less educated compared with the intervention group (Table 1).

Table 1.

Baseline sample characteristics of patients in the Evaluation of Routinely Measured Patient-Reported Outcomes in Hemodialysis Care trial

Mean±SD or N (%)	Overall, n=417	Intervention Group, n=308	Control Group, n=109
Age, yra	64±15	63±16	66±14
Sex, n (%)
Men	214 (51)	161 (52)	53 (49)
Women	168 (40)	116 (38)	52 (48)
Missing	35 (8)	31 (10)	4 (4)
Education level a , n (%)
Less than high school	107 (26)	70 (23)	37 (34)
High school diploma	144 (35)	102 (33)	42 (39)
More than high school	124 (30)	102 (33)	22 (20)
Missing	42 (10)	34 (11)	8 (7)
Employment status, n (%)
Employed	51 (12)	41 (13)	10 (9)
Retired	154 (37)	113 (37)	41 (38)
Unemployed/disabled	173 (42)	121 (39)	52 (48)
Missing	39 (9)	33 (11)	6 (6)
Total Communication Assessment Tool score	4±1	4±1	4±1

^aStatistically significant difference (P=0.05) between the intervention and control groups.

Overall, PROM use did not improve patient-clinician communication. There was a slight improvement in total Communication Assessment Tool scores over 12 months in the intervention group, with a mean change of 0.25; however, this change was nearly identical to the control group (0.21) (Figure 1). Patients in the intervention and control groups reported high mean total Communication Assessment Tool scores at baseline of 4.06 (SD 0.80) and 4.32 (SD 0.72), respectively, indicating relatively good communication prior to the EMPATHY trial intervention.

Figure 1.

Change in the mean total Communication Assessment Tool (CAT) score over 12 months for intervention versus control group participants. 95% CI, 95% confidence interval; PROM, patient-reported outcome measure.

Qualitative Results

Drawing upon multiple data sources, we conducted interviews with ten patients and eight nurses (demographics are reported in Table 2), observed six dialysis units with 23 interactions between nine nurses and 22 patients, and reviewed 779 open-ended narrative responses from 510 patients.

Table 2.

Characteristics of patients and nurses who participated in interviews

Interview Participant Characteristics	Percentage or Range
	Patients, n=10	Nurses, n=8
Sex, women	50	100
Age, yr	33–78	23–60
Years worked as clinician	N/A	3–27
Years on dialysis or working in nephrology setting	1–20	1–16
Highest level of education
Grade school (grades 1–9)	40	—
High school diploma	20	—
College, trade school, CEGEP diploma, or degree	30	—
Postgraduate degree	10	—
Employment status
Unable to work	60	—
Retired	10	—
Part-time employee	10	—
Full-time employee	20	—
Ethnic background
White	60	88
South Asian	30	0
Aboriginal	10	0
Latin American	0	13
EMPATHY trial study arm
IG	100	88
CG	0	13
EMPATHY trial site
Site 1: Urbana (IG)	30	—
Site 2: Urban (CG)	—	—
Site 3: Urban (IG)	—	—
Site 4: Urban (IG)	—	—
Site 5: Urban (IG)	10	—
Site 6: Ruralb (IG)	—	—
Site 7: Rural (IG)	10	—
Site 8: Rural (IG)	10	13
Site 9: Rural (CG)	—	—
Site 10: Rural (IG)	—	13
Site 11: Rural (IG)	—	—
Site 12: Rural (CG)	—	—
Site 13: Rural (IG)	30	25
Site 14: Rural (IG)	10	25
Site 15: Rural (CG)	—	13
Site 16: Rural (IG)	—	—
Site 17: Rural (IG)	—	—

N/A, not applicable; EMPATHY, Evaluation of Routinely Measured Patient-Reported Outcomes in Hemodialysis Care; IG, intervention group; CG, control group.

^aUrban site is defined as >10,000 people.

^bRural site is defined as <10,000 people.

The qualitative findings support the quantitative findings and provide in-depth nuances and insights into why PROM use did not improve patient-clinician communication. A summary of findings is provided in Table 3. Illustrative quotes (q1–q103) are provided in Table 4 along with examples of negative (minority) cases (q-nc1–q-nc16) to represent a diversity of perspectives.

Table 3.

Summary of qualitative findings

Qualitative Findings	Subtopics	Negative Cases
PROM use did not improve patient-clinician communication	Limited time for and/or quality of communication with clinicians Lack of continuity in care by the same dialysis health care providers, resulting in poorer communication A generic versus individualized approach to care meant patients had to advocate for their needs	Patients satisfied with care and communication
Insight 1: Purpose of PROM use was not always understood	Patients’ responses to PROMs conflicted with nurses’ clinical assessments of, communications with, and observations of patients Patients unsure of how their PROM responses were used and by whom Protectionism	Some understanding of the purpose and use of PROMs
Insight 2: Unintended implementation of PROMs	Negative presentation and reception of PROMs PROMs not always patient reported Barriers to patients completing PROMs PROM responses not reviewed or discussed	Neutral or positive presentation and reception of PROMs PROMs self-reported PROMs responses reviewed and discussed
Insight 3: Challenges using PROMs as a means to communicate	Difficulty understanding PROM questions Difficulty with PROM response scales Prefer to communicate through conversation rather than numeric expression on a PROM Patients prefer to initiate conversations with their clinicians if they have health concerns Patients do not want to be reminded of their symptoms/illness through PROM use Patients exert control of communication by reporting “no symptoms/problems” on PROMs	Easy completion of PROMs
Insight 4: PROM use was often perceived as having limited value	Frequent patient-clinician interaction already promoted communication Prior close relationships between patients and nurses Perception that symptoms cannot get any better PROMs administered too frequently PROMs do not ask relevant questions to each patient’s specific needs	PROM use prompted communication above usual care PROMs use provides insight into care that might not have been obtained during usual care

PROM, patient-reported outcome measure.

Table 4.

Illustrative quotes and observation field notes

Subtopic	Illustrative Quote
PROM use did not improve patient-clinician communication
PROM use did not improve patient-clinician communication	“I don’t feel like it’s made [communication] any better.” (nurse interview 11, q1)
	“[PROMs] are not productive.” (patient survey 381, q2)
	“[PROMs] are tedious.” (patient interview 20, q3)
	“Everyone [already] speaks outs about their issues.” (patient interview 540, q4)
Limited time for and/or quality of communication with clinicians	“Listen to patients more.” (patient survey 73, q5)
	“More staff quality time.” (patient survey 98, q6)
	“Have more time to provide information.” (patient survey 99, q7)
	“More information about the overall situation/sickness—more detailed information, navigation (i.e., time frames).” (patient survey 159, q8)
	“Spend time to find out what's going on.” (patient survey 354, q9)
	“That was the biggest frustration, the lack of information, especially from nurses, and getting questions answered was pretty depressing.” (patient interview 538, q10)
Lack of continuity in care by the same dialysis clinicians, resulting in poorer communication	“The next nephrologist comes in the next month, then you got to regurgitate everything that you told the last [one]. But then the next [nephrologist] is going to have a different opinion on what the last [one] did and then you start having a fight, ‘well, this was working, why you trying to change this?’ So, consistency I think is [key because] people get frustrated.” (patient interview 538, q11)
	“Communication and making sure nurses read my chart right, and not having to remind them on a consistent basis of meds given etc.” (patient survey 174, q12)
A generic versus individualized approach to care meant patients had to advocate for their needs	“Have a primary nurse whose job is to know you, not just to get the forms filled out.” (patient survey 361, q13)
	“Attention to [patients’] specific needs. More attention to senior issues—social worker visits.” (patient survey 304, q14)
	“I feel [PROMs are] really impersonal, and they were just doing it because they had to …. One thing I’ve noticed about dialysis is there’s a certain protocol. You’re treated like a senior [even if] you’re my age of 35, and you’ve got to really be your own advocate just to have things, like getting a chair instead of a bed so you can bike.” (patient interview 535, q15)
	“I think [clinicians] generally categorize everyone into one group rather than breaking up the age groups. There are different ways those groups handle mental health. When you start lumping everyone together, the 25-year-old is getting the same treatment as the 50, 60-year-old, and it’s not working. I think [assessing] everyone on a case by case basis will alleviate some mental health [issues] rather than generalizing things.” (patient interview 538, q16)
	“[I want] to have someone I can talk to about my concerns whom in turn could give me help and feedback but take into account my life and not just give me information I already know.” (patient survey 350, q17)
	“Really listen to the patient with regards to what they think is happening with their own body and taking what they say into consideration when deciding on a treatment plan.” (patient survey 495, q18)
	“I feel as though doctor visits once a week do very little when they aren't receptive to patient complaints, needs, and requests.” (patient survey 188, q19)
Insight 1: Purpose of PROM use was not always understood
Patients’ responses to PROMs conflicted with nurses’ clinical assessments of, communications with, and observations of patients	“[Patient responded] no [pain] on the [PROM] survey, but I remembered her telling me that she had pain … I find it is difficult to assess some patients using [PROMs] because I feel like they don't answer the questions appropriately …. Maybe we shouldn't, but you kind of have to almost override what they're [reporting on the PROM] for their own health. They can't just keep going on like this … but is it wrong to disagree with what the patient is [reporting]? I mean, if a patient says they're in pain and you say, ‘no, you're not in pain,’ obviously that's [wrong]. But then if the patient’s [reporting] they're not in pain when they are in pain, then what do you do? Ethically, that would be an issue.” (nurse interview 11, q20)
	“We had a patient who’s got some significant body odour, and we were concerned about his hygiene. First, we were just letting him know that it was there and asking if he had any concerns and just letting him know we’re here to help him if he needs anything. Then, the next time we moved to, ‘Do you need any homecare? Are there any challenges that you’re having showering?’ And I mean he’s done [the PROMs] assessments and he’s putting nothing down there as a problem, but we know he needs both knees replaced and he uses a walker, so we know he has some obstacles.” (nurse interview 13, q21)
	“On the mobility dimension, when patient responded he had no problems, nurse replied, ‘but you use a wheelchair.’ Patient agreed but still insisted he has no problems [with mobility].” (unit observation 5, q22)
Patients unsure of how their PROM responses were used and by whom	“I don’t remember anyone explaining [PROMs] to me. I mean, they might have. I was pretty new to dialysis when that was given to me. I just assumed it was part of regular protocol.” (patient interview 535, q23)
	“But who’s looking at it? Is that information just for research purposes only, or are you actually taking this information and giving it to a doctor?” (patient interview 540, q24)
	“I'm not sure [if anyone explained why I was asked to complete PROMs]—maybe at the beginning when I started dialysis. But, I think just to keep tabs on us, I think. I really don't really know.” (patient interview 536, q25)
Protectionism	“A lot of the people are scared of saying, ‘Yeah, I'm depressed’ [on a PROM] because we know where that leads. It starts delaying transplant because then it starts getting to the whole ‘Are you compliant, not compliant? Are you still with it or are you not with it?’ It’s a complex situation, asking those kinds of general questions … I just thought [PROMs] needed to be done on my end. I don't know if it’s like a checkmark in my file just to say okay, it’s been completed. I was really paranoid about being compliant about everything.” (patient interview 538, q26)
	“People don’t refuse to do certain things, or they’ll do pretty much whatever needs to be done, you know. They’re not shortening their runs, they’re what we would traditionally call ‘complaint with their treatment,’ so you know they’re not looking for ways out of doing things. They have in the past done things [to be compliant], but they’re starting to become more vocal now that they don’t want to be doing [PROMs].” (nurse interview 13, q27)
	“[Patients] may not admit to problems, [such as] mobility or self-care issues, because of fear of losing their independence and requiring assisted living support.” (unit observation 5, q28)
Negative cases for insight 1
Some understanding of the purpose and use of PROMs	“I just basically said, ‘Now we’re doing a symptom assessment just to know how you are doing overall so see if any symptoms have worsened and to see if we could help you with your treatment.’” (nurse interview 22, q-nc1)
	“They explained that the surveys were basically to help the patients and to figure out what’s best for our patients.” (patient interview 20, q-nc2)
Insight 2: Unintended implementation of PROMs
Negative presentation and reception of PROMs	“Site lead told patient, ‘It's that time of the month again’ in reference to completing the PROM.” (unit observation 6, q29)
	“We were really trying to be objective when we rolled it out and ask it in a way that was not ourselves rolling our eyes at it or poo-pooing it. We were trying to roll it out in a way that we should.” (nurse interview 13, q30)
	“Nurse approached this patient while she was sleeping and asked if it was ok if she ‘bugged her’ to go through the [PROM] …. The patient seemed slightly irritated or tired and said she just wanted to go back to sleep. The nurse said that she was ‘done bugging her.’” (unit observation 5, q31)
	“Patient replied with a comment, ‘Not again’ in reference to completing the PROM.” (unit observation 6, q32)
	“I just think [PROMs] are a pain in the ass [laughs].” (patient interview 20, q33)
	“Nurse confirmed that no patient was willing to be observed today because they do not like the surveys.” (unit observation 4, q34)
	“Nurse said that patients have a negative attitude toward the [PROMs]. The nurse said that many will roll their eyes when she administers it to them …. Nurse also said that she has [presented PROMs] in many different ways with patients, and each time, it elicits the same negative response from patients.” (unit observation 10, q35)
	“I think that [PROMs] are a waste of time.” (nurse interview 12, q36)
PROMs not always patient reported	“Nurse woke up patient to administer the survey. She knelt down beside the patient and administered the PROM orally …. Nurse generally asked the EQ-5D questions as affirmatives to patients (e.g., ‘you don’t have any problems with your mobility, right?’ or ‘nobody helps you bath, right?’).” (unit observation 5, q37)
	“Nurse asked each [Edmonton Symptom Assessment System, revised: Renal] symptom as a yes or no question—if they were experiencing that symptom or not. She asked probing questions thereafter. It seemed as though the nurse was then assigning ratings based on what the patient said to the [Edmonton Symptom Assessment System, revised: Renal] scales …. The nurse interpreted the patient's answers and filled in the responses to the PROMs.” (unit observation 2, q38)
Barriers to patients completing PROMs	“Some of them forget their glasses, then they can't read it on their own, and I have to read it for them … I mean even from the 60-year-olds. I guess it just kind of depends on like their health state.” (nurse interview 11, q39)
	“There’s always a mix and it depends, you know, there’s lots of factors, some is [eye] sight, some is cognitive impairment, another is literacy, another is fatigue.” (nurse interview 13, q40)
	“Some of the patients I would go down through the questions with them because they’re not able to read or write or [have] a language barrier and they don’t really understand.” (nurse interview 21, q41)
	“Well, we have very elderly people here that can’t really see very well …. So then in that case, we would help them with it, for sure.” (nurse interview 533, q42)
	“Some people have vision impairment and different things, and some people just want us to read the questions.” (nurse interview 534, q43)
	“Nurse administered EQ-5D to patient orally because patient didn't have his glasses. Nurse knelt down beside patient so she was eye level to him. Nurse recorded responses on a laminated sheet.” (unit observation 5, q44)
PROM responses not reviewed or discussed	“I think definitely those surveys would be helpful … if somebody gave you feedback on that survey.” (patient interview 536, q45)
	“I know it’s got to be done, but … can we have results?” (patient interview 537, q46)
	“And all the surveys I filled [in]. I don’t know if somebody looked at it or not.” (patient interview 544, q47)
	“I filled it [PROM] out. The nurse kind of looked at it funny and then just said ‘Are you okay?’ ’Yeah, I'm okay.’ And that was the end of it really. That’s how it went for administering that test.” (patient interview 538, q48)
	“I just thought it was hospital protocol, and they were just doing it because they had to. It doesn’t really feel like too much would come from it.” (patient interview 535, q49)
	“We have a task list for the month, and then most times, [PROMs] are left for last. So, it’s just scrambling to get them done before the month ends really. But typically, there is lots of time, but it’s usually like the last week we’re doing them, and then, we’re trying to get them all done on time.” (nurse interview 21, q50)
Negative cases for insight 2
Neutral or positive presentation and reception of PROMs	“There were a couple with eye rolling, but some [patients] didn’t mind.” (nurse interview, q-nc3)
	“It doesn’t matter to me, I don’t mind filling them [PROMs] out.” (patient interview 542, q-nc4)
PROMs self-reported	“It’s just basically on my own. She just said ‘can you fill this out whenever you have time during your run,’ so I just did it at the beginning.” (patient interview 538, q-nc5)
	“Most of our patients were actually just filling them [PROMs] out on their own, and then, we would look at the results.” (nurse interview 533, q-nc6)
PROM responses reviewed and discussed	“We’d punch [the PROMs responses] in the computer, and we’d do their report card, and then, we’d go back and talk to them about the different health issues they may have and also compare [results] from 1 month to another; see if it was any better or any worse, and if they needed any handouts. Most of the time most of these people were seeing their general practitioners anyhow, but oftentimes, we’d try and help them if they had issues. And then after we discussed that, we’d chart the handouts that we give them and any conversations we have with them.” (nurse interview 533, q-nc7)
Insight 3: Challenges communicating through PROM completion
Difficulty understanding PROM questions	“How do I rate this, my well-being?” (patient interview 539, q51)
	“I don't know [if patients would recognize symptoms as being ‘discomfort’]. I would probably have to explain that to them. Like, I would say, ‘discomfort is this and this’ …. Some [patients], I don't know if they understand the questions at all.” (nurse interview 11, q52)
	“The best well-being is always an incredibly difficult thing for people to answer. That as well as the ‘How good or bad your health is today.’ Best possible health and worst possible health you can imagine is another one that they don’t seem to like putting a number to [about] how they actually feel. And what does that mean, you know?” (nurse interview 13, q53)
	“We have very elderly people here … [who] have a hard time understanding the questions.” (nurse interview 533, q54)
	“Nurse asked patient if he had any feelings of anxiety. The patient asked for clarification …. Patient didn't understand the question about ‘best well-being.’ Nurse [rephrased the question and] asked, ‘How is your well-being, your overall health?’” (unit observation 2, q55)
	“Nurse commented that almost half of patients don’t understand the [PROM] questions …. The patient seemed confused by the [PROM] questions and was not able to provide the nurse with answers. For example, the patient wouldn't give a percentage for best overall health on the VAS, so nurse settled on 75% to which the patient said, ‘Sure, why not.’” (unit observation 5, q56)
Difficulty with PROM response scales	“Nurse said it’s hard for some patients to quantify their experience with symptoms.” (unit observation 10, q57)
	“The quantitative part of it is hard. They have a hard time, like when it's asking you to rate your health on a scale of zero to a hundred. They have a hard time saying a number. Usually, I have to suggest a number to them, and then, they will either agree or disagree.” (nurse interview 11, q58)
	“Nurse asked the patient to rate his overall health from zero to 100. Patient said “50/50.” Nurse asked patient why it is 50/50. Patient said he didn't know why. Patient said, ‘Maybe it was a three or four.’” (unit observation 2, q59)
	“[Some patients] were finding it like a test and found it hard to identify a number [for their response] … it would stress them out trying to actually identify a number …. It’s sort of a ‘Well, I wonder if this is accurate’ or ‘I wonder if I’m right’ or ‘What if it’s not exactly what I mean.’” (nurse interview 13, q60)
	“[PROM] does ask about levels of pain and it’s hard to [rate]. Some guys can stand more pain than others. I'm one that can stand quite a bit of pain, so my level’s a lot lower than that one over there …. [Pain] it’s a ten. How bad is that? Walk around screaming, or what? You know, [my pain] never was like that …. Some of the questions are a little hard to answer. It takes some thinking …. The scale is about [where] you think you are on the scale. I don’t know. I always try and put [my answers] all at the same place … I usually put about 50/50.” (patient interview 542, q61)
	“I think [responses would be more accurate] if we understood the scale, rather than just being an arbitrary one to ten scale because it’s—one to ten’s hard. I almost think one to five makes more sense. It’s really hard to tell if you’re a six or a seven …. Mini illustrations to line up with the one through ten [response scale] that would be helpful. Emoji faces would probably be really easy … I think it would be good to have little examples. Because, I think there’s certain pains that people can relate their chronic pain to, right? Like a bee sting versus suffering from an allergic reaction to a bee sting.” (patient interview 535, q62)
Prefer to communicate through conversation rather than numeric expression on a PROM	“Honestly, I think [it should be] an assessment of an individual, rather than just circling something [on a PROM] …. Actually talking to someone would nail down that mental health situation and I think alleviate a lot of those problems people have.” (patient interview 538, q63)
	“A good nursing assessment … it’s just a gentle way in rather than, ‘Okay here’s the clipboard, I want you to answer these things.’” (nurse interview 13, q64)
	“[Nurse] explained that she has the [PROM] results in mind, and she knows what she needs to follow up with patients about. She'll prompt patients about these things in conversations with patients in a more natural way [rather than reviewing the PROM results with patients].” (unit observation 6, q65)
	“I think that [patients] prefer that [bringing up issues in conversation] because they feel like they get more results from just talking to us, individually, than doing an actual [PROM] survey …. And they see, ‘Okay, I told my nurse this. This is what I got out of it. That's like a direct result from it.’” (nurse interview 11, q66)
Patients prefer to initiate conversations with their clinicians if they have health concerns	“I participated for the first couple of [PROMs], and then, I was like, ‘Okay, this is just getting monotonous.’ If something changes, I'll let you know, but to keep throwing me the same paper every month, you're wasting my time. I don't want to be woken up for a piece of paper that’s been filled out.” (patient interview 540, q67)
	Interviewer: “Is it fair to say that any concerns you have, you’ll bring up on your own with the nurse, or …” Participant: “Me, personally, yes” (patient interview 535, q68)
	“[Patients] started just telling us, ‘Do we have to do this? I’d rather not do this.’ And ‘I will ask you if we have a problem, but you guys are always asking us anyways every day when you do your assessment about many of these things, and we know we can come to talk to you if there’s a challenge and that you’ll help us.’” (nurse interview 13, q69)
	“[Patients] don’t want to be bothered. They go on dialysis and just want to be left alone for the most part … I find the patients that have been here longer are the ones that refuse to do it or they grumble about it.” (nurse interview 21, q70)
	“Nurse explained that some patients don't want to talk about symptoms that they perceive as being sensitive with dialysis staff. In addition, patients view their dialysis as treatment and just want to get their treatment completed and only talk to staff about their dialysis and that's it. Patients want to keep their dialysis separate from the rest of their lives, including other health issues or concerns. She said that if patients have a problem that they really want to address, they see their family doctor.” (unit observation 10, q71)
Patients do not want to be reminded of their symptoms/illness through PROM use	“Constantly filling out [PROMs], it’s just bringing us back down … we don’t like to talk about our illness all the time. I would rather tell my nurses what they need to know about me—if I’m having a bad day or feeling crappy or whatever, I tell them … I look at it this way; I’m just existing …. It’s a constant reminder that we have to go and do this survey and that basically we’re just basically existing. We don’t really have a life.” (patient interview 20, q72)
	“Some [patients] say when they come to dialysis that they don’t want to think about some of these things [on the PROMs]. They just want to come and run and have peace.” (nurse interview 12, q73)
	“Some [patients] expressed, ‘I don’t appreciate being asked these things or having to do these questionnaires when I come to dialysis. I’ll ask you if I have a concern, but I’m trying to forget about this chronic illness that I’m dealing with, so if you would like to have empathy and improve my quality of life, why don’t you get us nice TVs and things that we could use to escape some of the challenges we’re having already having to be here?’” (nurse interview 13, q74)
Patients exert control of communication by reporting “no symptoms/problems” on PROMs	“A lot of times [patients] just mark zeroes on everything so you knew that they were just trying to get through it and didn’t want to discuss anything.” (nurse interview 13, q75)
	“Nurse further mentioned that a lot of patients will report no symptoms/problems on their surveys because patients don't want to be followed up with.” (unit observation 10, q76)
Negative cases for insight 3
Easy completion of PROMs	“In contrast to these [examples of patients not understanding PROM questions], one patient was able to make a connection between itchiness and the [PROM] question about pain or discomfort.” (unit observation 5, q-nc8)
	“They're pretty simple questions, so a person can pretty well understand them and fill [the PROM] out themselves. If there are any problems or not understanding, then the nurse is there to help support me.” (patient interview 537, q-nc9)
Insight 4: PROM use was often perceived as having limited value
Frequent patient-clinician interaction already promoted communication	“[The nurses] are awesome. They ask us every day, and they know a lot of times just by the looks on [patients’] faces, you know, they don't even have to ask. They already know …. Everyone [already] speaks outs about their issues.” (patient interview 540, q77)
	“I usually am pretty good at knowing what's going on with my body. So the [PROM] really doesn't do anything for me because I pay attention … as soon as something is off, I say something to the nurse and then I usually talk to the doctor about it too.” (patient interview 536, q78)
	“[The nurses] know us better than anybody else.” (patient interview 20, q79)
	“We talk to these patients every single day. We know about every part of their life.” (nurse interview 11, q80)
	Interviewer: “Have you ever learned anything new or found out something about your patients based on their responses to the surveys?” Respondent: “No, I found that it was more in my assessment of the patient each day when they came in and just doing a good nursing assessment based on a natural conversation and just what you would do with any patient as a registered nurse.” (nurse interview 13, q81)
	“We have the same patients coming through all the time, and we’re such a small unit. We get to know all of the patients very well.” (nurse interview 21, q82)
	“I’ve known most of [the patients] for several years now, so there were no surprises on the [PROMs] when they filled them out.” (nurse interview 533, q83)
	“The nurse explained that any issues or concerns with a patient would come up during their time together on the unit. That is, either patients will bring up issues they’re concerned about and willing to discuss in the open on the dialysis unit, or the nurse will bring up issues based on the nurse’s observation (e.g., patients smell bad, so ask patients if they are willing to discuss it or open to a referral for help with self-care) …. When asked if [PROM surveys] ever helped her understand something new about her patients, she said nothing she wouldn't be able to find out through her usual care interactions with patients.” (unit observation 10, q84)
Prior close relationships between patients and nurses	“I’ve been here a long time, and the patients have been here a long time, so I kind of know the patients really well.” (nurse interview 533, q85)
	“I don't necessarily think that it helps us gather more information for our patients because most of them are pretty open to us, and they tell us lots of stuff, regardless of the survey.” (nurse interview 11, q86)
	“We’ve become a small family in here, you know. We seem to care about each other, you know. If somebody’s missing, we want to know what happened, is everybody okay, it’s nice though because like I say, it becomes a little family, a dialysis family.” (patient interview 541, q87)
	“I talk to the nurses … I can tell them how I feel, I don’t have to put it on paper … I have nothing holding me back to say if I feel something. It’s kind of like it’s a second home [here].” (patient interview 544, q88)
	“She feels that this survey might have value in the larger busier units, but in [their unit], they have a close relationship, and the survey doesn’t really help with the delivery of care.” (unit observation 5, q89)
	“[Observer] asked [the nurse] if any patients had ever become aware of an issue they were having because of the PROM; she said no. The nurse said that they are ‘like a little family,’ where you know about patients’ grandchildren and if they are upset about something in their lives or if something is not right.” (unit observation 6, q90)
Perception that symptoms cannot get any better	“There are some patients who are like, hey I don't really need this handout. Like I've been on dialysis for a billion years. I don't need the handout to help me with anything.” (nurse interview 11, q91)
	“Some have felt like because they were doing a symptom assessment that this is not going to change anything … because they have a chronic illness and they already know what the possible solutions are for fatigue, but they already know there isn’t really a fix for fatigue it just comes with the territory.” (nurse interview 13, q92)
	“She said that it is a constant reminder of some symptoms that cannot totally be fixed. For instance, the nurse gave the example of fatigue. Many dialysis patients experience tiredness and fatigue, and there are some things they can do to help preserve their energy, but it will never be totally resolved. The nurse thinks patients get tired of quantifying their fatigue when they know it can't get any better. She further mentioned that she thinks patients resent being asked questions that they can't do anything about.” (unit observation 10, q93)
	“There’s nothing you can really do to kind of mitigate that situation until obviously you get a transplant …. It’s just the situation, it’s just this is what those symptoms are on dialysis” (patient interview 538, q94)
	“Doing dialysis [for] 9 years is kind of sad, never get a transplant … doing all the surveys …. For me, it doesn’t change nothing in my health. Don’t ask. ‘Do you understand now?’” (patient interview 544, q95)
PROMs administered too frequently	“In conversation with the site lead, she mentioned that supportive care is completed too frequently and suggested that every 3 months or quarterly would be better.” (unit observation 3, q96)
	“It can be redundant, and doing it monthly, I found it very redundant, patients would [say] ‘Oh we have to do this again?’” (nurse interview 2, q97)
	“To me, nothing changes [every 2 months] …. Like once every 6 months, sure, because things do change in that time.” (patient interview 7, q98)
PROMs do not ask relevant questions specific to each patient’s specific needs	“Some people expressed, ‘I don’t appreciate being asked these things or having to do these questionnaires when I come to dialysis. I’ll ask you if I have a concern, but I’m trying to forget about this chronic illness that I’m dealing with, so if you would like to improve my quality of life, why don’t you get us nice TVs and things that we could use to escape some of the challenges we’re having already having to be here?’” (nurse interview 13, q99)
	“They don’t really know what I’m going through because number 1, none of this applies to me …. I’m just existing basically is how I feel.” (patient interview 20, q100)
	“But the getting around, that’s never asked—how do you get to your session? I don’t know if that should be a part of our nurses’ and the doctors’ care or—but it is an issue sometimes because, well, not all people are well off to take the transit or have somebody bring them there, or if they have to travel on their own, do they have enough gas money and all that. Some people travel an hour and a half just to get to a session and that’s three times a week and that’s quite a bit of gas. You know, there’s issues there too …. And that’s another thing that should be really addressed. You better be able to take a person’s mind off what the dialysis is doing.” (patient interview 537, q101)
	Interviewer: “What would be the right types of questions? Or can you tell us a bit more about that? Like what are your concerns?” Respondent: “I think your daily life stuff. How’s your family life, what’s your work situation like?” (patient interview 538, q102)
	“The nurse said that things like better TVs and a comfortable place to sleep during dialysis would improve patients’ quality of their care more than the surveys.” (unit observation 10, q103)
Negative cases for insight 4
Adequate PROMs frequency	Interviewer: “Do you like how often do you get the survey? Is it too much, is it not enough?” Respondent: “Not that often. It doesn’t seem that often. It’s good.” (patient interview 9, q-nc10)
PROM use prompted communication above usual care	“[Conversations] are not as in-depth as they are when you do [PROMs]. You know you ask any recent bleeding, any pain, any nausea or vomiting, swelling issues, but you don’t really touch on that restless legs until you start getting deep into those surveys … when you’re able to sit down and ask those questions with the patient you can go more in depth of why these symptoms, why would they rate it like an eight out of ten for restless legs.” (nurse interview 22, q-nc11)
	“[PROMs helped me discuss] skincare. I have extremely dry skin, and that’s part of the [PROM] and I realize oh, yeah, that’s happening. I’d better figure out what to do.” (patient interview 541, q-nc12)
	“The site lead mentioned that it is useful to compare current results to past PROMs results, where she can see how scores have changed and she can ask the patient what happened for the change to occur.” (unit observation 6, q-nc13)
PROMs use provided insight into care that might not have been obtained during usual care	“It does point out some issues that the patients are having … it does open your eyes to some things that maybe you get busy and you don’t think or the patient doesn’t think to tell you about it.” (nurse interview 12, q-nc14)
	“I mean if it was a newer patient then I would learn lots [from PROMs].” (nurse interview 533, q-nc15)
	“[PROMs] made me think how bad [symptoms] were, or you know, something like that. So yeah, it actually made me think a little more, yeah.” (patient interview 542, q-nc16)

PROM, patient-reported outcome measure; q, quote; q-nc, quote from a negative case; VAS, visual analogue scale.

Patient-Reported Outcome Measure Use Did Not Improve Patient-Clinician Communication.

Participants indicated that PROM use did not improve communication (q1–q4). As one nurse summarized, “I don’t feel like it’s made [communication] any better” (q1). Patients wanted more and better-quality patient-clinician communication (q5–q10). This included continuity of care from the same dialysis clinicians (q11 and q12), which was not the current care model. From patients’ perspectives, PROM use did not improve communication in this context. Patients wanted clinicians “whose job is to know you, not just get the [PROMs] filled out” (q13) and who could pay “attention to [their] specific needs” (q14), indicating that a generic approach to care meant patients had to advocate for their individual needs (q15–q19). Although routine PROM use is meant to standardize collection of patient-reported information for tailored follow-up care, patients often perceived that their care was not being personalized to them.

Insight 1: Purpose of Patient-Reported Outcome Measure Use Was Not Always Understood.

Despite recruitment of site champions and clinician workshops at the beginning of the trial, nurses and patients did not always understand why PROMs were being used in dialysis care, although there were a few exceptions (q-nc1 and q-nc2). This knowledge gap may have limited or even hindered patient-clinician communication. Nurses did not know what to do when patients’ PROM responses conflicted with their clinical assessments of, communications with, and physical observations of patients. This led some nurses to suggest that “[patients] don't answer the [PROMs] questions appropriately” (q20–q22), and the nurses were simultaneously aware of the ethical issues of questioning patients’ responses. Additionally, some nurses did not know what to do when patients reported “no problems/symptoms” on PROMs when they had verbally expressed problems/symptoms to nurses, resulting in tension about how to best support patients. Many patients were unsure of the purpose of PROMs, wondering if it was “just part of regular protocol” (q23), for “research purposes” (q24), or to “keep tabs on [us]” (q25). Some patients were unsure, at times fearful, about how their PROM responses were used and by whom. This may have resulted in protectionism or safeguarding against perceived consequences. To “protect” themselves, they completed PROMs to avoid being labeled “not compliant” and/or provided inaccurate responses to avoid possible consequences (e.g., losing independence or being denied a kidney transplant; q26–q28). Patients’ strategies to protect against possible consequences may have contributed to nurses’ perceptions that patients’ PROM responses did not accurately reflect patients’ symptoms/concerns in some cases.

Insight 2: Unintended Implementation of Patient-Reported Outcome Measures.

Overall, PROMs were not always administered, completed, or used as intended, despite clinician training and site champions. PROMs were routinely collected as intended but with a decline over time, specifically lower collection of the EQ-5D-5L in the study arm where both ESAS-r: Renal and EQ-5D-5L were administered (Figure 2). Although overall PROMs collection was successful, methods of administration, completion, and use varied. At times, PROMs were presented by nurses and received by patients in a negative light. For example, a nurse said, “It’s that time of the month again,” or a patient responded, “Not again” (q29–q36). It was unclear if nurses’ administration negatively influenced patients’ reception of PROMs or vice versa. PROMs were not always patient reported. Nurses often read PROMs to patients who needed or wanted assistance, but in doing so, nurses rephrased questions and response scales, interpreted, and then assigned a number (q37 and q38). Nurses identified barriers to patients completing PROMs unassisted, including “vision impairment,” poor “health state,” “cognitive impairment,” “fatigue,” “language barrier,” or literacy issues (q39–q44). Even when PROMs were patient reported, nurses did not always review responses or discuss them with patients, which patients noted with overall displeasure. More often, PROMs were considered simply a task to complete rather than something to inform or improve patient care (q45–q50). Although in the minority, a few exceptions were noted where patients expressed that they “didn’t mind” completing PROMs (q-nc3 and q-nc4), patients did complete PROMs independently (q-nc5 and q-nc6), and nurses reviewed PROM responses and discussed them with patients (q-nc7).

Figure 2.

Proportion of PROMs collected during the Evaluation of Routinely Measured Patient-Reported Outcomes in Hemodialysis Care trial study period by PROM type. EQ-5D-5L, European Quality of Life Five Dimension; ESAS-r: Renal, Edmonton Symptom Assessment System, Revised: Renal.

Insight 3: Challenges Using Patient-Reported Outcome Measures as a Means to Communicate.

Multiple aspects of PROM completion and discussion were regarded as hindrances to communication. In some cases, patients had difficulty understanding PROMs questions, including questions about “discomfort” or “overall well-being” (q51–q56). Some patients also experienced difficulty with the response scales, including how “to quantify their experience with symptoms” (q57–q60) or if their response should be compared and contextualized against other patients’ experiences (e.g., ability to tolerate pain; q61). A patient recommended providing “mini-illustrations” or “emoji faces” to help people understand the response scale (q62). Nevertheless, a few patients found it easy to complete and understand the PROMs (q-nc8 and q-nc9) Some participants preferred narrative discussions over numeric expression on a PROM (q63–q66). Lastly, patients wanted choice/agency in how and when to communicate about their symptoms or care, including not discussing symptoms/problems. Some patients preferred to initiate conversations if they have health concerns (q67–q71) or did not want PROMs to remind them of their symptoms/illness (q72–q74). Echoing previously discussed notions of self-protectionism, some patients reported “no symptoms/problems” on PROMs to avoid discussion (q75 and q76). Therefore, PROM use may reduce patient choice about when and what to discuss with clinicians.

Insight 4: Patient-Reported Outcome Measure Use Was Often Perceived as Having Limited Value.

PROM responses were not consistently seen by patients or nurses to contribute new information that was not already obtained through the predialysis assessment checklist or frequent patient-clinician interactions due to the nature of ongoing dialysis (q77–q82). PROM use was not perceived to prompt more in-depth communication about patients’ concerns because patients and nurses already knew each other so well (q83–q88), describing the experience as “a second home” or a “dialysis family.” Some patients and nurses perceived that symptoms/concerns identified through PROM use could not get any better because they were already optimally managed or the result of dialysis treatment (q89–q95). The frequency of PROM administration (every 2 months) was critiqued as too frequent to detect health changes (q96–q98), although this was not consistent (q-nc10). Some participants also perceived PROMs to measure items that were irrelevant to patients or their specific needs (q99–q103). However, some saw the value of PROMs to enhance care (q-nc11–q-nc13) or insights that might have otherwise been missed (q-nc14–q-nc16). Overall, communication may not have improved through PROM use because it was perceived to be of limited value for this patient population.

Discussion

We found that routine PROM use did not improve patient-clinician communication in in-center hemodialysis units in northern Alberta. Patients’ Communication Assessment Tool scores were high at baseline; thus, there may have been little room for improvement. Our findings are contradictory to the literature on PROM use theory (9–11,37,38) and from other clinical settings (12,13). The qualitative results corroborate the quantitative data but interrogate potential reasons for this divergence; each of the four insights is now discussed in relation to other scholarly work.

The overarching purpose for PROM use in hemodialysis care was often misunderstood by patients and nurses alike. Although nurses received training at the beginning of the trial, patients did not. Training and supporting clinicians to use PROMs have received the lion’s share of the focus in the extant field (38,39). Looking toward training patients may be a worthy pursuit that has received limited attention (40). Although PROM use is intended to be used in follow-up discussion and shared decision making to individualize care to a person’s needs, it was instead perceived by participants in our study to standardize care. Rather than seeing PROM use as a way to facilitate communication, many perceived it to curtail communication. Although this finding is unique to dialysis, it has previously been identified (41).

Implementation of PROM use did not always transpire as planned. It is possible we unearthed this finding because of our in-depth qualitative data, including observations in hemodialysis units and triangulation of data, a rare opportunity many PROMs studies do not have. It is also possible that this happens more often than is reported in larger PROMs studies. Confirmatory evidence is seen in similar observations about implementation raised in previous qualitative and mixed-methods PROM studies in dialysis (42,43) as well as by Greenhalgh et al. (14). Nevertheless, implementation of PROM use in our trial may have compromised fidelity of the intervention. Did these adaptations wash out potential mechanisms critical for improvements to communication? If so, what are these core elements of PROM use that need to be in place to support communication? Attention to and assessment of intervention fidelity should be considered in all evaluations of PROMs implementation.

One potentially important mechanism we found was that PROM completion was, at times, perceived as a barrier to communication. All questionnaires are premised upon the assumption that a number can be assigned to experience. Knowledge translation studies pertaining to PROM use have previously identified the difficulty some may encounter in assigning a number (40); this finding is not unique to our study. Our analysis further revealed a knowledge gap about how to use a numeric PROM score. Whereas scores were intended to provide a “snapshot” of the patient’s point of view to inform a discussion and more fulsome clinical assessment, participants often suspiciously questioned if it was intended to be a replacement. Patients expressed fears, protectionism, and agency about what that number would represent in their care, who would see it, how it would be used, and to what ends, such as no longer being allowed to receive a transplant. Clinicians, perhaps, saw it as a shoddy replacement for their clinical expertise. PROMs are not intended to stand alone but to be used as one important piece in a puzzle or as a catalyst to support care, potentially enhancing communication. Such concerns have not been previously identified in the kidney literature on PROM use (8), so this is an important contribution. Echoes of these concerns are, however, seen in the reluctance of patients with kidney disease to report stigmatized symptoms, such as depression or sexual dysfunction (44,45). Conversely, nurses have acknowledged that numerical scores via PROMs can support patients on dialysis in expressing a topic otherwise hard to talk about with words (46).

PROM use was often perceived as having limited value. The self-reports of patients on dialysis through PROMs can be regarded as powerful because they give permission to bring patients’ voices and perspectives into a clinical discussion (8,14), expressing what might otherwise be missed. However, this benefit was not perceived by the majority of participants in our study. Then again, patients were not always completing PROMs on their own, and they were not sure of who saw their responses or what was done with their responses. Dialysis clinicians also had uncertainty about how to use PROM responses in conjunction with their own clinician assessment, and clinicians and patients on dialysis questioned the valued added, findings previously identified in dialysis settings (43,47). Together, these knowledge gaps eroded the perceived value of PROM use and hence led to the skepticism about whether they could enhance communication.

The strengths of our study are in the large sample size with longitudinal data collection and our mixed-methods approach for triangulation, but it also has weaknesses. As previously noted, the fidelity of the intervention may have been compromised given that implementation did not unfold as intended. Adaptation of PROM implementation in clinical settings is not unique to our study; nevertheless, this may have “washed out” potential mechanisms to enhance communication. Additionally, nurses were the primary clinician participants, and none were from urban centers. Engagement of other multidisciplinary members as well as participants from all EMPATHY trial sites may have provided alternative perspectives. The quantitative measure of communication we used may have been insensitive and unable to detect specific aspects of communications. Additionally, Communication Assessment Tool responses are subject to selection bias.

Improving patient-clinician and clinician-clinician communication was identified as a top challenge in in-center hemodialysis in Canada (48); thus, investigation into whether or how PROM use may bolster communication will be an important contribution. An unanswered question from our analysis is what core mechanisms are needed for PROM use to support patient-clinician communication within the context of hemodialysis. Broadly, our findings speak to validation processes about PROM use, the consequences of using PROMs, and how PROMs data inform clinical decision making (49). Future research is needed to examine how PROM use in dialysis may be supported if patients received training and what mechanisms are needed to enhance communication with clinicians. PROM programs may have additional purposes beyond informing clinical decision making, where data are aggregated for assessing and comparing quality of care at the meso or macro level (50).

The results of this study reveal that routine PROM use did not improve patient-clinician communication in in-center hemodialysis units in northern Alberta. Although this finding is counterculture, it aligns with other PROMs literature raising a similar question. This paper offers an important contribution to the literature because we found four nuanced reasons to explain why this might have occurred. Our findings highlight the need for research into validation processes to articulate what core mechanisms are needed for PROM use to enhance patient-clinician communication within the context of hemodialysis.

Disclosures

R. Buzinski reports research funding from the Can-SOLVE CKD Network and the EuroQol Research Foundation. S.N. Davison reports research funding from Alberta Health Services, Alberta Innovates, Alberta Kidney Care, and the Canadian Institutes of Health Research and honoraria from Abbott Nutrition (April 2022) for an Abbott Experts Circle talk. J.A. Johnson reports research funding from the EuroQol Research Foundation (copyright holder for EQ-5D instruments), is a member of the EuroQol Group, and is a member of the Board of Directors for the EuroQol Research Foundation. S. Klarenbach reports consultancy agreements with the Canadian Agency for Drugs and Technology in Health. S. Klarenbach serves as the Director of the Real World Evidence Consortium (composed of the Institute of Health Economics, the University of Alberta, and the University of Calgary) that receives grants from industry to conduct research and also received research funding from Bayer; both are outside the scope of the submitted work. K. Schick-Makaroff reports an advisory or leadership role for the Kidney Foundation of Canada (not paid), a consultant role for the Health Quality Council of Alberta (not paid), and other interests or relationships with the Kidney Foundation of Canada. H. Short reports research funding from the Can-SOLVE CKD Network and the EuroQol Research Foundation. M. Walsh reports employment with the Ontario Renal Network; research funding from the British Heart Foundation, the Canadian Institutes of Health Research, the Health Research Council, the National Health and Medical Research Council, the National Institutes of Health Research, and Vifor (no salary support received through any research funding); advisory or leadership roles for Bayer (steering committee, payment to institution) and Otsuka (national leader, payment to institution); and other interests or relationships with Novo Nordisk (event adjudication, payment to institution). The remaining author has nothing to disclose.

Funding

The EMPATHY trial was independently peer reviewed and funded by Can-SOLVE CKD Network (grant number: SCA-145103) supported by the Canadian Institutes of Health Research under Canada’s Strategy for Patient-Oriented Research. This large network supports many research projects, including the EMPATHY trial, and provides basic infrastructure and resources. EuroQol Research Foundation (grant number: 20180170) provided funding for the qualitative substudy through the University of Alberta.

Authors Contributions

R. Buzinski, S.N. Davison, J.A. Johnson, S. Klarenbach, K. Schick-Makaroff, and M. Walsh conceptualized the study; H. Short and L.A. Wozniak were responsible for data curation; K. Schick-Makaroff, H. Short, and L.A. Wozniak were responsible for formal analysis; J.A. Johnson, K. Schick-Makaroff, M. Walsh, and L.A. Wozniak were responsible for methodology; H. Short was responsible for project administration; S.N. Davison, S. Klarenbach, and M. Walsh were responsible for clinical resources; J.A. Johnson and M. Walsh were responsible for funding acquisition; J.A. Johnson provided supervision; K. Schick-Makaroff, H. Short, and L.A. Wozniak wrote the original draft; and R. Buzinski, S.N. Davison, J.A. Johnson, S. Klarenbach, and M. Walsh reviewed and edited the manuscript.

Data Sharing Statement

The data that support the findings of this study are available from the University of Alberta, but restrictions apply to the availability of these data, which were used under license for this study and so, are not publicly available. Data are, however, available from the corresponding author upon reasonable request and with permission of the University of Alberta.

Supplemental Material

This article contains supplemental material online at http://cjasn.asnjournals.org/lookup/suppl/doi:10.2215/CJN.05940522/-/DCSupplemental.

Supplemental Material.