Background

Chronic kidney disease (CKD) affects approximately 12% of adults in Canada with significant morbidity and mortality [1]. CKD patients often receive suboptimal care [2], partly because identifying patients at highest risk of progression is challenging [3]. Although models to predict risk of progression have existed for many years [4, 5], the recently validated 4-variable kidney failure risk equation (KFRE) [6, 7] allows for a more accurate assessment of risk and has been implemented in a variety of clinical settings [8,9,10,11]. However, evaluation of the KFRE’s use as a tool to guide triage and care delivery for CKD patients has been limited.

In the past 20 years, many centres have established multidisciplinary clinics to care for patients with advanced CKD in a team-based setting, however the evidence to support the effectiveness of this resource-intensive care has been limited [12]. The KFRE provides the opportunity to identify patients at higher risk of progression, and tailor their care accordingly. We previously reported a qualitative description of patients’, family members’, and providers’ perceptions of the KFRE [13], which informed our implementation strategy. This follow-up study provides a description of patients’ and providers’ experiences and perceptions 1 year following KFRE implementation, as well as assessments of patient care experience and provider job satisfaction before and 1 year after implementation. Our goals were to better understand the use of the KFRE in clinical decision-making, evaluate patient experience and provider job satisfaction pre/post implementation, and assess the risk-stratification implementation process from a quality improvement perspective.

Methods

This study is one phase of our multiphase study evaluating use of the KFRE to guide CKD care in Alberta, Canada [14]. We report a concurrent mixed methods design [15] to describe and compare patients’ and providers’ experience 1 year following KFRE implementation, as well as report on patient care experience and provider job satisfaction before and after implementation (Fig. 1). Ethics approval was granted by the University of Calgary Conjoint Health Research Ethics Board.

Fig. 1
figure 1

Kidney Failure Risk Equation (KFRE) Implementation multiphase study design. Blue box denotes focus of this paper

Full size image

Study setting

The CKD multidisciplinary clinic provides team-based care to patients with advanced CKD. The primary goals of the clinic are to slow disease progression, manage symptoms and comorbidities, and prepare patients for renal replacement therapy. Beginning in 2017, the KFRE was used as a tool to guide access to multidisciplinary CKD care for all patients residing in the southern regions of Alberta, Canada. Patients at higher risk of kidney failure (2-year KFRE ≥10% or eGFR ≤15 ml/min/1.73m2) were recommended to receive multidisciplinary care, including a nurse case manager, dietitian, pharmacist, and social worker, while lower-risk patients (2-year KFRE < 10%) were recommended to receive care from general nephrology (remaining under the care of their primary nephrologist with referral-based access to allied health professionals, but without a nurse case manager).

Our pre-implementation findings [13] revealed both patient and provider concerns relating to care experience and quality for lower risk patients who would be risk-stratified out of the multidisciplinary clinic, or considered ineligible, following KFRE implementation. Consequently, our study focuses on the perspectives and experiences of low-risk multidisciplinary clinic patients who were discharged from multidisciplinary care following KFRE implementation and explores satisfaction of these low-risk patients as well as satisfaction of higher-risk patients who remain in, or were referred to, the multidisciplinary clinic during the study timeframe.

We conducted the patient care experience and provider job satisfaction surveys pre- and post-implementation to broadly assess potential changes to experience or satisfaction following this change to clinic operations.

Post-implementation - qualitative component

Patient and provider selection and recruitment

English-speaking, adult (age ≥ 18 years) non-dialysis CKD patients who were deemed low risk and discharged from multidisciplinary CKD care to general nephrology care following KFRE implementation in the Southern Alberta Renal Program were eligible to participate in a semi-structured telephone interview. Clinic staff screened patients in Calgary, Alberta, for eligibility and obtained consent for contact by a research coordinator. All multidisciplinary CKD healthcare providers (nephrologists, nurses, and allied health professionals) in Calgary, Alberta, were invited to participate by e-mail.

Data collection

Two investigators (M.D.S and H.T.) with qualitative research experience conducted all interviews. We used open-ended questions to explore benefits and challenges to KFRE implementation as well as patient and provider experiences transitioning from the CKD multidisciplinary clinic to general nephrology practice (Additional files 1 and 2). Patient and provider interviews were approximately 20 min. All interviews were audio-recorded and transcribed verbatim. Data collection was stopped once we reached data saturation, when no new additional concepts informing the research objectives were identified [16].

Four open-ended questions related to the benefits and challenges of KFRE implementation were completed by healthcare providers as part of the on-line survey, described below, and incorporated in the qualitative analysis (Additional file 3).

Qualitative data analysis

We summarized patients’, family members’ and providers’ perspectives from interviews and open-ended responses from provider surveys. Text data was imported into NVivo Version 12 software to facilitate data analysis (QSR International Ltd., Doncaster, Australia). Based on findings from the pre-implementation qualitative study [13], we applied a directed content analysis [17] approach to focus the interview guides, identify relationships between the pre- and post-implementation code definitions and to qualitatively describe [18] participant perspectives; we did not apply a theory to guide analysis. We also asked open-ended questions to identify new content categories and extend our conceptual framework relating to the benefits and challenges of a risk-based approach to multidisciplinary CKD care [17].

Three investigators (M.D.S, H. T, and J.F) independently reviewed the transcripts and developed a coding scheme guided by relevant findings from the pre-implementation qualitative study. The investigators went through several iterations of transcript review and coding scheme refinement to ensure the full range of data was considered (investigator triangulation [19, 20]) and until consensus was reached.

Pre and post-implementation - quantitative component

Patient and provider selection and data collection

All patients attending the CKD multidisciplinary clinics in Calgary as well as low-risk patients discharged from multidisciplinary care to general nephrology care following implementation of the KFRE were eligible to complete an anonymous, paper-based care experience survey (the Care Experience Feedback Improvement Tool [CEFIT]) [21] . Clinic staff provided the survey to eligible patients for completion during clinic visits. The surveys were distributed from November 2016 through January 2017 (pre-implementation) and January 2018 through March 2018 (post-implementation).

All multidisciplinary CKD healthcare providers (nephrologists, nurses, and allied health professionals) from the multidisciplinary clinics in Calgary were invited by email to participate in an online anonymous job satisfaction survey (the Andrews and Withey Job Satisfaction questionnaire [22]). Survey responses were collected online from November 2016 through January 2017 (pre-implementation) and in April 2018 (post-implementation) (OutSideSoft Solutions, Inc., Quebec, Canada).

The patient and provider surveys were collected anonymously and therefore it was not possible to link pre- and post-implementation survey responses.

Quantitative data analysis

Nonparametric statistics were used for the Likert-type survey data [23]; pre- and post-implementation differences in survey responses were compared using the Wilcoxon Mann-Whitney rank sum test for unmatched data [24, 25]. Analyses were performed using Stata version 14 (StataCorp, College Station, TX).

Results

Qualitative component

Patient and provider characteristics

Of the 23 eligible patients approached by clinic nurses to obtain consent for contact by the research team, 13 consented to be contacted and 10 refused. Of the 13 patients contacted by the research team, nine provided consent and participated in a telephone interview; one family member self-referred and participated in a telephone interview (Table 1). Participants were all low-risk (2-year KFRE < 10%) patients discharged from the multidisciplinary clinic following KFRE implementation and followed by their nephrologist only. The majority of participants were ≥ 65 years of age and had previously spent > 5 years as patients in the multidisciplinary clinic. Males and females participated equally and the majority reported their health status as ‘good’ or ‘very good’.

Table 1 Demographic characteristics of patient participants (pre- and post-implementation surveys and post-implementation interviews)
Full size table

Of the 75 healthcare providers eligible for an interview, 17 responded to the email invitation and participated in an interview (Table 2). The majority of participants were female and had worked in their profession for > 10 years with > 5 years experience working at the multidisciplinary clinic. Of the 33 healthcare providers responding to the online survey, 27 completed the open-ended questions included in the qualitative analysis.

Table 2 Demographic characteristics of healthcare provider participants (pre- and post-implementation surveys and post-implementation interviews)
Full size table

Content categories

Based on the patient and provider interviews we categorized participant perspectives into seven content categories: targeted care; access to resources outside the multidisciplinary clinics; self-efficacy; patient reassurance and reduced stress; transition process for low-risk patients; anticipated concerns (providers only); and job satisfaction (providers only). Within these categories we describe and compare patients’ and providers’ experiences and perceptions of barriers and facilitators following KFRE implementation. Illustrative quotes are provided in Table 3; an expanded table of quotes and a summary of pre- and post-implementation category relationships and definitions are available in the supplementary material (Additional files 4 and 5).

Table 3 Selected illustrative quotations
Full size table

Content categories identified by patients and providers

Targeted care : this category includes reference to personalized medicine, targeting care based on individual needs, ensuring patients are seen in the most appropriate setting, and sustainability of risk-based model of care.

Both patients and providers reported that using the KFRE to target care to patients most likely to progress was a key strength. Providers indicated that the KFRE was a ‘useful tool’ to help utilize clinic resources optimally, limit ‘inappropriate referrals’, and ensure maximum benefit to patients in need of targeted intervention. Patients discharged from multidisciplinary care following KFRE implementation echoed this sentiment, indicating that as long as they are ‘stable’, ‘there’s no point’ going to the clinic more often, or accessing more resources. Some providers stated they changed their referral practice as a result of the KFRE implementation, suggesting they may have more confidence in their ability to identify high-risk patients.

Access to resources outside of the multidisciplinary clinic: this category includes examples of resources that may or may not exist outside of the multidisciplinary clinics and patient/provider awareness, need (or lack of need) and comfort accessing those resources.

Most patients reported they had adequate access to reliable resources outside of the clinic to help manage their CKD including primary care providers, nurse practitioners, dietitians, and pharmacists. Moreover, many patients reported that they did not ‘really have to access’ additional services because their disease was ‘stable’. However, some nurses indicated that patients discharged from the CKD clinic were still phoning them, primarily to inquire about blood work results, suggesting that ‘patients are calling us back because they don’t have the same support anymore’. Although patients reported that they ‘did rely on their nurse case managers’ in the past, they did not perceive a substantial impact to their care following discharge from the multidisciplinary clinic, suggesting they are able to receive the care they need.

A few providers expressed concern about access to preventive education and adequate monitoring outside of the multidisciplinary clinic; one provider in particular indicated they were reluctant to discharge low-risk patients because they were not ‘confident’ that the ‘right amount of supervision and care’ would be provided. However, several providers reported there were many resources in the community, such as primary care networks and nutrition classes/resources. Some providers suggested that the perception of inadequate monitoring in the community might be due to a lack of ‘respect’ toward primary care in general.

Self-efficacy: this category includes patient and provider comments on, or examples of, self-management and health literacy following implementation of risk-based care.

Many of the low-risk patients demonstrated good self-management and self-efficacy, reporting that they ‘keep track of’ their diet and lab information and know how to manage and help slow progression of their CKD. However, these patients also reported that one aspect of care they missed was timely access to their laboratory test results, commenting that the ability to access those values on their own would be ‘marvellous’ so they ‘don’t have to wait to see’ their nephrologist and can ‘do things before’ their appointment to manage their CKD. Similarly, nurses also indicated that access to laboratory test results was one of the primary reasons patients called them and that an online patient portal to access bloodwork would be helpful because most patients ‘don’t ask us about anything [when they call]’, ‘they just want to know their numbers’.

Patient reassurance and reduced stress : this category includes references to reassurance gained from knowing the KFR value and patient or provider comments on patient emotional well-being following implementation of the risk-based approach.

Most patients indicated that knowing their KFRE value was very ‘reassuring’. Providers also articulated the relief experienced by patients, particularly when the patient’s risk of kidney failure was low, and indicated that the KFRE was a ‘great educational tool’ to communicate individual risk. Additionally, both patients and providers reported that some patients were quite relieved to have fewer appointments when transitioned back to general nephrology or primary care, particularly when their CKD was stable, and that they were ‘not unhappy about not going’. Several providers echoed this sentiment, suggesting many patients ‘are really happy to go back to their GP [general practitioner], it’s just one less appointment … they feel like they are getting better somehow and they are happy with it’.

Transition process for low-risk patients: this category includes comments related to the transition process, challenges experienced, and reluctance to discharge lower-risk patients.

Providers identified several challenges related to the implementation, particularly relating to discharge of prevalent multidisciplinary clinic patients to general nephrology. Most issues were administrative, related to relocation of patient medical charts, updating medication renewals, changing the bloodwork requisition and frequency, and general communication specific to this transition. A few patients indicated they would have appreciated ‘more information or explanation’ to process the transition. Other patients, although recognizing they were stable and may not require multidisciplinary care, still expressed disappointment about not having access to their nurse anymore, commenting that ‘more access to a nurse would probably be a little more helpful … even though … nothing has really changed in three years’.

Content categories identified by providers only

Anticipated concerns : this category includes perceived concerns about the KFRE’s predictive ability and potential, perceived long-term risks for low-risk patients due to lost access to a nurse case manager and multidisciplinary resources.

Some providers were concerned about long-term outcomes for low-risk CKD patients who will not receive ‘education and supervision’ from the multidisciplinary team to keep them ‘focused’ on slowing progression. These concerns were amplified by a perception that a low KFRE value might provide a false sense of security for patients and their community healthcare providers. A few providers also expressed concern about those patients who might ‘tank right away’, and whether the KFRE would identify a rapidly progressing CKD patient in time. Patients did not express specific concerns about future progression of their kidney disease.

Job satisfaction: this category includes references to workload changes and overall job satisfaction following implementation of the risk-based approach.

In general, nephrologists reported that although the KFRE created additional ‘workload’ for them, they acknowledged that it was ‘the right thing to do’. In contrast, nurses and other staff indicated that the KFRE had a slight positive impact, or no noticeable impact, on their workload and job satisfaction. Many nurses reported they were grateful for ‘more time to focus on high-risk patients’; however, some nurses felt that as a result of the KFRE their caseload was more acute and, since patients who were discharged required minimal monitoring, their workload was relatively unchanged. Nurses’ workload, and possibly job satisfaction, may have been impacted by a number of concurrent staffing issues that arose during the post-implementation study period; several staff went on leave (medical and maternity) and as a result, caseloads were re-distributed.

Quantitative component

Patient care experience

Overall, 176 patients completed the care experience survey in the pre-implementation phase and 237 in the post-implementation phase. Most participants in both phases were male, married, > 65 years of age, and reported their health status as “Fair” or “Good” (Table 1).

The majority of patients were satisfied with their care in both the pre- and post-implementation periods, with no significant differences in any of the care experience domains when analyzing all five categories (Fig. 2a). When considering the responses “Always” and “Often” together versus not (Fig. 2b), there were slight, but statistically significant, improvements in three of five domains in the post-implementation period: access to care (p = 0.01), caring staff (p = 0.02), and safety of care (p = 0.03).

Fig. 2
figure 2

a Chronic Kidney Disease (CKD) multidisciplinary clinic patient experience before and after KFRE implementation b Proportion of respondents who selected “Always” or “Often”

Full size image

Provider job satisfaction

Of the 75 healthcare providers, 40 (53%) completed the workplace satisfaction survey in the pre-implementation phase and 33 (44%) in the post-implementation phase. The majority of respondents were female with > 10 years experience in their profession and > 5 years working at the CKD multidisciplinary clinic (Table 2).

There were no significant differences in workplace satisfaction items pre- and post-implementation of the KFRE, with the majority of participants reporting they were satisfied (Fig. 3a). When considering “Very” and “Mostly” satisfied responses together (versus mixed or unsatisfied responses) for each scale item there were no significant differences (Fig. 3b).

Fig. 3
figure 3

a Chronic Kidney Disease (CKD) multidisciplinary clinic healthcare provider satisfaction before and after KFRE implementation b Proportion of respondents who selected “Very” or “Mostly”

Full size image

Discussion

In this mixed methods study, patients and healthcare providers described benefits and challenges associated with using the KFRE to stratify CKD patients to care providers based on their risk of kidney failure. Overall, patients and providers felt that a KFRE-based approach improved sustainability and quality of the multidisciplinary clinics by targeting care to patients at highest risk; however, some providers expressed concern that care for low risk CKD patients may be of lower quality, and that access to care may be inadequate. There were no significant differences in job satisfaction for healthcare providers, whereas patient results suggested slight improvement in care experience following KFRE implementation.

Similar to our pre-implementation study [13], the primary concern identified by healthcare providers regarding use of the KFRE to risk stratify patients for multidisciplinary care was potential for patients with lower risk of kidney failure experiencing more rapid progression over time without multidisciplinary care. These provider concerns may stem from poor perceptions of patient self-efficacy and primary care in general and is consistent with other studies [26,27,28,29,30,31]. Although challenges related to identification and management of CKD patients in primary care have been reported [2], subsequent strategies to build CKD awareness, education, and access to specialist advice, such as an online CKD clinical pathway [32,33,34] and electronic access to specialist advice [35], have been implemented to build healthcare provider capacity to care for CKD patients in the community.

In contrast, low-risk patients who were transitioned out of the multidisciplinary care clinics to general nephrology care reported they had adequate access to, and confidence in, their community-based primary care physicians, dietitians and pharmacists, and did not perceive a substantial impact to their care. These patients described a number of strategies they used to manage their CKD, such as diet and clinical lab results tracking. The threshold we used to consider multidisciplinary care was relatively low (≥ 10% risk of kidney failure in 2 years), and the majority of patients discharged from the clinic continue to be managed in general nephrology. Our ongoing prospective evaluation using our provincial data sources [36] will address this concern by assessing the risk of kidney failure for patients not followed in the multidisciplinary care clinic. Additionally, a self-management eHealth tool [37], co-developed with patients and providers, and a provincial patient portal [38] to access individual medical information online, are both in development. These tools will complement existing resources, support patients, and encourage self-efficacy.

A number of providers remained concerned regarding the KFRE’s accuracy and consequent long-term patient outcomes. Although some studies suggest multidisciplinary CKD care leads to improved patient outcomes, decreased hospitalizations, and reduced healthcare costs in pre-dialysis CKD patients [39,40,41,42], a recent systematic review [12] concluded that most published studies demonstrate a high risk of bias. While there was improvement in blood pressure control and prescribing practices in nurse-led and pharmacist-led care models, they found limited evidence of a significant effect on most outcomes [12]. Moreover, previous research indicates that CKD trajectories in the 2 years preceding dialysis are substantially heterogeneous [43] and the cost-effectiveness of multidisciplinary care may be greatest in younger patients and in those with higher urine albumin levels [44]. As a result, further targeting multidisciplinary clinic services based on individual characteristics has been suggested to optimize patient outcomes and resource use [45]. Assessment of long-term outcomes (hospitalizations, death, modality choice, kidney transplantation, physician visits, and process-based indicators of appropriate CKD care) using provincial administrative data is the focus of the next phase of this study [14].

Strengths of this study were inclusion of both patient and provider perspectives, and the integration of qualitative and quantitative components. However, there are limitations that should be recognized when interpreting the results. We acknowledge that the qualitative component of this study included a larger number of healthcare providers than patients; however, data saturation was achieved in both groups despite differences in sample size. Most of the patients who participated in the interviews had been exposed to multidisciplinary care for > 5 years prior to discharge and may be inherently more proactive, which may have impacted our findings relating to self-efficacy and may limit transferability of these findings to patients with lesser multidisciplinary clinic experience. Moreover, interviewed patients had been discharged from multidisciplinary care within the previous 12 months, following implementation of the risk-based approach, and therefore their described experiences and perceptions are reflective of this limited period. The pre- and post-implementation surveys were anonymous, so it was not possible to determine if participants completed both surveys to pair survey responses, or to determine the patient survey response rates. Finally, these results are based on a single Canadian centre, which may limit generalizability to other settings.

Conclusion

In summary, evaluation of risk-based models to guide triage and care delivery for CKD patients has been limited. Our findings suggest patients and healthcare providers perceive that a KFRE risk-based approach may improve sustainability and focus of team-based multidisciplinary CKD clinics by targeting patients at highest risk, with survey results suggesting no difference in healthcare provider job satisfaction and potential slight improvement in patient care experience. However, some providers remain concerned regarding the KFRE’s accuracy and consequent long-term patient outcomes for lower-risk patients risk-stratified to receive care in general nephrology only. This study is one component of a multiphase study evaluating use of a KFRE-based approach to guide CKD care. In the final phase, provincial administrative data will be used to prospectively evaluate the effectiveness of this risk-based approach on important patient care and clinical outcomes.