Impact of educational interventions on clinician knowledge, adherence, and satisfaction utilizing a CKD care pathway in primary care

Abstract

Background

Primary care clinicians are central to early chronic kidney disease (CKD) detection and management, yet often lack structured tools to support guideline-based care. CKD is common, underdiagnosed, and frequently undertreated in primary care, despite guideline-recommended screening and disease-modifying therapies. Structured CKD care pathways may improve clinician knowledge, confidence, and adherence to evidence-based management, yet real-world implementation data remain limited.

Methods

A prospective, multi-site pilot study was conducted at 2 health systems implementing a shared CKD Care Pathway with a 3-month educational intervention. Clinicians at the University of Pittsburgh Medical Center (UPMC) Matilda H. Theiss Primary Care Practice and Geisinger Community Medicine Service Line (CMSL) participated in the study. Standardized educational content was delivered at both sites, with implementation tailored to local workflows. Pre- and post-intervention surveys assessed clinician knowledge, confidence, perceived barriers, and satisfaction. Patient-level outcomes and study populations differed by site. At UPMC, electronic medical record (EMR) data were analyzed to assess CKD screening and diagnosis among adults with diabetes, whereas at Geisinger, EMR dashboard data were used to descriptively evaluate sodium-glucose cotransporter-2 inhibitor (SGLT2i) prescribing as a proxy for uptake of guideline-directed therapy in patients at risk of or with CKD.

Results

Clinician confidence improved across multiple CKD identification and management domains. At UPMC, the proportion of patients receiving urine albumin-to-creatinine ratio (uACR) screening increased from 49% pre-intervention to 60% post-intervention at UPMC. At Geisinger, the percentage of patients with diabetes and an active SGLT2i prescription increased descriptively from approximately 17% pre-intervention to approximately 19% during and following the intervention.

Conclusions

A standardized yet adaptable CKD Care Pathway was feasible to implement in primary care and was associated with improved clinician confidence and selected care processes.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12882-026-05141-3.

Introduction

Chronic kidney disease (CKD), defined as abnormalities in kidney structure or function (glomerular filtration rate [GFR] < 60 mL/min/1.73 m² [2] or urine albumin-to-creatinine ratio [uACR] ≥ 30 mg/g) for at least 3 months, represents an underdiagnosed public health concern [1, 2]. The Centers for Disease Control and Prevention estimates that 1 in 7 United States adults have stage 1 to 4 CKD, with 90% unaware of their condition due to the asymptomatic nature of early stages [3, 4]. By 2040, CKD is expected to become the fifth leading cause of years of life lost worldwide [5].

Despite recognition of the burden of CKD, there is controversy and a lack of consensus on the value and frequency of screening for at-risk individuals [1]. Disease and guideline-based education within health systems is limited, with resources primarily focused on advanced stages rather than early detection [6]. Limited healthcare provider (HCP) knowledge in primary care further hinders CKD diagnosis and monitoring, including underutilization of screening tests such as serum creatinine and uACR [7–10]. Additional barriers to CKD screening include low awareness among HCPs and patients about CKD risk factors and complications, minimal primary care involvement in CKD management, inappropriate nephrology referrals based on GFR targets, and underutilized electronic medical record (EMR) tools [11–14]. The reliance on estimated glomerular filtration rate (eGFR) for the detection of CKD is also limiting, as it identifies the disease but does not prompt necessary follow-up care, leaving many patients without further intervention. A significant proportion of patients with abnormal eGFR results do not receive timely follow-up, highlighting the need for improved systems to ensure consistent and proactive CKD management [15]. Importantly, screening recommendations are strongest for high-risk populations, including individuals with diabetes and hypertension, where early detection and intervention have established clinical benefit.

A systematic review has suggested proactive CKD screening can be cost-effective for individuals with diabetes and hypertension (HTN), the 2 most common causes of CKD worldwide [10]. However, clinical trials have not yet confirmed whether interventions to detect, risk stratify, and treat CKD improve health outcomes in these populations [1]. A recent cost-effectiveness analysis of population-wide screening, incorporating evidence-based treatments like sodium-glucose cotransporter-2 inhibitors (SGLT2i), concluded that screening for albuminuria in adults could be cost-effective in the United States [15]. This aligns with the Kidney Disease: Improving Global Outcomes (KDIGO) Controversies Conference on Early Detection and Intervention in CKD, which suggested that early identification of CKD in asymptomatic at-risk individuals would likely benefit primary care and community settings, especially when coupled with risk stratification and treatment [1].

The CKD Leaders Network, established in 2020, is a multidisciplinary network of health-system leaders in CKD focused on defining and promoting a best-practice model for population health–driven management [16]. In April 2023, the Network hosted a roundtable discussion to address challenges in CKD screening, diagnosis, risk stratification, and management [16]. Key barriers identified included challenges in diagnosis, stratification, management, and monitoring of patients with CKD, as well as gaps requiring immediate attention [16]. Experts also highlighted innovative care models and best practices for CKD care delivery [16].

The CKD Care Pathway Within the Primary Care Setting was developed to integrate evidence-based treatments with best practices to identify high-risk patients, encourage appropriate screening and medication use, and ensure appropriate referrals to nephrology [16]. To bridge the gap between clinical guidelines and real-world implementation, the CKD Care Pathway translates the 2024 KDIGO Clinical Practice Guidelines for the Evaluation and Management of CKD into standardized processes for patient care, ensuring that management of CKD within the primary care setting reflects the most current clinical practices [1, 16]. This study evaluates implementation of the CKD Care Pathway at 2 health systems and assesses its impact on clinician knowledge, confidence, satisfaction, and selected patient-level outcomes.

Methods

Study design and settings

The CKD Care Pathway (Fig. 1) was piloted by 2 institutions: University of Pittsburgh Medical Center (UPMC) Matilda H. Theiss Primary Care Practice and Geisinger Community Medicine Service Line (CMSL). UPMC piloted the CKD Care Pathway under the name Increasing CKD Screening in Primary Care for Diabetics (INSPIRED) to clearly communicate the intervention’s emphasis on improving CKD screening among patients with diabetes, the primary population targeted at this site. In contrast, Geisinger retained the CKD Care Pathway nomenclature, reflecting its broader implementation across patients with CKD, diabetes, and hypertension. The UPMC site was selected for its large practice size, inner-city location, and low pre-intervention uACR screening rate among patients with diabetes mellitus (DM). Geisinger piloted the CKD Care Pathway across its service line. Geisinger was selected due to its integrated health system structure, established EMR infrastructure, and existing performance dashboards that enabled evaluation of medication prescribing trends at a system level. Both sites represent primary care settings within larger health systems, with interventions focused on frontline outpatient care.

Fig. 1.

CKD Care pathway and interventional plan overview of the CKD Care Pathway as outlined by the CKD Leaders Network and interventions assessed by Geisinger, and as part of the INSPIRED study by UPMC. ACEi, angiotensin-converting enzyme inhibitor; ARB, angiotensin receptor blocker; CKD, chronic kidney disease; eGFR, estimated glomerular filtration rate; EMR, electronic medical record; GLP-1 RA, glucagon-like peptide 1 receptor agonist; ICD-10, International Classification of Diseases version 10; INSPIRED, Increasing CKD Screening in Primary Care for Diabetics; BMP, basic metabolic panel; SGLT2i, sodium-glucose cotransoporter-2 inhibitor; uACR, urine albumin-creatinine ratio; UPMC, University of Pittsburgh Medical Center

Although site-specific nomenclature and implementation strategies differed (INSPIRED at UPMC and CKD Care Pathway at Geisinger), both institutions piloted the same underlying CKD Care Pathway framework as part of a single, coordinated pilot. Outcome selection differed by site based on availability of data infrastructure and local implementation priorities. UPMC focused on screening and diagnostic outcomes using patient-level EMR data, whereas Geisinger evaluated prescribing trends using existing system-level dashboards. Although prescribing trends were evaluated among patients with diabetes based on available dashboard data, the intervention targeted broader guideline-defined populations at risk for CKD, including individuals with CKD, diabetes, and hypertension. Pre- and post-surveys assessed clinicians’ knowledge, confidence, adherence, and satisfaction with the Pathway. The survey instruments were developed for this study and are provided in Supplementary Material 1. Although study populations differed between institutions, patient-level outcomes were assessed among individuals with diabetes in both settings to ensure comparability of high-risk populations.

Study design at UPMC (INSPIRED)

INSPIRED was a single-arm prospective pilot conducted over three months (July–October 2024) focusing on adults aged 18–85 years with DM, selecting this population because patients with diabetes are at high risk for developing CKD. Current national guidelines recommend routine screening for CKD with annual assessment of eGFR and urine albuminuria testing [17]. EMR data were analyzed to assess CKD screening, diagnosis, and nephrology referral before and after intervention.

Study design at Geisinger CMSL

Geisinger implemented a department-wide educational intervention from September to December 2024 focused on guideline-recommended management of patients with CKD, diabetes, and hypertension, including SGLT2i use for eligible patients. The intervention aimed to improve clinicians’ confidence in prescribing guideline-supported medications, with the pathway recommending the use of SGLT2i for patients with an eGFR ≥ 20 mL/min/1.73 m² [2]. Clinicians prescribing trends were evaluated using data from an existing EMR dashboard.

Institutional review board (IRB)

IRB approval forms were submitted at both institutions. Protocols were reviewed and qualified for exemption under United States Department of Health and Human Services regulations [(45 CFR 46.104)].

Pre-intervention surveys

The survey instrument was developed for this study based on key domains identified in CKD guideline implementation, site specific focus areas and prior educational interventions. Pre-intervention surveys were anonymous at both institutions and assessed similar domains, including clinicians’ knowledge of CKD, screening for high-risk populations, utilization of SGLT2i, and referral processes. While survey administration differed by site, the content and constructs assessed were comparable across institutions (see Supplementary Material for survey instruments). For UPMC, surveys were administered on paper prior to the launch meeting and entered into the REDCap system. At Geisinger, pre-surveys were administered during a mandatory CMSL quarterly meeting. Surveys were distributed electronically, with a request for completion before the meeting start. Only responses submitted prior to the meeting were included in the analysis. Clinician self-reported confidence was selected as a primary outcome to assess perceived readiness to apply guideline-based care in clinical practice. Although objective measures such as clinical vignettes may provide additional insight, such approaches were not a focus within the scope of this pilot implementation study.

Pre-survey instruments included a multi-item, close-ended questionnaire designed to gather information on clinician confidence in managing CKD, screening for high-risk populations, utilization of SGLT2i for appropriate patients, referral processes, and barriers to patient management. Most items used 5-point rating scales to assess knowledge and understanding about the processes. Response scale formats varied based on the nature of the constructs assessed, with some items measuring frequency (e.g., “never” to “always”) and others assessing agreement or perceived impact, which aligned better with alternative scale formats. For the role of SGLT2i in CKD management, scores for providers who felt “confident enough to practice in this area without support” and those who were “fully confident and felt they could teach others” were combined.

Post-intervention surveys

Post-surveys assessed the impact of the interventional materials on adherence to and satisfaction with the CKD Care Pathway. The pre- and post-intervention surveys contained the same core provider knowledge and confidence items to allow for direct comparison over time. The post-intervention survey additionally included items assessing satisfaction with the educational materials and perceived impact of the intervention, which were not included in the pre-intervention survey. At both institutions, surveys were conducted 3 months after the start of intervention.

Intervention

At both institutions, the educational intervention consisted of a standardized core curriculum based on the CKD Care Pathway, including guideline-recommended approaches to CKD screening, diagnosis, risk stratification, pharmacologic management, and nephrology referral. The core content and learning objectives were consistent across sites and delivered through a 1-hour educational session, followed by a series of monthly educational emails and publicly available webinars referencing national guidelines and professional society resources.

Implementation strategies were adapted to local workflows and infrastructure. At UPMC, the intervention incorporated EMR-based workflow facilitation, including health maintenance flags and pre-pending of screening laboratories by ancillary staff, as well as patient-facing educational materials. At Geisinger, implementation focused on clinician education and medication optimization without EMR workflow modification, reflecting site-specific clinical priorities and workflow adaptation.

At both institutions, implementation included structured delivery of the educational session (live or virtual), followed by reinforcement through monthly email communications and publicly available webinars. Attendance was tracked via sign-in logs or electronic participation records where available. Educational sessions were delivered by multidisciplinary faculty supporting implementation of the pilot, including nephrology and primary care clinicians.

EMR analysis

For the INSPIRED study, data were extracted before (January 2023 to July 2024) and during (July to October 2024) the interventional periods. The following data were tracked for analysis: patient demographics, vitals, encounter diagnoses, medication orders, and outpatient/inpatient laboratory orders and results. Eligible patients were aged 18 to 85 years with documented type 1 or type 2 DM and had at least 1 primary care visit during both the intervention and pre-intervention periods. For each patient, the baseline was defined as the date of pre-intervention, and characteristics were assessed in a window spanning from 12 months prior to 1 month after the index date. Outcomes of interest included CKD screening (e.g., eGFR, uACR testing), CKD diagnosis (International Classification of Diseases, Tenth Revision [ICD-10] codes), and nephrology referral. Changes in outcome rates between the pre-intervention and during/post-intervention periods were analyzed using Generalized Estimating Equations with robust standard errors, clustering by patient. An autoregressive working correlation structure was specified to account for within-subject correlation across time points. Binary outcomes were modeled using a logit link and binomial distribution. All analyses were conducted using R version 4.4.2, and statistical significance was defined as P < 0.05.

At Geisinger, investigators extracted data from a preexisting EMR dashboard that tracked CKD screening and SGLT2i trends. The average percentage of patients with DM and SGLT2i in their active medication list were compared from pre-intervention (June to August 2024) to the 3 months during (September to December 2024) and following (January to March 2025) intervention.

Results

Clinician survey respondents

Survey respondents represented a multidisciplinary sample of clinicians and care team members involved in CKD management, including physicians, advanced practice providers (APPs), nurses, medical assistants (MAs), pharmacists, and administrative staff at both sites. At UPMC, participants included physicians, an APP, MAs, nurses, office staff, and a pharmacist (pre-intervention, N = 13; post-intervention, N = 15), reflecting varied roles (Table 1). Among Geisinger participants at the CMSL meeting, 11 were physicians, 5 were APPs, and 1 was a nurse (pre-intervention, N = 17; post-intervention, N = 18). The increase in post-intervention respondents at both institutions reflects additional staff participation during the pilot intervention (Table 1).

Table 1.

Survey respondent demographics

Respondents N (%)		INSPIRED - UPMC	Geisinger CMSL
	Pre-intervention (N = 13)	Post-intervention* (N = 15)	Pre-intervention (N = 17)	Post-intervention* (N = 18)
Physician	2 (15%)	3 (20%)	11 (65%)	13 (72%)
APP	1 (8%)	1 (7%)	5 (29%)	5 (28%)
MA	2 (23%)	3 (20%)	0 (0%)	0 (0%)
Nurse	3 (15%)	3 (20%)	1 (6%)	0 (0%)
Pharmacist	1 (8%)	1 (7%)	0 (0%)	0 (0%)
Other†	4 (31%)	4 (27%)	0 (0%)	0 (0%)

*The number of post-intervention survey respondents was higher than pre-intervention due to additional staff participating during the course of the pilot intervention

†At UPMC, “Other” included front desk staff, dietitian, office manager, and 1 unknown

APP, Advanced Practice Provider; CMSL, Community Medicine Service Line; MA, medical assistant; UPMC, University of Pittsburgh Medical Center

Impact of the educational intervention on clinician confidence in CKD identification and management

Across both institutions, clinician confidence in CKD identification and management domains shifted toward higher confidence categories following the educational intervention, as assessed by pre- and post-intervention surveys. These clinician-level outcomes were evaluated at both UPMC and Geisinger and reflect self-reported confidence, rather than objective measures of clinical performance.

At UPMC, post-intervention responses demonstrated a redistribution across confidence categories toward “confident without support” and “fully confident” across multiple domains, including knowledge of CKD diagnostic criteria, use of eGFR, understanding of the importance of uACR testing, awareness of CKD as a cardiovascular risk multiplier, and familiarity with CKD-related complications (Fig. 2a). For example, the percentage of respondents reporting “fully confident” in interpreting uACR increased from 10% pre-intervention to 13% post-intervention, while the proportion reporting “confident without support” increased from 20% to 33%. Despite these shifts, a substantive proportion of respondents continued to select “need to know more” or “confident with support,” indicating persistent knowledge gaps.

Similar redistribution toward higher confidence categories was observed among clinicians at Geisinger across the same CKD identification domains (Fig. 2b). Post-intervention responses showed increases in the percentages of respondents selecting “confident without support” and “fully confident” for diagnostic criteria, eGFR-based identification, and interpretation of uACR. However, moderate confidence categories remained common, particularly for more complex diagnostic and prognostic concepts.

Across both sites, clinicians were also surveyed regarding confidence in predicting CKD severity and prognosis and selecting appropriate management strategies. Post-intervention responses demonstrated increases in the proportion of respondents reporting higher confidence for interpreting uACR and treatment selection at both UPMC (Fig. 2c) and Geisinger (Fig. 2d). In contrast, confidence related to predicting CKD prognosis and recognizing signs of advanced disease remained largely distributed across moderate confidence categories, highlighting continued areas for educational emphasis.

At Geisinger only, clinicians were additionally surveyed regarding comfort with the role of SGLT2i in CKD management. Pre-intervention, most respondents reported moderate or partial confidence, with 35.3% selecting “confident to practice without support.” Post-intervention, the proportion selecting this category increased to 66.7%, while no respondents selected “not confident” (Fig. 2e). Self-reported frequency of SGLT2i use also shifted following the intervention, with an increase in respondents reporting “frequent” use from 23.5% pre-intervention to 55.6% post-intervention, accompanied by decreases in “rarely” and “never” use categories (Fig. 2f).

Impact of the INSPIRED pathway on CKD screening, diagnosis, and referral (UPMC only)

Screening and diagnostic outcomes were evaluated only at UPMC because patient-level longitudinal EMR data were available for analysis at this site. The INSPIRED cohort consisted of 110 adults with diabetes receiving care in an urban primary care setting. Patients were predominantly Black and female, with a mean age of 58.9 years, and had a high area deprivation index, consistent with an inner-city population (Table 2). At baseline, 78% of patients were classified as CKD stages 1 or 2, and uACR values were missing for a substantial proportion of patients.

Table 2.

Patient baseline characteristics pre-intervention in the INSPIRED study

Variable	All patients (N = 110) Mean (SD) or N (%)
Age (years)	58.9 (14.0)
Gender Female	79 (72%)
Race
Black	84 (76%)
White	22 (20%)
Others/unreported	4 (3.6%)
Ethnicity
Non-Hispanic	102 (93%)
Hispanic	2 (1.8%)
Unreported	6 (5.5%)
Area Deprivation Index*	79.5 (19.7)
BMI†	33.6 (7.9)
BP Systolic‡	135.4 (15.3)
BP Diastolic‡	80.7 (7.0)
Cr (mg/dL)*	1.2 (0.9)
eGFR*	75.8 (26.1)
uACR (mg/g)¶,#	10.0 (5.9, 37.0)
HbA1c||	7.2 (1.6)
KFRE-2¶,#	0.01 (0.00, 0.04)
KFRE-5¶,#	0.04 (0.01, 0.12)
Type 2 DM	94 (85%)
HTN	82 (75%)
CVD	54 (49%)
Missing albuminuria data	57 (52%)
Albuminuria Stages¶
A1	39 (74%)
A2	10 (19%)
A3	4 (7.5%)
CKD Stages*
1	31 (30%)
2	49 (48%)
3a	8 (7.8%)
3b	11 (11%)
4	2 (1.9%)
5	2 (1.9%)

*ADI, Cr, eGFR, and CKD stages have 7 missing values

^†BMI has 1 missing value

^‡BP Systolic/Diastolic has 1 missing value

^||HbA1c has 18 missing values

^¶uACR, KFRE-2, KFRE-5, and albuminuria stages have 57 missing values

^#Listed as median (first quantile, third quantile)

ADI, area deprivation index; BMI, body mass index; BP, blood pressure; CKD, chronic kidney disease; Cr, creatinine; DM, diabetes mellitus; eGFR, estimated glomerular filtration rate; HbA1c, hemoglobin A1c; HTN, hypertension; KFRE-2, kidney failure risk equation; INSPIRED, Increasing CKD Screening in Primary Care for Diabetics; SD, standard deviation; uACR, urine albumin-creatine ratio

Following implementation of the INSPIRED pathway, changes were observed across multiple patient-level care processes (Fig. 3). The proportion of patients with documented eGFR testing within the prior year remained high and stable, decreasing slightly from 95% pre-intervention to 93% post-intervention (N = 110 for both periods). In contrast, uACR screening increased from 49% pre-intervention to 60% post-intervention (N = 110 for both periods).

Fig. 3.

Impact of INSPIRED intervention on screening, diagnosis, and referral for patients with CKD and DM Changes in the percentage of patients 3 months prior to and post-intervention with screening values, an ICD-10 code, or referral to nephrology. CKD, chronic kidney disease; DM, diabetes mellitus; eGFR, estimated glomerular filtration rate; ICD-10, International Classification of Diseases version 10; INSPIRED, Increasing CKD Screening in Primary Care for Diabetics

The proportion of patients with an ICD-10 diagnosis code for CKD increased modestly from 12% pre-intervention to 15% post-intervention (N = 110). Among patients meeting criteria for advanced kidney disease (eGFR < 30 mL/min/1.73 m² or CKD stage 4), all patients (100%) had evidence of nephrology referral both before (N = 10) and after (N = 7) the intervention.

SGLT2i prescribing trends (Geisinger only)

SGLT2i prescribing trends were evaluated only at Geisinger due to the availability of system-level prescribing dashboards, which were not the focus at UPMC for this analysis. The proportion of patients with diabetes and an active SGLT2i prescription showed a modest descriptive increase from approximately 17% during the pre-intervention period to approximately 19% during and following the educational intervention (Fig. 4). Formal inferential statistical testing was not performed, as the study was not powered to detect statistically significant differences in prescribing rates; therefore, results are presented as descriptive trends.

Fig. 4.

Percentage of patients with active SGLT2i prescriptions on a rolling monthly basis at Geisinger Trends reflect prescribing patterns within the context of a multidisciplinary intervention emphasizing guideline-directed use of SGLT2i, sodium-glucose cotransporter-2 inhibitor, among eligible patients, including those with CKD risk factors or reduced eGFR (≥20 mL/min/1.73 m²), consistent with the CKD Care Pathway

Provider satisfaction with educational materials and workflows

Most providers at both systems were “satisfied” or “very satisfied” with the educational materials (85% and 61%, respectively). They appreciated the increased awareness of when and whom to screen for CKD, information on appropriate treatment options, and felt more confident in CKD guidelines with this educational intervention. The few providers who were only “somewhat satisfied” cited reasons such as already being familiar with the guidelines.

While 33% of respondents at UPMC were “satisfied” or “very satisfied” with the proposed eGFR/uACR workflow, 50% were “indifferent” to this update. Some providers requested more education on communicating topics to patients and consistent use of uACR screening for patients with HTN alone. Despite these reservations, 75% were either “likely” or “extremely likely” to recommend these educational materials and workflow changes to others.

Respondents at Geisinger were mostly “very satisfied” (6%), “satisfied” (39%), or “indifferent” (38%) to the proposed SGLT2i workflow. Of those who were at least satisfied, 2 were unsure of the impact these changes would have on long-term patient outcomes. Despite this, > 83% were either “likely” or “extremely likely” to recommend this workflow to others, and 78% stated it was “easy to incorporate into their current workstreams.”

Discussion

The 3-month educational intervention was associated with improvements in clinicians’ knowledge and confidence in CKD care at both systems, with shifts toward higher confidence levels across multiple domains, including identification and interpretation of uACR values. However, a subset of clinicians continued to report moderate or lower confidence, particularly in managing severe CKD, indicating ongoing educational needs. At Geisinger, confidence in prescribing SGLT2i improved, accompanied by a modest increase in prescribing rates. Barriers to care also evolved over the study period, with time constraints and patient awareness cited less frequently, while challenges related to regular CKD screening and medication cost persisted. Clinicians reported high satisfaction with the educational materials and workflows, with most indicating that these resources were easy to implement and recommending them to others.

Our findings are consistent with prior studies demonstrating that structured educational interventions improve clinician knowledge and confidence in CKD care. For example, Smekal et al. reported that targeted training programs improved providers’ knowledge and confidence in CKD management [8]. Similarly, Donald et al. demonstrated that implementation of an online CKD clinical pathway was associated with improved adherence to guideline-recommended diagnostic testing, particularly in regions with the highest dissemination [9]. While these prior studies focused on targeted or digital interventions, our approach emphasized department-wide educational initiatives combined with implementation of a shared care pathway. In our study, integration of the pathway with EMR-supported workflows was associated with increased uACR screening at UPMC and modest improvements in SGLT2i prescribing at Geisinger. Collectively, these findings support the value of multifaceted interventions that combine education with systems-level changes to enhance CKD care delivery. However, additional research is needed to evaluate the long-term sustainability of these improvements and their impact on clinical outcomes.

These findings highlight the importance of aligning educational interventions with local workflows and practice needs when implementing clinical pathways in primary care. From an implementation perspective, consistent delivery of core educational content combined with locally tailored workflow integration appeared critical to adoption. Key operational components included leveraging existing clinical meetings, integrating pathway elements into EMR workflows (when available), and reinforcing learning through repeated educational touchpoints. Educational interventions should also be informed by analyses of local learning needs, goal-oriented with measurable objectives, and adapted to the specific practice environment. Facilitators of implementation included leadership engagement, multidisciplinary care team involvement, access to EMR data for performance monitoring, and alignment with national guideline recommendations. These elements may inform replication of similar interventions in other health system settings [6].

Scalability of this approach may vary across healthcare systems due to differences in EMR capabilities, quality improvement infrastructure, clinician time constraints, and medication access barriers. Adaptation to local workflows and resources will likely be necessary for broader implementation. Although improvements were observed within the 3-month study period, the sustainability of these effects over time remains uncertain and warrants further longitudinal evaluation.

This pilot study has the following limitations. The small sample size limits statistical power, generalizability, and the ability to perform subgroup analyses by provider type. Survey participation was voluntary and may not represent all clinicians involved in patient care, introducing response and nonresponse bias. Practice-level factors such as payor mix and participation in alternative payment models were not formally assessed and may have influenced baseline CKD awareness. The use of multiple response scale formats across survey items may have introduced variability in interpretation. Changes in SGLT2i prescribing were evaluated descriptively using dashboard data without inferential statistical testing. Although a patient-facing component was included in INSPIRED, its effect on patient knowledge or CKD screening rates was not evaluated.

Conclusion

CKD presents a global burden, particularly in its early stages, with many barriers to care delivery. Gaps in HCP knowledge contribute to limited use of appropriate screening tests [7–10, 16]. Other barriers include limited primary care involvement, insufficient nephrology consultations based on target GFR/uACR levels, lack of evidence-based interventions for HCPs, and underutilization of EMR tools [7–10, 16]. These gaps hinder early diagnosis and appropriate patient risk stratification [7–10, 16]. Integrating a standardized care pathway within diverse systems, supported by targeted educational interventions, addresses key barriers to CKD management. The CKD Care Pathway, implemented at 2 institutions with different patient populations and workflows, enhances HCP knowledge, confidence, and adherence to guidelines. This facilitates earlier diagnosis and improved management of CKD, with the potential to improve patient outcomes and reduce the disease’s long-term burden.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

Conceptualization, J.G, A.C., T.C., M.S.K, A.E., M.J.; methodology, J.G, A.C., T.C., M.S.K, A.E., M.J.; writing—review and editing, J.G, A.C., T.C., M.S.K, A.E., M.J.; supervision, J.G, A.C., T.C., M.S.K, A.E., M.J. All authors have read and agreed to the published version of the manuscript.

Funding

Funding and support for the development of this manuscript were provided by Boehringer Ingelheim and Eli Lilly.

Data availability

The datasets generated and analyzed during the current study are not publicly available due to institutional restrictions and the need to protect participant confidentiality. Data may be available from the corresponding author on reasonable request.

Declarations

Human ethics and consent to participate

IRB approval forms were submitted to the University of Pittsburgh Medical Center’s Institutional Review Board Office of Research Protections and Geisinger Institutional Review Board. Protocols were reviewed and qualified for exemption under United States Department of Health and Human Services regulations [(45 CFR 46.104)]. Informed consent to participate was waived by the IRB because the study qualified for exemption under 45 CFR 46.104. The study was conducted in accordance with the ethical principles outlined in the Declaration of Helsinki and in compliance with applicable institutional and federal regulations.

Consent for publication

Not applicable.

Competing interests

Manisha Jhamb, MD, MPH reports grants from Networks of Excellence during the conduct of the study and personal fees from Eli Lilly, Boehringer Ingelheim, Networks of Excellence, and Clinical Care Targeted Communications, LLC, as well as grants from Pfizer, Bayer, and Dialysis Clinic Inc., outside the submitted work. Alexander Chang, MD reports research funding from Boehringer Ingelheim, Novartis, and Bayer. Tracey Conti, MD reports consulting fees from Bayer.Evan R. Norfolk, MD reports consulting relationships with Petauri Kinect and Fresenius Medical Care. Ahlam Elbedewe is an employee of Petauri Kinect. The remaining authors (Jamie Green, MD, MS; Maria S. Kobylinski, MD; and Melissa Weimer, MS) declare that they have no competing interests.Funding and support for the development of this manuscript were provided by Boehringer Ingelheim and Eli Lilly. The funders had no role in the study design; data collection, analysis, or interpretation; manuscript preparation; or the decision to submit the manuscript for publication.