Community-acquired AKI (CA-AKI) is a common but under-recognized condition developed outside of the hospital that increases the risk of future hospitalization and death.1 Despite this, little is known about nephrology care patterns after such an event. Using a national retrospective cohort from the Veterans Health Administration (VA), we evaluated outpatient nephrology care ≤90 days after an episode of CA-AKI.
We assessed the cumulative CA-AKI incidence among VA primary care users in 2013–2017 with a recorded outpatient serum creatinine. AKI was identified from two consecutive creatinine values ≤12 months, with each creatinine serving as the index value for the subsequent creatinine.2 Our sample comprised Veterans with stage 2 and 3 CA-AKI, defined as ≥2.0–2.9-fold or ≥3-fold relative increase, respectively, in outpatient or inpatient creatinine (≤24 hours from admission), from the index value ≤12 months earlier.
The primary outcome was outpatient nephrology follow-up visit after CA-AKI, ascertained from in-person and telehealth VA outpatient encounters. This outcome was modeled in two ways: a nephrology encounter ≤90 days from the AKI event with (1) any diagnosis code and (2) an AKI-specific diagnosis code (584.x, N17.x). A Cox model was used to estimate associations between CA-AKI event characteristics (stage, setting) and nephrology care, adjusting for baseline sociodemographics and comorbidities.
Of 43,473 Veterans with stage 2 (73%) or stage 3 (27%) CA-AKI between 2013 and 2017, the mean age was 67 years, 6% were female, 21% were of Black race, and 7% were of Hispanic ethnicity. Mean eGFR was 71 ml/min per 1.73 m2, and 16% had documented CKD. In the 90 days before their CA-AKI event, 63% of Veterans had a primary care visit and 1% had a nephrology visit. Overall, 4% of Veterans had an outpatient nephrology visit ≤90 days after AKI (3% stage 2, 4% stage 3; 4% outpatient, 3% inpatient) before death or hospitalization. Mean qualifying creatinine levels were 4.4 and 3.1 mg/dl, respectively, for Veterans with and without outpatient nephrology follow-up. Of the 1519 outpatient nephrology visits after CA-AKI, 725 (48%) included a diagnosis code for AKI (44% of visits after stage 2, 57% after stage 3, 51% after outpatient, 43% after inpatient). There were 3740 (9%) and 10,485 (24%) Veterans with death or hospitalization, respectively, in ≤90 days.
In a Cox model (Table 1), Veterans with inpatient versus outpatient CA-AKI (hazard ratio, 0.64 [95% confidence interval, 0.57 to 0.72]) and advancing age (hazard ratio, 0.88 [0.86 to 0.91] per 5 years) had lower rates of outpatient nephrology visits ≤90 days after AKI. Those with stage 3 AKI (1.60 [1.43 to 1.80]), CKD (1.58 [1.38 to 1.80]), lower baseline eGFR (1.13 [1.12 to 1.15] per five-unit decrease), or prior nephrology visit ≤90 days (3.40 [2.74 to 4.22]) had higher rates of having a nephrology visit in ≤90 days. Similar patterns were observed in the outpatient and inpatient settings, and when requiring the nephrology visit to have a concomitant diagnosis code for AKI, with the exception of a documented diagnosis of CKD, which was not statistically significant.
Table 1.
Factors associated with nephrology visit ≤90 days after community-acquired AKIa
| Parameter | Hazard Ratio (95% Confidence Interval) | |||||
|---|---|---|---|---|---|---|
| Overall CA-AKI | Outpatient CA-AKI | Inpatient CA-AKI | ||||
| Model 1: with Any Diagnosis Code No. events=1519 |
Model 2: with AKI Diagnosis Code No. events=725 |
Model 1: with Any Diagnosis Code No. events=940 |
Model 2: with AKI Diagnosis Code No. events=477 |
Model 1: with Any Diagnosis Code No. events=579 |
Model 2: with AKI Diagnosis Code No. events=248 |
|
| Inpatient CA-AKI (ref: outpatient CA-AKI) | 0.64 (0.57 to 0.72)b | 0.51 (0.43 to 0.60)b | — | — | — | — |
| Stage 3 (ref: CA-AKI stage 2) | 1.60 (1.43 to 1.80)b | 2.10 (1.79 to 2.46)b | 1.46 (1.24 to 1.71)b | 1.90 (1.54 to 2.34)b | 1.78 (1.50 to 2.10)b | 2.44 (1.88 to 3.15)b |
| Age at index, per 5 yr | 0.88 (0.86 to 0.91)b | 0.91 (0.87 to 0.95)b | 0.92 (0.89 to 0.96)b | 0.95 (0.90 to 1.00)b | 0.83 (0.80 to 0.87)b | 0.85 (0.79 to 0.90)b |
| Black (ref: White) | 0.90 (0.79 to 1.03) | 0.91 (0.75 to 1.10) | 0.94 (0.80 to 1.12) | 0.98 (0.77 to 1.24) | 0.84 (0.68 to 1.03) | 0.78 (0.57 to 1.08) |
| Other race (ref: White) | 1.34 (0.98 to 1.82) | 1.77 (1.17 to 2.68) | 1.27 (0.85 to 1.91) | 1.59 (0.93 to 2.72) | 1.54 (0.94 to 2.52) | 2.27 (1.19 to 4.34)b |
| Unknown race (ref: White) | 0.89 (0.69 to 1.13) | 0.99 (0.70 to 1.40) | 0.95 (0.70 to 1.29) | 1.11 (0.74 to 1.66) | 0.73 (0.48 to 1.11) | 0.70 (0.35 to 1.41) |
| Hispanic or Latino (ref: Non-Hispanic) | 1.09 (0.88 to 1.35) | 0.93 (0.68 to 1.29) | 1.05 (0.80 to 1.38) | 0.94 (0.63 to 1.39) | 1.13 (0.79 to 1.60) | 0.92 (0.53 to 1.60) |
| Other ethnicity (ref: Non-Hispanic) | 1.28 (0.91 to 1.81) | 1.06 (0.64 to 1.78) | 1.22 (0.80 to 1.88) | 1.19 (0.67 to 2.12) | 1.44 (0.80 to 2.60) | 0.74 (0.23 to 2.41) |
| Documented diagnosis of CKD (any stage) | 1.58 (1.38 to 1.80)b | 1.06 (0.86 to 1.30) | 1.46 (1.23 to 1.74)b | 0.91 (0.70 to 1.19) | 1.78 (1.43 to 2.20)b | 1.40 (0.99 to 1.98) |
| eGFR, ml/min per 1.73 m2, per five unit decrease | 1.13 (1.12 to 1.15)b | 1.10 (1.08 to 1.12)b | 1.13 (1.11 to 1.15)b | 1.10 (1.07 to 1.13)b | 1.15 (1.12 to 1.18)b | 1.10 (1.06 to 1.14)b |
| Proton pump inhibitors | 1.09 (0.98 to 1.21) | 1.19 (1.02 to 1.38)b | 1.08 (0.94 to 1.24) | 1.19 (0.99 to 1.44) | 1.13 (0.95 to 1.35) | 1.24 (0.95 to 1.61) |
| NSAIDs | 0.97 (0.87 to 1.08) | 0.94 (0.80 to 1.10) | 0.95 (0.82 to 1.09) | 0.90 (0.74 to 1.10) | 1.01 (0.85 to 1.21) | 1.01 (0.78 to 1.33) |
| Diuretics | 1.01 (0.90 to 1.14) | 0.90 (0.77 to 1.06) | 1.09 (0.94 to 1.26) | 1.00 (0.82 to 1.22) | 0.88 (0.74 to 1.06) | 0.70 (0.52 to 0.93)b |
| RAAS blockers | 1.11 (0.98 to 1.25) | 1.06 (0.89 to 1.27) | 1.30 (1.10 to 1.52)b | 1.25 (1.00 to 1.57)b | 0.82 (0.67 to 1.00) | 0.73 (0.54 to 0.99)b |
| Number of outpatient evaluation and management visits in the prior year, per 5 | 1.07 (1.04 to 1.09)b | 1.07 (1.04 to 1.10)b | 1.09 (1.06 to 1.12)b | 1.08 (1.04 to 1.12)b | 1.03 (0.99 to 1.08) | 1.06 (1.00 to 1.12)b |
| 1–14 total inpatient days in the prior year (ref: 0 d) | 0.93 (0.81 to 1.06) | 0.94 (0.77 to 1.13) | 0.84 (0.71 to 1.00) | 0.88 (0.69 to 1.12) | 1.11 (0.90 to 1.37) | 1.04 (0.75 to 1.43) |
| >14 total inpatient days in the prior year (ref: 0 d) | 0.85 (0.68 to 1.06) | 0.95 (0.68 to 1.31) | 0.89 (0.67 to 1.17) | 0.92 (0.62 to 1.36) | 0.84 (0.55 to 1.27) | 1.06 (0.59 to 1.91) |
| PCP visit in prior 90 d (ref: no) | 1.03 (0.92 to 1.15) | 1.07 (0.91 to 1.25) | 0.99 (0.87 to 1.14) | 1.02 (0.84 to 1.24) | 1.11 (0.93 to 1.33) | 1.21 (0.91 to 1.60) |
| Outpatient nephrology visit prior 90 d (ref: no) | 3.40 (2.74 to 4.22)b | 1.92 (1.24 to 2.98)b | 3.12 (2.31 to 4.21)b | 1.81 (0.99 to 3.32) | 3.97 (2.90 to 5.44)b | 2.17 (1.14 to 4.14)b |
CA-AKI, community-acquired AKI; NSAID, nonsteroidal anti-inflammatory drug; PCP, primary care provider; RAAS, renin-angiotensin-aldosterone system.
Select variables shown; the full model includes dichotomous indicators of 28 comorbid conditions.
Results from associations that met statistical significance.
In this study, few Veterans with stages 2–3 CA-AKI had an outpatient nephrology follow-up ≤90 days after an AKI event. Emerging research, such as the ongoing Caring for OutPatiEnts after AKI trial,3 will yield important findings on the role of postdischarge nephrology care after hospital-based episodes of advanced AKI. However, AKI developed outside of the hospital setting is infrequently studied. Given that CA-AKI is estimated to be twice as common as AKI in the hospital and itself is associated with disproportionate risk of hospitalization and death,1 our work supports the need for care models that also prioritize post–CA-AKI care. Of note, Veterans with AKI detected in the outpatient setting were more likely to see nephrology in ≤90 days, which may be because of differences in acute management of AKI in the inpatient versus outpatient setting. This finding aligns with prior work demonstrating the need for care coordination for AKI after hospital discharge.4–6 Study limitations include that results may not generalize to non-Veteran groups, inability to confirm provider recognition of a CA-AKI event when it occurs that may affect follow-up patterns, absence of data beyond 2017, and lack of granularity regarding content of nephrology follow-up care visits (e.g., medication changes). However, the national and longitudinal nature of VA data provides an unparalleled opportunity to examine follow-up care patterns after CA-AKI in a US cohort.
Encouragingly, approximately half of CA-AKI nephrology encounters included an AKI diagnosis, indicating recognition of AKI by nephrologists that exceeds overall documentation for inpatient AKI, despite the potentially extended period between creatinine measurements.7 Given the lack of acuity and reduced visit frequency in the outpatient setting, early recognition and implementation of mitigation strategies are critical to attenuate poor outcomes associated with CA-AKI. Care models that facilitate recognition of CA-AKI, integrate primary care providers in AKI care, and prioritize nephrology care for advanced cases may improve adverse outcomes associated with CA-AKI.
Disclosures
Disclosure forms, as provided by each author, are available with the online version of the article at http://links.lww.com/CJN/B885.
Funding
This study was funded by National Institutes of Health grant R01DK120732 from the National Institute of Diabetes and Digestive and Kidney Diseases, and M.L. Maciejewski received funding from VA HSR&D (RCS 10-391).
Author Contributions
Conceptualization: C. Barrett Bowling, M. Alan Brookhart, Clarissa J. Diamantidis, Matthew L. Maciejewski, Virginia Wang.
Formal analysis: M. Alan Brookhart, Lindsay Zepel.
Funding acquisition: Clarissa J. Diamantidis, Virginia Wang.
Investigation: Matthew L. Maciejewski.
Methodology: C. Barrett Bowling, M. Alan Brookhart, Clarissa J. Diamantidis, Matthew L. Maciejewski, Virginia Wang, Lindsay Zepel.
Supervision: M. Alan Brookhart, Clarissa J. Diamantidis, Virginia Wang.
Writing – original draft: C. Barrett Bowling, M. Alan Brookhart, Clarissa J. Diamantidis, Matthew L. Maciejewski, Virginia Wang, Lindsay Zepel.
Writing – review & editing: C. Barrett Bowling, M. Alan Brookhart, Clarissa J. Diamantidis, Matthew L. Maciejewski, Virginia Wang, Lindsay Zepel.
Data Sharing Statement
Data cannot be shared. VA restrictions.
References
- 1.Wang V Zepel L Maciejewski ML, et al. Outcomes Following Community-Acquired Acute Kidney Injury: A National Cohort Study of US Veterans. Orlando, FL: Poster Presentation American Society of Nephrology Annual Meeting; November 2022. [Google Scholar]
- 2.Diamantidis CJ Zepel L Smith VA, et al. National epidemiology of community-acquired acute kidney injury among US veterans. Am J Kidney Dis;82(3):300–310. doi: 10.1053/j.ajkd.2023.01.448. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.A Patient-Centered Trial of a Process-Of-Care Intervention in Hospitalized AKI Patients: The COPE-AKI Trial (COPE-AKI). 2023. Accessed July 6 2023. https://classic.clinicaltrials.gov/ct2/show/NCT05805709 [Google Scholar]
- 4.Silver SA Adhikari NK Jeyakumar N, et al. Association of an acute kidney injury follow-up clinic with patient outcomes and care processes: a cohort study. Am J Kidney Dis. 2023;81(5):554–563.e1. doi: 10.1053/j.ajkd.2022.10.011 [DOI] [PubMed] [Google Scholar]
- 5.Siew ED Liu KD Bonn J, et al. Improving care for patients after hospitalization with AKI. J Am Soc Nephrol. 2020;31(10):2237–2241. doi: 10.1681/ASN.2020040397 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Silver SA, Siew ED. Follow-up care in acute kidney injury: lost in transition. Adv Chronic Kidney Dis. 2017;24(4):246–252. doi: 10.1053/j.ackd.2017.05.008 [DOI] [PubMed] [Google Scholar]
- 7.Grams ME, Waikar SS, MacMahon B, Whelton S, Ballew SH, Coresh J. Performance and limitations of administrative data in the identification of AKI. Clin J Am Soc Nephrol. 2014;9(4):682–689. doi: 10.2215/CJN.07650713 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
Data cannot be shared. VA restrictions.
