Maggy Spolnik, Indika Mallawaarachchi, Binu Sharma, Michael Ellwood, Jennie Z Ma, Jason A Lyman, Julia J Scialla
{"title":"Implementation of a Kidney Screening Intervention to Improve Early CKD Detection in Adults with Diabetes.","authors":"Maggy Spolnik, Indika Mallawaarachchi, Binu Sharma, Michael Ellwood, Jennie Z Ma, Jason A Lyman, Julia J Scialla","doi":"10.34067/KID.0000000922","DOIUrl":null,"url":null,"abstract":"<p><strong>Background: </strong>Clinical guidelines recommend that patients with diabetes mellitus (DM) are screened annually for kidney disease with estimated glomerular filtration rate (eGFR) and urine albumin-to-creatinine ratio (UACR). To improve screening, we implemented clinical decision support (CDS) at UVA Health in April 2022. This included: (1) auto-enrollment of primary care patients with DM in an electronic health record (EHR)-based health maintenance plan; and (2) prompting one-click ordering of the Kidney Profile (KP; panel including UACR and eGFR) or UACR alone, as needed.</p><p><strong>Methods: </strong>We assessed effectiveness of the CDS using an interrupted time series approach across 3 periods (pre-COVID-19 control: January 2019-February 2020; post-COVID-19 control: March 2021-April 2022; post-CDS: May 2022 to April 2023). All non-acute office and telehealth encounters in primary care for patients aged ≥22 years with DM, no coded diagnosis of CKD in the prior 4 years, and due for screening (i.e., not screened for CKD in past 365 days). Screening was assessed as orders placed for UACR within 30 days of the encounter and aggregated by calendar months.</p><p><strong>Results: </strong>There were 66,388 encounters (23,419 pre-COVID-19 control; 22,611 post-COVID-19 control; 20,358 post-CDS). The screening trend in both control periods was similar, therefore only the post-COVID-19 control was considered further. Demographics, encounter types, and clinic distribution were similar in the control and post-CDS periods. There was an immediate screening difference of 3.02% (95% CI, 0.37- 5.68; p=0.03) after the CDS, and screening acceleration with a difference in screening rate of 0.57% each month compared to 0.06% per month prior to the CDS (p<0.01). Results were similar if encounters for patients with prior CKD by laboratory criteria were removed.</p><p><strong>Conclusions: </strong>Roll out of CDS coincided with immediate and ongoing improvement in annual screening for CKD among adult patients with DM. These results suggest that simple CDS may be an effective intervention to promote CKD screening.</p>","PeriodicalId":17882,"journal":{"name":"Kidney360","volume":" ","pages":""},"PeriodicalIF":3.0000,"publicationDate":"2025-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Kidney360","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.34067/KID.0000000922","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"UROLOGY & NEPHROLOGY","Score":null,"Total":0}
引用次数: 0
Abstract
Background: Clinical guidelines recommend that patients with diabetes mellitus (DM) are screened annually for kidney disease with estimated glomerular filtration rate (eGFR) and urine albumin-to-creatinine ratio (UACR). To improve screening, we implemented clinical decision support (CDS) at UVA Health in April 2022. This included: (1) auto-enrollment of primary care patients with DM in an electronic health record (EHR)-based health maintenance plan; and (2) prompting one-click ordering of the Kidney Profile (KP; panel including UACR and eGFR) or UACR alone, as needed.
Methods: We assessed effectiveness of the CDS using an interrupted time series approach across 3 periods (pre-COVID-19 control: January 2019-February 2020; post-COVID-19 control: March 2021-April 2022; post-CDS: May 2022 to April 2023). All non-acute office and telehealth encounters in primary care for patients aged ≥22 years with DM, no coded diagnosis of CKD in the prior 4 years, and due for screening (i.e., not screened for CKD in past 365 days). Screening was assessed as orders placed for UACR within 30 days of the encounter and aggregated by calendar months.
Results: There were 66,388 encounters (23,419 pre-COVID-19 control; 22,611 post-COVID-19 control; 20,358 post-CDS). The screening trend in both control periods was similar, therefore only the post-COVID-19 control was considered further. Demographics, encounter types, and clinic distribution were similar in the control and post-CDS periods. There was an immediate screening difference of 3.02% (95% CI, 0.37- 5.68; p=0.03) after the CDS, and screening acceleration with a difference in screening rate of 0.57% each month compared to 0.06% per month prior to the CDS (p<0.01). Results were similar if encounters for patients with prior CKD by laboratory criteria were removed.
Conclusions: Roll out of CDS coincided with immediate and ongoing improvement in annual screening for CKD among adult patients with DM. These results suggest that simple CDS may be an effective intervention to promote CKD screening.