Nephrology Providers’ Perspective and Use of Mortality Prognostic Tools in Dialysis Patients

Nephrology Providers’ Perspective and Use of Mortality Prognostic Tools in Dialysis Patients | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Nephrology Providers’ Perspective and Use of Mortality Prognostic Tools in Dialysis Patients Jennifer Bergeron, Katharine L. Cheung, Christina Marchese, Colton Jensen, and 3 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4249542/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 26 Nov, 2024 Read the published version in BMC Nephrology → Version 1 posted 12 You are reading this latest preprint version Abstract Background: Mortality prognostic tools exist to aid in shared decision making with kidney failure patients but are underutilized. This study aimed to elucidate nephrology providers’ practice patterns and understand barriers to prognostic tool use. Methods: Nephrology providers (8 physicians and 2 nurse practitioners) at an academic medical center underwent semi-structured interviews regarding their experience and perspective on the utility of mortality prognostic tools. Common themes were identified independently by 2 reviewers using grounded theory. Three six-month mortality prognostic tools were applied to the 279 prevalent dialysis patients that the interviewed providers care for. The C statistic was calculated for each tool based on via logistic regression and subsequent ROC analysis. Nephrology providers reviewed the performance of the prognostication tools in their own patient population. A post interview reassessed perspectives and any change in attitudes regarding the tools. Results: Nephrology providers did not use these mortality prognostic tools in their practice. Key barriers identified were provider concern that the tools were not generalizable to their patients, providers’ trust in their own clinical judgement over that of a prognostic tool, time constraints, and lack of knowledge about the data behind these tools. When re-interviewed with the results of the three prognostic tools in their patients, providers thought the tools performed as expected, but still did not intend to use the tools in their practice. They reported that these tools are good for populations, but not individual patients. The providers preferred to use clinical gestalt for prognostication. Conclusion: Although several well validated prognostic tools are available for predicting mortality, the nephrology providers studied do not use them in routine practice, even after an educational intervention. Other approaches should be explored to help incorporate prognostication in shared-decision-making for patients receiving dialysis. Mortality Tools Prognostication Dialysis Kidney Failure Qualitative Perspectives Figures Figure 1 Introduction Kidney disease is common and highly morbid, with over 3 million people worldwide receiving dialysis. The mortality rate among patients receiving maintenance dialysis is a staggering 60% at 5 years 1 . However, much heterogeneity exists 1 , making it difficult to predict patients’ outcomes, particularly in older adults 2 . Accurately predicting mortality is essential for prognostication and honest conversations may enhance advance care planning. In fact, studies have shown that patients with chronic and end stage kidney disease desire this prognostic information in shared decision making 3–7 . In addition, the ASN Choosing Wisely Campaign 8 , the RPA Clinical Practice Guidelines 9 , and the KDIGO 2012 CKD guidelines 10 support that individualized prognostic information should be included in the decision to initiate dialysis. Because prognostication is challenging, several prognostic tools have been developed to help make an accurate prognosis that can be used in these conversations. However, a recent study of Canadian nephrology providers found that > 80% of providers use clinical gestalt to prognosticate and 70% never or rarely use clinical prediction tools 11 . To our knowledge, there is little research focusing on Nephrology providers’ perspectives about and method of use of these tools in the real world. This study aimed to elucidate Nephrology providers’ attitudes about and practice patterns of mortality prognostic tools in their care of patients on dialysis. This study also aimed to discover whether their perspectives and use of these tools changed after they were presented with data on how these tools performed in their own patients and the patients in their state. Methods Study setting and participants This study was conducted at the University of Vermont Medical Center (UVMMC), located in Burlington, Vermont. UVMMC is Vermont’s only academic medical center and serves over 1 million patients in Vermont and northern New York. There are six UVMMC affiliated, non-profit dialysis units. All Nephrology providers (8 physicians and 2 nurse practitioners) caring for patients receiving maintenance dialysis were eligible to participate. Qualitative study methods Semi-structured interviews were conducted via Zoom (Zoom Video Communications, Inc., San Jose, CA) by the first author (JB), in May 2020. Two of the Nephrology providers had worked with JB (medicine resident) before as the attending on Nephrology consults. All the providers knew JB and knew that she was doing this project to support her application to nephrology fellowship. Providers were asked about their knowledge of and experience with mortality prognostic tools for patients receiving dialysis (see interview guide - Supplement 1). The interviews were 20 minutes ± 10 minutes. No field notes were made. The interviews were recorded and transcribed verbatim by CD and JB. The transcripts were not returned to the participants for comment or correction. Qualitative study analysis Two members of the study team, the principal investigator (JB) and a medical student who did not know any of the providers (CD) performed a thematic analysis for content using the transcripts. The backbone of the code tree was created using the questions from the semi-structed interview guide, but the data for each question was analyzed using grounded theory. The initial codes were generated independently and then they were reviewed together for each interview and themes were identified. Disagreements about themes, the coding tree, and final coding were resolved by discussion. Mortality Prognostic Tool Selection : Three mortality prognostic tools commonly reported in the literature and available without cost were selected (See Table 1 ). Cohen et. al’s 2010 model was derived from 514 prevalent hemodialysis patients in New England using age, albumin, dementia, peripheral vascular disease and the surprise question: “Would I be surprised if this patient died in the next six months?”. 12 Charlson et al.’s “Charlson Comorbidity Index” (CCI) was derived from 559 medical patients in the US using age and 16 comorbidites. 13 Couchoud et. al’s algorithm in 2015 was derived from 24,348 incident elderly ESKD patients over 75 years old in France using age, gender, albumin, five comorbidities, and mobility. 14 The three prognostic tools were chosen because they focus on different aspects of prognostication: Cohen’s tool includes provider gestalt with use of the surprise question, Charlson is heavily weighted by comorbidities and is the most commonly used prognostic tool 15 , and Couchoud tool includes mobility and was designed to be used in older adults. Table 1 Mortality prognostic tools, and their required data elements, selected for this study. Cohen Charlson Couchoud Age X X X Gender X Albumin X X Comorbidities Dementia, peripheral vascular disease Myocardial infarction, congestive heart failure, peripheral vascular disease, dementia, COPD, connective tissue disease, diabetes, hemiplegia, chronic kidney disease, solid tumor, lymphoma, leukemia, AIDS Congestive heart failure, peripheral vascular disease, dysrhythmia, active cancer, severe behavioral disorder Mobility X Surprise Question (Would I be surprised if this patient dies in the next six months) X Mortality Prognostication and Measurement In April 2020, 279 prevalent dialysis patients cared for by these Nephrology providers were identified and prospectively followed for six months. All patients receiving maintenance dialysis were included. Data were extracted through chart review of the dialysis electronic medical record (CyberRen) and UVMMC’s EMR (EPIC) in April 2020. Most patients had data in both EMRs. A standardized approach to identify comorbid conditions from the EMRs was used. To capture the most complete assessment of burden of comorbid conditions, a patient was considered to have a comorbidity if it was listed in at least one of the EMRs (as problem list completeness in EMRs varies anywhere from 60–99% 16 ). A patient was considered to have the more severe disease stage if the stages differed in the two EMRs. The most recent serum albumin resulted before May 1st, 2020 was chosen. The providers were given a list of their patients and asked to answer the surprise question 12 for each. The responses and patient characteristics were used in the corresponding online calculators for the prognostic tools. 17–19 Each patient had a score calculated for each of the three tools (Cohen’s result was a percentage from 0 to 100, Charlson’s was a score from 0 to 37, and Couchoud’s was a score from 0 to 28). At six months follow up, EMR review was used to identify patients who had died. The C statistic, or discrimination, for each tool was calculated via logistic regression and subsequent receiver operating characteristic (ROC) analysis using Stata (Stata 16.1, Stata Corp, LLC. College Station, TX). A C statistic of 0.5 is no better than flipping a coin, 0.7 is considered a good model and a C statistic of 0.8 is considered a “strong” or “excellent” model. Brief Intervention and Follow Up Interviews A similar process of email invitation, semi-structured interview (Supplement 2), transcription, and coding was used for the follow up interviews. Providers received the results of the prognostic tools and patients’ outcome at the time of the email invitation (Supplement 3). Results were also reviewed with the providers at the beginning of the interview before the follow-up questions were asked. The follow up interviews were shorter, on average 10 minutes ± 5 minutes. Results The providers (8 MDs and 2 NPs) who participated in the study were 50% female, 60% Caucasian, 30% Asian, and 10% Black and had a mean age of 54 (range 36–73). They had an average of 17 years of practice (range 2 years to 43) and had been trained in a wide variety of locations. They each cared for an average of 34 patients (range 6–55). Providers' Views on Mortality Prognostic Tools Providers were only aware of 2 tools to predict mortality in dialysis patients. 80% of the providers had heard of Cohen’s mortality prognostic tool, especially regarding the surprise question. 10% of the nephrology providers had heard of Charlson comorbidity index. None of the nephrologists used these tools in their current practice. Representative quotes of providers’ views on mortality prognostic tools can be seen in Table 2 . The main barrier identified to use was provider concern that the tool was not applicable or accurate in their specific patients. Most providers also noted that the disease course itself is unpredictable. Time restraints and the addition of more “work” was a barrier identified by all the providers. Lack of knowledge of the tools and the data behind them were also acknowledged by 6 of the 10 providers. All the providers identified clinical experience as their main source of prognostication. ‘ Table 2 Nephrology providers’ perspective on the barriers to use of mortality prediction tools Tools are not applicable or accurate in their patients “Of course, with any calculator, there’s going to be variability. Some people might live longer than what the calculator predicted, some less.” “I don't think I'll live long enough to see a predictive system that I will really believe is that accurate in terms of predicting how people will do.” “I see a big problem with all risk assessment tools. Not that the variables themselves are not valid, but how are they weighted in terms of driving the end number.” “What are the barriers to using the tool? The first one is believing it.” “I've never used it again for predicting for particular person because I don't think one can be that certain.” Disease course is unpredictable “Sometimes I'm surprised that there are people who don't show up for dialysis for weeks, are non-compliant with medications, and they do fine. While some people who follow all the rules suddenly die. It feels unpredictable.” “I worry more that people want to try to develop tools to give certainty, when I know there really isn't any.” Time restraints and the addition of more “work” “It takes time.” “Is it just adding more burden, adding more noise without telling you much.” “I'm sitting with a patient and, oh yeah, I've got this tool and oh my goodness, I can't remember where I where I’ve hidden it and it's somewhere in the computer, but I don't know which one it’s in and isn't in. And has it been calculated and what numbers do I have to put into it. Where do I find those numbers. And then I have to do all that. And for somebody who's technologically challenged like me that gets put to the side really quickly.” “Actually stopping to take the time to import it into a tool doesn't seem very helpful for me, especially since I know these patients well.” Lack of knowledge “Well, number one, not knowing about it.” “I don’t know how hard it is [to use] honestly.” Clinical experience as the main source of prognostication “I don't want to quantitate those sorts of things because I just think about it.” “When you’ve done this for a number of years, your clinical experience helps you out here.” “I guess I’m looking at the same points, but not putting it in the calculator.” “So I'm able to make that judgment without resorting to the tool.” Uncertainty if mortality prognostic information would change patients’ decisions “It would give us an idea of who would do badly or not, but I know that doesn't change the fact that if patients want to continue, they will continue.” “But in the end, it still depends on the patient’s wishes, whether they want to try something.” “I’m not going to say you can’t do it just because you would do poorly, so it gives us an idea of you know how to talk to them and which way we should probably lean towards but it's still you know their choice.” “People don't make decisions based on that information very often. They make decisions based on whether they are risk averse or risk tolerant.” “It's mostly what do people want, what do the people around them want?” The providers identified a few advantages to using mortality prognostic tools. They noted that some patients are number-oriented and being able to provide that information may help those patients in decision making. Providers also noted that these prognostic tools, if predicting a poor prognosis, would be a reminder to have a goals of care conversation and make providers more likely to encourage supportive care over dialysis. The majority of the providers reported they were open to the idea of using these tools in their prognostication if further evidence for the validity and education about the use of these tools was provided. Providers noted that if a tool was shown to have strong discrimination and predicted a high mortality, it would change how they discuss management options with the patient- i.e. make them more likely to encourage supportive care over dialysis. At the same time though, providers expressed that mortality was not the key factor on which to prognosticate and that patients will make decisions based on a variety of quality of life measures. Validation of Mortality Prognostic Tools The overall 6-month mortality in Vermont’s prevalent dialysis population was 14%. Couchoud had the best discrimination of 6-month mortality in Vermont’s dialysis patients with a C statistic of 0.77 compared to Cohen and Charlson at 0.68 (Fig. 1 ). Post Intervention Interview 5 physicians and 2 nurse practitioners participated in the follow up interviews as of March 2023. Of the remaining 3 physicians, 1 no longer worked at the study site, 1 was on maternity leave, and 1 did not respond to emails to arrange a second interview. The providers overall thought the tools performed about “as well as expected”. “There were no surprises.” “I think they’re about what you would expect because I’ll never be that excited about predictive tools.” Providers speculated that the Couchoud model performed the best because of its inclusion patient mobility and tied that in with frailty as a risk factor for not doing well on dialysis. I do think that mobility is a major factor for a lot of patients, so I do think that it was a good idea for the Couchoud model to include that. I like these, these factors in this tool, with the albumin-nutrition and the frailty, because I know those are independent predictors of those not doing well on dialysis. “ It actually makes me think more of mobility as an important index of patient wellness.” Though some acknowledged that it may be because the original Couchoud cohort had the largest study population. Still, while most providers endorsed a “role” for using risk assessment tools, none of the providers routinely used the tools or had plans to implement it into their practice. Providers again voiced concern that although tools are good for populations, and even their specific population, but that they were not accurate for any one specific patient. I think the tools are reasonably good at predicting what will happen in the population, not particularly for what will happen in an individual. So obviously that makes the utility of that somewhat questionable when you’re dealing with the individual rather than planning for the population. I mean they’re nice for studies if you are trying to look at large populations or and you have to have a particular reason for wanting to understand that particular prediction. But for individuals they're never terribly good so I'm not totally surprised. Providers still identified clinical experience and gestalt as their main determinants of prognostication, though a few providers noted that after seeing this data, they might try to incorporate some of the individual risk factors from the tools into their clinical assessment. “ I would place it in my ‘subjective-ometer’ when I’m thinking about these things with the patient.” Discussion This study assessed nephrology providers’ perspectives and use of mortality prognostic tools in dialysis patients. We found that nephrology providers had some knowledge of prognostic tools but did not routinely use them in practice. The main barrier identified to using prognostication tools was the perspective that they are not generalizable nor specific enough for a given patient. After external validation of three routinely available prognostic tools in these providers’ practice, perspectives were seemingly unchanged reflecting a lack of trust in mortality prognostic tools. Provider interviews elicited an interest in mobility as a factor to improve prognostication and suggested a need for better training on how to incorporate prognostic information into serious illness conversations with patients. There was also interest in the tools automatically generating prognostic information in the EMR or using a suite of prognostic tools that include other outcomes besides mortality. Several prognostic tools for mortality on dialysis exist, but few have been externally validated. 20 This study revealed that generalizability was a major concern for nephrology providers. The Vermont population is predominantly rural, white, older, and of a lower socioeconomic status than the derivation cohorts and other studies have shown that these tools don’t perform well in older populations. 2 To respond to this concern, three prognostic tools were validated in providers’ dialysis patient population. In this study, Couchoud’s tool had the highest discrimination for 6-month mortality with a C statistic of 0.765, which is comparable to the highest C statistic found in the 2019 meta-analysis of 32 indices to predict mortality in incident dialysis patients (C statistic 0.74). 15 There, the overall C statistic was 0.71 for any prediction length for mortality and had high heterogeneity, with the sub group analysis for models predicting 6 month mortality range having a C statistic of 0.540-0.896. It is worth noting that the metanalysis was in incident patients rather than the prevalent population in our study. The current study showed that Couchoud’s tool had strong discrimination for 6 mortality and should have assuaged providers’ concerns about external validation, allowing other barriers to be identified in the follow up interviews. Our study confirmed findings by Forzley et al 11 that nephrology providers do not use prognostic tools to provide prognostic information, preferring clinical gestalt. In addition, this study demonstrated that provider preference did not change even after validation of the tools in their patients. Therefore, creating more accurate prognostic tools or making them easier to implement may not increase providers’ use. Provider perspectives suggest a disconnect in patient-physician communication around prognosis, as providers report they are comfortable using gestalt to prognosticate, but other studies show patients aren’t receiving the prognostic information they desire. 3,5,6 Providers may need help implementing prognostic tools to incorporate this information into shared decision making. A recent pilot study that found training nephrologists to use best case/worst case communication improved shared decision making about dialysis and may increase access to palliative care. 21 . As more evidence mounts that dialysis does not confer morbidity or mortality benefits for all patients with kidney failure, 22,23 future studies are needed to help bridge this prognostication gap. This study, as the first to evaluate Nephrology providers’ perceptions and barriers to use of mortality prognostic tools, had several strengths. Foremost was this study’s use of mixed methods and a brief intervention. Externally validating the tools addressed a major concern that the providers identified in the first interview and allowed the subsequent interviews to capture other unresolved barriers. Furthermore, giving the providers the results of their patients’ 6-month mortality next to their predictions and the predictions from the tools (Supplement 3) yielded more grounded and real-world discussion of their perceptions. Performing the second interviews allowed for analysis of any dynamic perceptions and verified previous themes which is often not done in qualitative studies. Lastly, the choice of prognostic tools with different aspects of prognostication allowed the interviews to capture provider perspectives on which parts of prognostication are highest yield. There were limitations to this study. First, it is a small sample size of both providers and patients from one state, and not all providers were available for the second interview. Therefore, the interviewed providers’ responses may not be generalizable. However, the providers do have a wide variety of training backgrounds, employment history, and practice length. Second, social acceptability bias may have been at play as the interviews were not blinded and the primary author conducting the interviews was a resident interested in Nephrology at their institution. Lastly, the interviews were semi-structured which gave the opportunity for more in-depth conversation, but may have introduced interviewer bias with leading questions, wording bias, or confirmation bias. In conclusion, several well validated prognostic tools are available for predicting mortality in dialysis patients, but nephrology providers do not use them in routine practice due to concerns about their applicability in their patients. Addressing the barriers of external validity and lack of knowledge of the tools did not change the nephrology providers’ use or attitude towards the tools. Implementation research is needed to help providers share prognosis and enhance shared-decision-making dialysis. Declarations Ethics approval and consent to participate This study was approved as exempt research by the University of Vermont IRB Committees on Human Research (STUDY00001356). This Committee determined that informed consent was not needed from subjects. Consent for publication I declare that the authors all consent for publication. I confirm that I understand BMC Nephrology is an open access journal that levies an article processing charge per articles accepted for publication. By submitting my article I agree to pay this charge in full if my article is accepted for publication. The results/data/figures in this manuscript have not been published elsewhere, nor are they under consideration by another publisher. I have read the Nature Portfolio journal policies on author responsibilities and submit this manuscript in accordance with those policies. All of the material is owned by the authors and/or no permissions are required. Availability of Data and Materials The dataset generated and/or analyzed during the current study are not publicly available due to this research being considered exempt and we do not have consent from participants to share raw data. Deidentified data may be available from the corresponding author on reasonable request. Competing interests I declare that the authors have no competing interests as defined by BMC, or other interests that might be perceived to influence the results and/or discussion reported in this paper. Author Contributions JB, KC, and AK made substantial contributions to the conceptions and design of the work. JB, CM, CJ, and SM acquired the data. JB, CM, CJ, and BT analyzed the data. JB and KC wrote the main manuscript with substantial revisions by AK. JB, AK, and BT prepared the figures. All authors reviewed the manuscript. Funding None Acknowledgments None References United States Renal Data System. 2023 Annual Data Report: Epidemiology of Kidney Disease in the United States . National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases; 2023. Cheung KL, Montez-Rath ME, Chertow GM, Winkelmayer WC, Periyakoil VS, Kurella Tamura M. Prognostic stratification in older adults commencing dialysis. J Gerontol A Biol Sci Med Sci . 2014;69(8):1033-1039. doi:10.1093/gerona/glt289 Wachterman MW, Marcantonio ER, Davis RB, et al. Relationship Between the Prognostic Expectations of Seriously Ill Patients Undergoing Hemodialysis and Their Nephrologists. JAMA Intern Med . 2013;173(13):1206-1214. doi:10.1001/jamainternmed.2013.6036 Oestreich T, Sayre G, O’Hare AM, Curtis JR, Wong SPY. Perspectives on Conservative Care in Advanced Kidney Disease: A Qualitative Study of US Patients and Family Members. Am J Kidney Dis Off J Natl Kidney Found . 2021;77(3):355-364.e1. doi:10.1053/j.ajkd.2020.07.026 Davison SN. End-of-life care preferences and needs: perceptions of patients with chronic kidney disease. Clin J Am Soc Nephrol CJASN . 2010;5(2):195-204. doi:10.2215/CJN.05960809 Saeed F, Sardar MA, Davison SN, Murad H, Duberstein PR, Quill TE. Patients’ perspectives on dialysis decision-making and end-of-life care. Clin Nephrol . 2019;91(5):294-300. doi:10.5414/CN109608 Thorsteinsdottir B, Espinoza Suarez NR, Curtis S, et al. Older Patients with Advanced Chronic Kidney Disease and Their Perspectives on Prognostic Information: a Qualitative Study. J Gen Intern Med . 2022;37(5):1031-1037. doi:10.1007/s11606-021-07176-8 Williams AW, Dwyer AC, Eddy AA, et al. Critical and honest conversations: the evidence behind the “Choosing Wisely” campaign recommendations by the American Society of Nephrology. Clin J Am Soc Nephrol CJASN . 2012;7(10):1664-1672. doi:10.2215/CJN.04970512 Moss AH. Shared Decision-Making in the Appropriate Initiation of and Withdrawal from Dialysis . Renal Physicians Association; 2010. Accessed March 4, 2023. https://cdn.ymaws.com/www.renalmd.org/resource/resmgr/Store/Shared_Decision_Making_Recom.pdf KDIGO 2012 Clinical Practice Guideline for the Evaluation and Management of Chronic Kidney Disease. Kidney Int Suppl . 2013;3(1). Accessed March 4, 2023. https://kdigo.org/wp-content/uploads/2017/02/KDIGO_2012_CKD_GL.pdf Forzley B, Chiu HHL, Djurdjev O, et al. A Survey of Canadian Nephrologists Assessing Prognostication in End-Stage Renal Disease. Can J Kidney Health Dis . 2017;4:2054358117725294. doi:10.1177/2054358117725294 Cohen LM, Ruthazer R, Moss AH, Germain MJ. Predicting Six-Month Mortality for Patients Who Are on Maintenance Hemodialysis. Clin J Am Soc Nephrol CJASN . 2010;5(1):72-79. doi:10.2215/CJN.03860609 Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis . 1987;40(5):373-383. doi:10.1016/0021-9681(87)90171-8 Couchoud CG, Beuscart JBR, Aldigier JC, Brunet PJ, Moranne OP. Development of a risk stratification algorithm to improve patient-centered care and decision making for incident elderly patients with end-stage renal disease. Kidney Int . 2015;88(5):1178-1186. doi:10.1038/ki.2015.245 Anderson RT, Cleek H, Pajouhi AS, et al. Prediction of Risk of Death for Patients Starting Dialysis: A Systematic Review and Meta-Analysis. Clin J Am Soc Nephrol . 2019;14(8):1213-1227. doi:10.2215/CJN.00050119 Wright A, McCoy AB, Hickman TTT, et al. Problem list completeness in electronic health records: a multi-site study and assessment of success factors. Int J Med Inf . 2015;84(10):784-790. doi:10.1016/j.ijmedinf.2015.06.011 Month Mortality on HD | QxMD. Calculate by QxMD. Accessed February 4, 2024. https://qxmd.com/calculate/calculator_135/6-month-mortality-on-hd Charlson Comorbidity Index (CCI). MDCalc. Accessed February 4, 2024. https://www.mdcalc.com/calc/3917/charlson-comorbidity-index-cci Month Mortality in Incident Elderly ESRD Patients | QxMD. Calculate by QxMD. Accessed February 4, 2024. https://qxmd.com/calculate/calculator_286/3-month-mortality-in-incident-elderly-esrd-patients Ramspek CL, Voskamp PW, van Ittersum FJ, Krediet RT, Dekker FW, van Diepen M. Prediction models for the mortality risk in chronic dialysis patients: a systematic review and independent external validation study. Clin Epidemiol . 2017;9:451-464. doi:10.2147/CLEP.S139748 Zimmermann CJ, Jhagroo RA, Wakeen M, et al. Opportunities to Improve Shared Decision Making in Dialysis Decisions for Older Adults with Life-Limiting Kidney Disease: A Pilot Study. J Palliat Med . 2020;23(5):627-634. doi:10.1089/jpm.2019.0340 O’Connor NR, Kumar P. Conservative Management of End-Stage Renal Disease without Dialysis: A Systematic Review. J Palliat Med . 2012;15(2):228-235. doi:10.1089/jpm.2011.0207 Voorend CGN, van Oevelen M, Verberne WR, et al. Survival of patients who opt for dialysis versus conservative care: a systematic review and meta-analysis. Nephrol Dial Transplant . 2022;37(8):1529-1544. doi:10.1093/ndt/gfac010 Additional Declarations No competing interests reported. Supplementary Files Supplementarymaterial.docx Cite Share Download PDF Status: Published Journal Publication published 26 Nov, 2024 Read the published version in BMC Nephrology → Version 1 posted Editorial decision: Revision requested 03 Sep, 2024 Reviews received at journal 31 Aug, 2024 Reviewers agreed at journal 27 Aug, 2024 Reviews received at journal 22 Aug, 2024 Reviewers agreed at journal 20 Aug, 2024 Reviewers agreed at journal 15 Aug, 2024 Reviewers agreed at journal 14 Aug, 2024 Reviewers invited by journal 12 Aug, 2024 Editor assigned by journal 09 Jul, 2024 Editor invited by journal 23 Apr, 2024 Submission checks completed at journal 15 Apr, 2024 First submitted to journal 10 Apr, 2024 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-4249542","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":291178421,"identity":"4457fe0b-6243-419c-a449-66d85c54bb24","order_by":0,"name":"Jennifer Bergeron","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA6klEQVRIiWNgGAWjYBACxgYeGJP54AMgycNHgha2ZAOQFjbC9sC18JhJgHUS0sA87ezBxxW/bPL5pXvMKr/m2MmwMTA/fHQDn8Nm5yUbnu1Ls5w551jZbdltyUCHsRkb5+DVkmMm2dhz2MDgRvK225LbmIFaeNikCWgx/9nY8x+oJcGsWHJbPVFazBgbfhwAakkxY/y47TAxWvKSJRsbkg0kZ6QlSzNuO87DxkzAL4azcw9+bPhjZ8AvkXzw489t1fb87M0PH+PV0gCyqg3CYQbHETMe5SAgDyb/QF35g4DqUTAKRsEoGJkAAPVJR0clUfghAAAAAElFTkSuQmCC","orcid":"","institution":"West Virginia University School of Medicine","correspondingAuthor":true,"prefix":"","firstName":"Jennifer","middleName":"","lastName":"Bergeron","suffix":""},{"id":291178422,"identity":"c9f0d4d3-088e-407e-a773-099f19332ee8","order_by":1,"name":"Katharine L. Cheung","email":"","orcid":"","institution":"University of Vermont Medical Center","correspondingAuthor":false,"prefix":"","firstName":"Katharine","middleName":"L.","lastName":"Cheung","suffix":""},{"id":291178423,"identity":"b778effb-2f50-48bd-a18f-c1809a77d303","order_by":2,"name":"Christina Marchese","email":"","orcid":"","institution":"University of Vermont Medical Center","correspondingAuthor":false,"prefix":"","firstName":"Christina","middleName":"","lastName":"Marchese","suffix":""},{"id":291178424,"identity":"01d82d4e-e172-45b3-98b1-1f968304e86e","order_by":3,"name":"Colton Jensen","email":"","orcid":"","institution":"University of Vermont Medical Center","correspondingAuthor":false,"prefix":"","firstName":"Colton","middleName":"","lastName":"Jensen","suffix":""},{"id":291178425,"identity":"3bbf7afd-a277-4ace-a5ab-e93fff71f99d","order_by":4,"name":"Sean Meagher","email":"","orcid":"","institution":"Beth Israel Deaconess Medical Center","correspondingAuthor":false,"prefix":"","firstName":"Sean","middleName":"","lastName":"Meagher","suffix":""},{"id":291178426,"identity":"4b741230-808f-4f9d-a5e8-b6ae67a8bd2a","order_by":5,"name":"Amanda G. Kennedy","email":"","orcid":"","institution":"The Robert Larner, MD College of Medicine at The University of Vermont","correspondingAuthor":false,"prefix":"","firstName":"Amanda","middleName":"G.","lastName":"Kennedy","suffix":""},{"id":291178427,"identity":"d9334183-5d48-4476-b73d-963a53d1e2af","order_by":6,"name":"Bradley Tompkins","email":"","orcid":"","institution":"The Robert Larner, MD College of Medicine at The University of Vermont","correspondingAuthor":false,"prefix":"","firstName":"Bradley","middleName":"","lastName":"Tompkins","suffix":""}],"badges":[],"createdAt":"2024-04-11 00:44:16","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-4249542/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4249542/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1186/s12882-024-03861-y","type":"published","date":"2024-11-26T15:58:06+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":55059065,"identity":"4c49bd30-d521-408f-8a8a-42785e37cd89","added_by":"auto","created_at":"2024-04-22 02:06:43","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":92569,"visible":true,"origin":"","legend":"\u003cp\u003eComparison of receiver operating characteristics curves for predicting 6-month mortality for patients on dialysis among the Cohen, Charlson, and Couchoud tools\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-4249542/v1/87baa10a7594c168d346ac9b.png"},{"id":70382833,"identity":"17ffd9d6-fca9-4ef3-8ef1-f47e8a2efcb5","added_by":"auto","created_at":"2024-12-02 16:32:45","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":838261,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4249542/v1/14f0075f-902c-4071-b2a6-c6c27149feec.pdf"},{"id":55059066,"identity":"560aba10-6104-4b71-9080-b71c38d834dd","added_by":"auto","created_at":"2024-04-22 02:06:43","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":300900,"visible":true,"origin":"","legend":"","description":"","filename":"Supplementarymaterial.docx","url":"https://assets-eu.researchsquare.com/files/rs-4249542/v1/4957883f464b2a6277979914.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Nephrology Providers’ Perspective and Use of Mortality Prognostic Tools in Dialysis Patients","fulltext":[{"header":"Introduction","content":"\u003cp\u003eKidney disease is common and highly morbid, with over 3\u0026nbsp;million people worldwide receiving dialysis. The mortality rate among patients receiving maintenance dialysis is a staggering 60% at 5 years\u003csup\u003e1\u003c/sup\u003e. However, much heterogeneity exists\u003csup\u003e1\u003c/sup\u003e, making it difficult to predict patients\u0026rsquo; outcomes, particularly in older adults\u003csup\u003e2\u003c/sup\u003e. Accurately predicting mortality is essential for prognostication and honest conversations may enhance advance care planning. In fact, studies have shown that patients with chronic and end stage kidney disease desire this prognostic information in shared decision making\u003csup\u003e3\u0026ndash;7\u003c/sup\u003e. In addition, the ASN Choosing Wisely Campaign\u003csup\u003e8\u003c/sup\u003e, the RPA Clinical Practice Guidelines\u003csup\u003e9\u003c/sup\u003e, and the KDIGO 2012 CKD guidelines\u003csup\u003e10\u003c/sup\u003e support that individualized prognostic information should be included in the decision to initiate dialysis.\u003c/p\u003e \u003cp\u003eBecause prognostication is challenging, several prognostic tools have been developed to help make an accurate prognosis that can be used in these conversations. However, a recent study of Canadian nephrology providers found that \u0026gt;\u0026thinsp;80% of providers use clinical gestalt to prognosticate and 70% never or rarely use clinical prediction tools\u003csup\u003e11\u003c/sup\u003e. To our knowledge, there is little research focusing on Nephrology providers\u0026rsquo; perspectives about and method of use of these tools in the real world.\u003c/p\u003e \u003cp\u003eThis study aimed to elucidate Nephrology providers\u0026rsquo; attitudes about and practice patterns of mortality prognostic tools in their care of patients on dialysis. This study also aimed to discover whether their perspectives and use of these tools changed after they were presented with data on how these tools performed in their own patients and the patients in their state.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003e \u003cstrong\u003eStudy setting and participants\u003c/strong\u003e \u003cp\u003eThis study was conducted at the University of Vermont Medical Center (UVMMC), located in Burlington, Vermont. UVMMC is Vermont\u0026rsquo;s only academic medical center and serves over 1\u0026nbsp;million patients in Vermont and northern New York. There are six UVMMC affiliated, non-profit dialysis units. All Nephrology providers (8 physicians and 2 nurse practitioners) caring for patients receiving maintenance dialysis were eligible to participate.\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cstrong\u003eQualitative study methods\u003c/strong\u003e \u003cp\u003eSemi-structured interviews were conducted via Zoom (Zoom Video Communications, Inc., San Jose, CA) by the first author (JB), in May 2020. Two of the Nephrology providers had worked with JB (medicine resident) before as the attending on Nephrology consults. All the providers knew JB and knew that she was doing this project to support her application to nephrology fellowship. Providers were asked about their knowledge of and experience with mortality prognostic tools for patients receiving dialysis (see interview guide - Supplement 1). The interviews were 20 minutes\u0026thinsp;\u0026plusmn;\u0026thinsp;10 minutes. No field notes were made. The interviews were recorded and transcribed verbatim by CD and JB. The transcripts were not returned to the participants for comment or correction.\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cstrong\u003eQualitative study analysis\u003c/strong\u003e \u003cp\u003eTwo members of the study team, the principal investigator (JB) and a medical student who did not know any of the providers (CD) performed a thematic analysis for content using the transcripts. The backbone of the code tree was created using the questions from the semi-structed interview guide, but the data for each question was analyzed using grounded theory. The initial codes were generated independently and then they were reviewed together for each interview and themes were identified. Disagreements about themes, the coding tree, and final coding were resolved by discussion.\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cb\u003eMortality Prognostic Tool Selection\u003c/b\u003e: Three mortality prognostic tools commonly reported in the literature and available without cost were selected (See Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Cohen et. al\u0026rsquo;s 2010 model was derived from 514 prevalent hemodialysis patients in New England using age, albumin, dementia, peripheral vascular disease and the surprise question: \u0026ldquo;Would I be surprised if this patient died in the next six months?\u0026rdquo;.\u003csup\u003e12\u003c/sup\u003e Charlson et al.\u0026rsquo;s \u0026ldquo;Charlson Comorbidity Index\u0026rdquo; (CCI) was derived from 559 medical patients in the US using age and 16 comorbidites.\u003csup\u003e13\u003c/sup\u003e Couchoud et. al\u0026rsquo;s algorithm in 2015 was derived from 24,348 incident elderly ESKD patients over 75 years old in France using age, gender, albumin, five comorbidities, and mobility.\u003csup\u003e14\u003c/sup\u003e The three prognostic tools were chosen because they focus on different aspects of prognostication: Cohen\u0026rsquo;s tool includes provider gestalt with use of the surprise question, Charlson is heavily weighted by comorbidities and is the most commonly used prognostic tool\u003csup\u003e15\u003c/sup\u003e, and Couchoud tool includes mobility and was designed to be used in older adults.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eMortality prognostic tools, and their required data elements, selected for this study.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCohen\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eCharlson\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eCouchoud\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eX\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eX\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eX\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGender\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eX\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAlbumin\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eX\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eX\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eComorbidities\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDementia, peripheral vascular disease\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eMyocardial infarction, congestive heart failure, peripheral vascular disease, dementia, COPD, connective tissue disease, diabetes, hemiplegia, chronic kidney disease, solid tumor, lymphoma, leukemia, AIDS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eCongestive heart failure, peripheral vascular disease, dysrhythmia, active cancer, severe behavioral disorder\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMobility\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eX\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSurprise Question (Would I be surprised if this patient dies in the next six months)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eX\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cstrong\u003eMortality Prognostication and Measurement\u003c/strong\u003e \u003cp\u003eIn April 2020, 279 prevalent dialysis patients cared for by these Nephrology providers were identified and prospectively followed for six months. All patients receiving maintenance dialysis were included. Data were extracted through chart review of the dialysis electronic medical record (CyberRen) and UVMMC\u0026rsquo;s EMR (EPIC) in April 2020. Most patients had data in both EMRs. A standardized approach to identify comorbid conditions from the EMRs was used. To capture the most complete assessment of burden of comorbid conditions, a patient was considered to have a comorbidity if it was listed in at least one of the EMRs (as problem list completeness in EMRs varies anywhere from 60\u0026ndash;99%\u003csup\u003e16\u003c/sup\u003e). A patient was considered to have the more severe disease stage if the stages differed in the two EMRs. The most recent serum albumin resulted before May 1st, 2020 was chosen.\u003c/p\u003e \u003c/p\u003e \u003cp\u003eThe providers were given a list of their patients and asked to answer the surprise question\u003csup\u003e12\u003c/sup\u003e for each. The responses and patient characteristics were used in the corresponding online calculators for the prognostic tools.\u003csup\u003e17\u0026ndash;19\u003c/sup\u003e Each patient had a score calculated for each of the three tools (Cohen\u0026rsquo;s result was a percentage from 0 to 100, Charlson\u0026rsquo;s was a score from 0 to 37, and Couchoud\u0026rsquo;s was a score from 0 to 28).\u003c/p\u003e \u003cp\u003eAt six months follow up, EMR review was used to identify patients who had died. The C statistic, or discrimination, for each tool was calculated via logistic regression and subsequent receiver operating characteristic (ROC) analysis using Stata (Stata 16.1, Stata Corp, LLC. College Station, TX). A C statistic of 0.5 is no better than flipping a coin, 0.7 is considered a good model and a C statistic of 0.8 is considered a \u0026ldquo;strong\u0026rdquo; or \u0026ldquo;excellent\u0026rdquo; model.\u003c/p\u003e \u003cp\u003e \u003cstrong\u003eBrief Intervention and Follow Up Interviews\u003c/strong\u003e \u003cp\u003eA similar process of email invitation, semi-structured interview (Supplement 2), transcription, and coding was used for the follow up interviews. Providers received the results of the prognostic tools and patients\u0026rsquo; outcome at the time of the email invitation (Supplement 3). Results were also reviewed with the providers at the beginning of the interview before the follow-up questions were asked. The follow up interviews were shorter, on average 10 minutes\u0026thinsp;\u0026plusmn;\u0026thinsp;5 minutes.\u003c/p\u003e \u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003eThe providers (8 MDs and 2 NPs) who participated in the study were 50% female, 60% Caucasian, 30% Asian, and 10% Black and had a mean age of 54 (range 36\u0026ndash;73). They had an average of 17 years of practice (range 2 years to 43) and had been trained in a wide variety of locations. They each cared for an average of 34 patients (range 6\u0026ndash;55).\u003c/p\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003eProviders' Views on Mortality Prognostic Tools\u003c/h2\u003e \u003cp\u003eProviders were only aware of 2 tools to predict mortality in dialysis patients. 80% of the providers had heard of Cohen\u0026rsquo;s mortality prognostic tool, especially regarding the surprise question. 10% of the nephrology providers had heard of Charlson comorbidity index. None of the nephrologists used these tools in their current practice.\u003c/p\u003e \u003cp\u003eRepresentative quotes of providers\u0026rsquo; views on mortality prognostic tools can be seen in Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e. The main barrier identified to use was provider concern that the tool was not applicable or accurate in their specific patients. Most providers also noted that the disease course itself is unpredictable. Time restraints and the addition of more \u0026ldquo;work\u0026rdquo; was a barrier identified by all the providers. Lack of knowledge of the tools and the data behind them were also acknowledged by 6 of the 10 providers. All the providers identified clinical experience as their main source of prognostication. \u0026lsquo;\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eNephrology providers\u0026rsquo; perspective on the barriers to use of mortality prediction tools\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"2\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTools are not applicable or accurate in their patients\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003e\u0026ldquo;Of course, with any calculator, there\u0026rsquo;s going to be variability. Some people might live longer than what the calculator predicted, some less.\u0026rdquo;\u003c/em\u003e\u003c/p\u003e \u003cp\u003e\u003cem\u003e\u0026ldquo;I don't think I'll live long enough to see a predictive system that I will really believe is that accurate in terms of predicting how people will do.\u0026rdquo;\u003c/em\u003e\u003c/p\u003e \u003cp\u003e\u003cem\u003e\u0026ldquo;I see a big problem with all risk assessment tools. Not that the variables themselves are not valid, but how are they weighted in terms of driving the end number.\u0026rdquo;\u003c/em\u003e\u003c/p\u003e \u003cp\u003e\u003cem\u003e\u0026ldquo;What are the barriers to using the tool? The first one is believing it.\u0026rdquo;\u003c/em\u003e\u003c/p\u003e \u003cp\u003e\u003cem\u003e\u0026ldquo;I've never used it again for predicting for particular person because I don't think one can be that certain.\u0026rdquo;\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDisease course is unpredictable\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003e\u0026ldquo;Sometimes I'm surprised that there are people who don't show up for dialysis for weeks, are non-compliant with medications, and they do fine. While some people who follow all the rules suddenly die. It feels unpredictable.\u0026rdquo;\u003c/em\u003e\u003c/p\u003e \u003cp\u003e\u003cem\u003e\u0026ldquo;I worry more that people want to try to develop tools to give certainty, when I know there really isn't any.\u0026rdquo;\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTime restraints and the addition of more \u0026ldquo;work\u0026rdquo;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003e\u0026ldquo;It takes time.\u0026rdquo;\u003c/em\u003e\u003c/p\u003e \u003cp\u003e\u003cem\u003e\u0026ldquo;Is it just adding more burden, adding more noise without telling you much.\u0026rdquo;\u003c/em\u003e\u003c/p\u003e \u003cp\u003e\u003cem\u003e\u0026ldquo;I'm sitting with a patient and, oh yeah, I've got this tool and oh my goodness, I can't remember where I where I\u0026rsquo;ve hidden it and it's somewhere in the computer, but I don't know which one it\u0026rsquo;s in and isn't in. And has it been calculated and what numbers do I have to put into it. Where do I find those numbers. And then I have to do all that. And for somebody who's technologically challenged like me that gets put to the side really quickly.\u0026rdquo;\u003c/em\u003e\u003c/p\u003e \u003cp\u003e\u003cem\u003e\u0026ldquo;Actually stopping to take the time to import it into a tool doesn't seem very helpful for me, especially since I know these patients well.\u0026rdquo;\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLack of knowledge\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003e\u0026ldquo;Well, number one, not knowing about it.\u0026rdquo;\u003c/em\u003e\u003c/p\u003e \u003cp\u003e\u003cem\u003e\u0026ldquo;I don\u0026rsquo;t know how hard it is [to use] honestly.\u0026rdquo;\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eClinical experience as the main source of prognostication\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003e\u0026ldquo;I don't want to quantitate those sorts of things because I just think about it.\u0026rdquo;\u003c/em\u003e\u003c/p\u003e \u003cp\u003e\u003cem\u003e\u0026ldquo;When you\u0026rsquo;ve done this for a number of years, your clinical experience helps you out here.\u0026rdquo;\u003c/em\u003e\u003c/p\u003e \u003cp\u003e\u003cem\u003e\u0026ldquo;I guess I\u0026rsquo;m looking at the same points, but not putting it in the calculator.\u0026rdquo;\u003c/em\u003e\u003c/p\u003e \u003cp\u003e\u003cem\u003e\u0026ldquo;So I'm able to make that judgment without resorting to the tool.\u0026rdquo;\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eUncertainty if mortality prognostic information would change patients\u0026rsquo; decisions\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003e\u0026ldquo;It would give us an idea of who would do badly or not, but I know that doesn't change the fact that if patients want to continue, they will continue.\u0026rdquo;\u003c/em\u003e\u003c/p\u003e \u003cp\u003e\u003cem\u003e\u0026ldquo;But in the end, it still depends on the patient\u0026rsquo;s wishes, whether they want to try something.\u0026rdquo;\u003c/em\u003e\u003c/p\u003e \u003cp\u003e\u003cem\u003e\u0026ldquo;I\u0026rsquo;m not going to say you can\u0026rsquo;t do it just because you would do poorly, so it gives us an idea of you know how to talk to them and which way we should probably lean towards but it's still you know their choice.\u0026rdquo;\u003c/em\u003e\u003c/p\u003e \u003cp\u003e\u003cem\u003e\u0026ldquo;People don't make decisions based on that information very often. They make decisions based on whether they are risk averse or risk tolerant.\u0026rdquo;\u003c/em\u003e\u003c/p\u003e \u003cp\u003e\u003cem\u003e\u0026ldquo;It's mostly what do people want, what do the people around them want?\u0026rdquo;\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eThe providers identified a few advantages to using mortality prognostic tools. They noted that some patients are number-oriented and being able to provide that information may help those patients in decision making. Providers also noted that these prognostic tools, if predicting a poor prognosis, would be a reminder to have a goals of care conversation and make providers more likely to encourage supportive care over dialysis.\u003c/p\u003e \u003cp\u003eThe majority of the providers reported they were open to the idea of using these tools in their prognostication if further evidence for the validity and education about the use of these tools was provided. Providers noted that if a tool was shown to have strong discrimination and predicted a high mortality, it would change how they discuss management options with the patient- i.e. make them more likely to encourage supportive care over dialysis. At the same time though, providers expressed that mortality was not the key factor on which to prognosticate and that patients will make decisions based on a variety of quality of life measures.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003eValidation of Mortality Prognostic Tools\u003c/h2\u003e \u003cp\u003eThe overall 6-month mortality in Vermont\u0026rsquo;s prevalent dialysis population was 14%. Couchoud had the best discrimination of 6-month mortality in Vermont\u0026rsquo;s dialysis patients with a C statistic of 0.77 compared to Cohen and Charlson at 0.68 (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003ePost Intervention Interview\u003c/h2\u003e \u003cp\u003e5 physicians and 2 nurse practitioners participated in the follow up interviews as of March 2023. Of the remaining 3 physicians, 1 no longer worked at the study site, 1 was on maternity leave, and 1 did not respond to emails to arrange a second interview.\u003c/p\u003e \u003cp\u003eThe providers overall thought the tools performed about \u0026ldquo;as well as expected\u0026rdquo;. \u003cem\u003e\u0026ldquo;There were no surprises.\u0026rdquo; \u0026ldquo;I think they\u0026rsquo;re about what you would expect because I\u0026rsquo;ll never be that excited about predictive tools.\u0026rdquo;\u003c/em\u003e\u003c/p\u003e \u003cp\u003eProviders speculated that the Couchoud model performed the best because of its inclusion patient mobility and tied that in with frailty as a risk factor for not doing well on dialysis.\u003cdiv class=\"BlockQuote\"\u003e\u003cp\u003eI do think that \u003cb\u003emobility is a major factor\u003c/b\u003e for a lot of patients, so I do think that it was a good idea for the Couchoud model to include that.\u003c/p\u003e\u003cp\u003eI like these, these factors in this tool, with the albumin-nutrition and the frailty, because I know those are independent predictors of those not doing well on dialysis.\u003c/p\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003e \u003cem\u003e\u0026ldquo;\u003c/em\u003e \u003cb\u003eIt actually makes me think more of mobility\u003c/b\u003e \u003cem\u003eas an important index of patient wellness.\u0026rdquo;\u003c/em\u003e\u003c/p\u003e \u003cp\u003eThough some acknowledged that it may be because the original Couchoud cohort had the largest study population.\u003c/p\u003e \u003cp\u003eStill, while most providers endorsed a \u0026ldquo;role\u0026rdquo; for using risk assessment tools, none of the providers routinely used the tools or had plans to implement it into their practice.\u003c/p\u003e \u003cp\u003eProviders again voiced concern that although tools are good for populations, and even their specific population, but that they were not accurate for any one specific patient.\u003cdiv class=\"BlockQuote\"\u003e\u003cp\u003eI think the tools are reasonably good at predicting what will happen in the population, not particularly for what will happen in an individual. So obviously that makes the utility of that somewhat questionable when you\u0026rsquo;re dealing with the individual rather than planning for the population.\u003c/p\u003e\u003cp\u003eI mean they\u0026rsquo;re nice for studies if you are trying to look at large populations or and you have to have a particular reason for wanting to understand that particular prediction. But for individuals they're never terribly good so I'm not totally surprised.\u003c/p\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003eProviders still identified clinical experience and gestalt as their main determinants of prognostication, though a few providers noted that after seeing this data, they might try to incorporate some of the individual risk factors from the tools into their clinical assessment. \u0026ldquo;\u003cem\u003eI would place it in my \u0026lsquo;subjective-ometer\u0026rsquo; when I\u0026rsquo;m thinking about these things with the patient.\u0026rdquo;\u003c/em\u003e\u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eThis study assessed nephrology providers’ perspectives and use of mortality prognostic tools in dialysis patients. We found that nephrology providers had some knowledge of prognostic tools but did not routinely use them in practice. The main barrier identified to using prognostication tools was the perspective that they are not generalizable nor specific enough for a given patient. After external validation of three routinely available prognostic tools in these providers’ practice, perspectives were seemingly unchanged reflecting a lack of trust in mortality prognostic tools. Provider interviews elicited an interest in mobility as a factor to improve prognostication and suggested a need for better training on how to incorporate prognostic information into serious illness conversations with patients. There was also interest in the tools automatically generating prognostic information in the EMR or using a suite of prognostic tools that include other outcomes besides mortality. \u0026nbsp;\u003c/p\u003e\n\u003cp\u003eSeveral prognostic tools for mortality on dialysis exist, but few have been externally validated.\u003csup\u003e20\u003c/sup\u003e This study revealed that generalizability was a major concern for nephrology providers. The Vermont population is predominantly rural, white, older, and of a lower socioeconomic status than the derivation cohorts and other studies have shown that these tools don’t perform well in older populations.\u003csup\u003e2\u003c/sup\u003e To respond to this concern, three prognostic tools were validated in providers’ dialysis patient population. In this study, Couchoud’s tool had the highest discrimination for 6-month mortality with a C statistic of 0.765, which is comparable to the highest C statistic found in the 2019 meta-analysis of 32 indices to predict mortality in incident dialysis patients (C statistic 0.74).\u003csup\u003e15\u003c/sup\u003e There, the overall C statistic was 0.71 for any prediction length for mortality and had high heterogeneity, with the sub group analysis for models predicting 6 month mortality range having a C statistic of 0.540-0.896. It is worth noting that the metanalysis was in incident patients rather than the prevalent population in our study. The current study showed that Couchoud’s tool had strong discrimination for 6 mortality and should have assuaged providers’ concerns about external validation, allowing other barriers to be identified in the follow up interviews. \u0026nbsp;\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eOur study confirmed findings by Forzley et al\u003csup\u003e11\u003c/sup\u003e that nephrology providers do not use prognostic tools to provide prognostic information, preferring clinical gestalt. In addition, this study demonstrated that provider preference did not change even after validation of the tools in their patients. Therefore, creating more accurate prognostic tools or making them easier to implement may \u003cem\u003enot\u003c/em\u003e increase providers’ use. Provider perspectives suggest a disconnect in patient-physician communication around prognosis, as providers report they are comfortable using gestalt to prognosticate, but other studies show patients aren’t receiving the prognostic information they desire.\u003csup\u003e3,5,6\u003c/sup\u003e\u0026nbsp; Providers may need help implementing prognostic tools to incorporate this information into shared decision making. A recent pilot study that found training nephrologists to use best case/worst case communication improved shared decision making about dialysis and may increase access to palliative care.\u003csup\u003e21\u003c/sup\u003e. As more evidence mounts that dialysis does not confer morbidity or mortality benefits for all patients with kidney failure,\u003csup\u003e22,23\u003c/sup\u003e future studies are needed to help bridge this prognostication gap.\u003c/p\u003e\n\u003cp\u003eThis study, as the first to evaluate Nephrology providers’ perceptions and barriers to use of mortality prognostic tools, had several strengths. Foremost was this study’s use of mixed methods and a brief intervention. Externally validating the tools addressed a major concern that the providers identified in the first interview and allowed the subsequent interviews to capture other unresolved barriers. Furthermore, giving the providers the results of their patients’ 6-month mortality next to their predictions and the predictions from the tools (Supplement 3) yielded more grounded and real-world discussion of their perceptions. Performing the second interviews allowed for analysis of any dynamic perceptions and verified previous themes which is often not done in qualitative studies. Lastly, the choice of prognostic tools with different aspects of prognostication allowed the interviews to capture provider perspectives on which parts of prognostication are highest yield.\u003c/p\u003e\n\u003cp\u003eThere were limitations to this study. First, it is a small sample size of both providers and patients from one state, and not all providers were available for the second interview. Therefore, the interviewed providers’ responses may not be generalizable. However, the providers do have a wide variety of training backgrounds, employment history, and practice length. Second, social acceptability bias may have been at play as the interviews were not blinded and the primary author conducting the interviews was a resident interested in Nephrology at their institution. Lastly, the interviews were semi-structured which gave the opportunity for more in-depth conversation, but may have introduced interviewer bias with leading questions, wording bias, or confirmation bias.\u003c/p\u003e\n\u003cp\u003eIn conclusion, several well validated prognostic tools are available for predicting mortality in dialysis patients, but nephrology providers do not use them in routine practice due to concerns about their applicability in their patients. Addressing the barriers of external validity and lack of knowledge of the tools did not change the nephrology providers’ use or attitude towards the tools. Implementation research is needed to help providers share prognosis and enhance shared-decision-making dialysis.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study was approved as exempt research by the University of Vermont IRB Committees on Human Research (STUDY00001356). This Committee determined that informed consent was not needed from subjects.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eI declare that the authors all consent for publication.\u0026nbsp;I confirm that I understand BMC Nephrology is an open access journal that levies an article processing charge per articles accepted for publication. By submitting my article I agree to pay this charge in full if my article is accepted for publication. The results/data/figures in this manuscript have not been published elsewhere, nor are they under consideration by another publisher. I have read the Nature Portfolio journal policies on author responsibilities and submit this manuscript in accordance with those policies. All of the material is owned by the authors and/or no permissions are required.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of Data and Materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe dataset generated and/or analyzed during the current study are not publicly available due to this research being considered exempt and we do not have consent from participants to share raw data. Deidentified data may be available from the corresponding author on reasonable request.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eI declare that the authors have no competing interests as defined by BMC, or other interests that might be perceived to influence the results and/or discussion reported in this paper.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor Contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eJB, KC, and AK made substantial contributions to the conceptions and design of the work. JB, CM, CJ, and SM acquired the data. JB, CM, CJ, and BT analyzed the data. JB and KC wrote the main manuscript with substantial revisions by AK. JB, AK, and BT prepared the figures. All authors reviewed the manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNone\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgments\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNone\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eUnited States Renal Data System. \u003cem\u003e2023 Annual Data Report: Epidemiology of Kidney Disease in the United States\u003c/em\u003e. National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases; 2023. \u003c/li\u003e\n\u003cli\u003eCheung KL, Montez-Rath ME, Chertow GM, Winkelmayer WC, Periyakoil VS, Kurella Tamura M. Prognostic stratification in older adults commencing dialysis. \u003cem\u003eJ Gerontol A Biol Sci Med Sci\u003c/em\u003e. 2014;69(8):1033-1039. doi:10.1093/gerona/glt289\u003c/li\u003e\n\u003cli\u003eWachterman MW, Marcantonio ER, Davis RB, et al. Relationship Between the Prognostic Expectations of Seriously Ill Patients Undergoing Hemodialysis and Their Nephrologists. \u003cem\u003eJAMA Intern Med\u003c/em\u003e. 2013;173(13):1206-1214. doi:10.1001/jamainternmed.2013.6036\u003c/li\u003e\n\u003cli\u003eOestreich T, Sayre G, O\u0026rsquo;Hare AM, Curtis JR, Wong SPY. Perspectives on Conservative Care in Advanced Kidney Disease: A Qualitative Study of US Patients and Family Members. \u003cem\u003eAm J Kidney Dis Off J Natl Kidney Found\u003c/em\u003e. 2021;77(3):355-364.e1. doi:10.1053/j.ajkd.2020.07.026\u003c/li\u003e\n\u003cli\u003eDavison SN. End-of-life care preferences and needs: perceptions of patients with chronic kidney disease. \u003cem\u003eClin J Am Soc Nephrol CJASN\u003c/em\u003e. 2010;5(2):195-204. doi:10.2215/CJN.05960809\u003c/li\u003e\n\u003cli\u003eSaeed F, Sardar MA, Davison SN, Murad H, Duberstein PR, Quill TE. Patients\u0026rsquo; perspectives on dialysis decision-making and end-of-life care. \u003cem\u003eClin Nephrol\u003c/em\u003e. 2019;91(5):294-300. doi:10.5414/CN109608\u003c/li\u003e\n\u003cli\u003eThorsteinsdottir B, Espinoza Suarez NR, Curtis S, et al. Older Patients with Advanced Chronic Kidney Disease and Their Perspectives on Prognostic Information: a Qualitative Study. \u003cem\u003eJ Gen Intern Med\u003c/em\u003e. 2022;37(5):1031-1037. doi:10.1007/s11606-021-07176-8\u003c/li\u003e\n\u003cli\u003eWilliams AW, Dwyer AC, Eddy AA, et al. Critical and honest conversations: the evidence behind the \u0026ldquo;Choosing Wisely\u0026rdquo; campaign recommendations by the American Society of Nephrology. \u003cem\u003eClin J Am Soc Nephrol CJASN\u003c/em\u003e. 2012;7(10):1664-1672. doi:10.2215/CJN.04970512\u003c/li\u003e\n\u003cli\u003eMoss AH. \u003cem\u003eShared Decision-Making in the Appropriate Initiation of and Withdrawal from Dialysis\u003c/em\u003e. Renal Physicians Association; 2010. Accessed March 4, 2023. https://cdn.ymaws.com/www.renalmd.org/resource/resmgr/Store/Shared_Decision_Making_Recom.pdf\u003c/li\u003e\n\u003cli\u003eKDIGO 2012 Clinical Practice Guideline for the Evaluation and Management of Chronic Kidney Disease. \u003cem\u003eKidney Int Suppl\u003c/em\u003e. 2013;3(1). Accessed March 4, 2023. https://kdigo.org/wp-content/uploads/2017/02/KDIGO_2012_CKD_GL.pdf\u003c/li\u003e\n\u003cli\u003eForzley B, Chiu HHL, Djurdjev O, et al. A Survey of Canadian Nephrologists Assessing Prognostication in End-Stage Renal Disease. \u003cem\u003eCan J Kidney Health Dis\u003c/em\u003e. 2017;4:2054358117725294. doi:10.1177/2054358117725294\u003c/li\u003e\n\u003cli\u003eCohen LM, Ruthazer R, Moss AH, Germain MJ. Predicting Six-Month Mortality for Patients Who Are on Maintenance Hemodialysis. \u003cem\u003eClin J Am Soc Nephrol CJASN\u003c/em\u003e. 2010;5(1):72-79. doi:10.2215/CJN.03860609\u003c/li\u003e\n\u003cli\u003eCharlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. \u003cem\u003eJ Chronic Dis\u003c/em\u003e. 1987;40(5):373-383. doi:10.1016/0021-9681(87)90171-8\u003c/li\u003e\n\u003cli\u003eCouchoud CG, Beuscart JBR, Aldigier JC, Brunet PJ, Moranne OP. Development of a risk stratification algorithm to improve patient-centered care and decision making for incident elderly patients with end-stage renal disease. \u003cem\u003eKidney Int\u003c/em\u003e. 2015;88(5):1178-1186. doi:10.1038/ki.2015.245\u003c/li\u003e\n\u003cli\u003eAnderson RT, Cleek H, Pajouhi AS, et al. Prediction of Risk of Death for Patients Starting Dialysis: A Systematic Review and Meta-Analysis. \u003cem\u003eClin J Am Soc Nephrol\u003c/em\u003e. 2019;14(8):1213-1227. doi:10.2215/CJN.00050119\u003c/li\u003e\n\u003cli\u003eWright A, McCoy AB, Hickman TTT, et al. Problem list completeness in electronic health records: a multi-site study and assessment of success factors. \u003cem\u003eInt J Med Inf\u003c/em\u003e. 2015;84(10):784-790. doi:10.1016/j.ijmedinf.2015.06.011\u003c/li\u003e\n\u003cli\u003eMonth Mortality on HD | QxMD. Calculate by QxMD. Accessed February 4, 2024. https://qxmd.com/calculate/calculator_135/6-month-mortality-on-hd\u003c/li\u003e\n\u003cli\u003eCharlson Comorbidity Index (CCI). MDCalc. Accessed February 4, 2024. https://www.mdcalc.com/calc/3917/charlson-comorbidity-index-cci\u003c/li\u003e\n\u003cli\u003eMonth Mortality in Incident Elderly ESRD Patients | QxMD. Calculate by QxMD. Accessed February 4, 2024. https://qxmd.com/calculate/calculator_286/3-month-mortality-in-incident-elderly-esrd-patients\u003c/li\u003e\n\u003cli\u003eRamspek CL, Voskamp PW, van Ittersum FJ, Krediet RT, Dekker FW, van Diepen M. Prediction models for the mortality risk in chronic dialysis patients: a systematic review and independent external validation study. \u003cem\u003eClin Epidemiol\u003c/em\u003e. 2017;9:451-464. doi:10.2147/CLEP.S139748\u003c/li\u003e\n\u003cli\u003eZimmermann CJ, Jhagroo RA, Wakeen M, et al. Opportunities to Improve Shared Decision Making in Dialysis Decisions for Older Adults with Life-Limiting Kidney Disease: A Pilot Study. \u003cem\u003eJ Palliat Med\u003c/em\u003e. 2020;23(5):627-634. doi:10.1089/jpm.2019.0340\u003c/li\u003e\n\u003cli\u003eO\u0026rsquo;Connor NR, Kumar P. Conservative Management of End-Stage Renal Disease without Dialysis: A Systematic Review. \u003cem\u003eJ Palliat Med\u003c/em\u003e. 2012;15(2):228-235. doi:10.1089/jpm.2011.0207\u003c/li\u003e\n\u003cli\u003eVoorend CGN, van Oevelen M, Verberne WR, et al. Survival of patients who opt for dialysis versus conservative care: a systematic review and meta-analysis. \u003cem\u003eNephrol Dial Transplant\u003c/em\u003e. 2022;37(8):1529-1544. doi:10.1093/ndt/gfac010\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"bmc-nephrology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bnep","sideBox":"Learn more about [BMC Nephrology](http://bmcnephrol.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/bnep/default.aspx","title":"BMC Nephrology","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Mortality, Tools, Prognostication, Dialysis, Kidney Failure, Qualitative, Perspectives","lastPublishedDoi":"10.21203/rs.3.rs-4249542/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4249542/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eBackground: \u003c/strong\u003eMortality prognostic tools exist to aid in shared decision making with kidney failure patients but are underutilized. This study aimed to elucidate nephrology providers’ practice patterns and understand barriers to prognostic tool use.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods: \u003c/strong\u003eNephrology providers (8 physicians and 2 nurse practitioners) at an academic medical center underwent semi-structured interviews\u003cstrong\u003e \u003c/strong\u003eregarding their experience and perspective on the utility of mortality prognostic tools. Common themes were identified independently by 2 reviewers using grounded theory. Three six-month mortality prognostic tools were applied to the 279 prevalent dialysis patients that the interviewed providers care for. The C statistic was calculated for each tool based on via logistic regression and subsequent ROC analysis. Nephrology providers reviewed the performance of the prognostication tools in their own patient population. A post interview reassessed perspectives and any change in attitudes regarding the tools.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults: \u003c/strong\u003eNephrology providers did not use these mortality prognostic tools in their practice. Key barriers identified were provider concern that the tools were not generalizable to their patients, providers’ trust in their own clinical judgement over that of a prognostic tool, time constraints, and lack of knowledge about the data behind these tools. When re-interviewed with the results of the three prognostic tools in their patients, providers thought the tools performed as expected, but still did not intend to use the tools in their practice. They reported that these tools are good for populations, but not individual patients. The providers preferred to use clinical gestalt for prognostication.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusion: \u003c/strong\u003eAlthough several well validated prognostic tools are available for predicting mortality, the nephrology providers studied do not use them in routine practice, even after an educational intervention. Other approaches should be explored to help incorporate prognostication in shared-decision-making for patients receiving dialysis.\u003c/p\u003e","manuscriptTitle":"Nephrology Providers’ Perspective and Use of Mortality Prognostic Tools in Dialysis Patients","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-04-22 02:06:38","doi":"10.21203/rs.3.rs-4249542/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2024-09-03T17:27:07+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-08-31T22:18:16+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"118710906232337775138268490369616621715","date":"2024-08-27T19:43:35+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-08-22T18:31:03+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"296789558250489546530765407019828145328","date":"2024-08-20T12:09:45+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"134175900042536215646177052148902907089","date":"2024-08-15T22:31:13+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"306150872894856196430760393142743665483","date":"2024-08-14T19:22:56+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2024-08-12T13:23:29+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2024-07-09T07:28:41+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2024-04-23T11:16:20+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2024-04-15T05:30:05+00:00","index":"","fulltext":""},{"type":"submitted","content":"BMC Nephrology","date":"2024-04-11T00:31:29+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"bmc-nephrology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bnep","sideBox":"Learn more about [BMC Nephrology](http://bmcnephrol.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/bnep/default.aspx","title":"BMC Nephrology","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"1a65bb53-892c-4e63-b530-895567331a9a","owner":[],"postedDate":"April 22nd, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2024-12-02T16:04:30+00:00","versionOfRecord":{"articleIdentity":"rs-4249542","link":"https://doi.org/10.1186/s12882-024-03861-y","journal":{"identity":"bmc-nephrology","isVorOnly":false,"title":"BMC Nephrology"},"publishedOn":"2024-11-26 15:58:06","publishedOnDateReadable":"November 26th, 2024"},"versionCreatedAt":"2024-04-22 02:06:38","video":"","vorDoi":"10.1186/s12882-024-03861-y","vorDoiUrl":"https://doi.org/10.1186/s12882-024-03861-y","workflowStages":[]},"version":"v1","identity":"rs-4249542","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4249542","identity":"rs-4249542","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}