BRIEF REVIEW www.jasn.org

Pragmatic Clinical Trials in CKD: Opportunities and Challenges

† ‡ | †† Ian H. de Boer,* Csaba P. Kovesdy, § Sankar D. Navaneethan, ¶ Carmen A. Peralta,** †† ‡‡ || Delphine S. Tuot, Miguel A. Vazquez, and Deidra C. Crews,§§ for the American Society of Nephrology Chronic Kidney Disease Advisory Group

*Division of Nephrology and †Department of Medicine, Kidney Research Institute, University of Washington, Seattle, WA; ‡Division of Nephrology, University of Tennessee Health Science Center, Memphis, TN; §Nephrology Section, Memphis VA Medical center, Memphis, TN; |Section of Nephrology, Department of Medicine, Selzman Institute for Kidney Health, Baylor College of Medicine; ¶Section of Nephrology, Michael E. DeBakey Veterans Affairs Medical Center, Houston, TX; **Kidney Health Research Collaborative and ††Division of Nephrology, University of California San Francisco, San Francisco, CA; ‡‡Division of Nephrology, Department of Internal Medicine, UT Southwestern Medical Center, Dallas, TX; §§Division of Nephrology, Department of Medicine, Johns Hopkins University School of Medicine; and ||Welch Center for Prevention, Epidemiology and Clinical Research, Johns Hopkins Medical Institutions, Baltimore, MD

ABSTRACT Randomized controlled trials in CKD lag in number behind those of other chronic demonstration of treatment efficacy and diseases, despite the high morbidity and mortality experienced by patients with kidney implementation of proven in disease and the exorbitant costs of their health care. Observational studies of CKD clinical care, and high-quality RCT evi- frequently yield seemingly paradoxic associations of traditional risk factors with dence has remained scarce for many im- outcomes, making it difficult to extrapolate the results of trials conducted in people portant and practical clinical questions. with normal kidney function to patients with CKD. However, many completed trials in Upon this general background, con- CKD have been limited by intermediate outcomes of unclear clinical significance or ducting investigations in CKD has faced narrow eligibility criteria that limit external validity, and implementation of proven additional challenges. The number of RCTs therapies remains a challenge. It is therefore imperative that the nephrology com- conducted in patients with CKD or ESRD munity capitalize on recent interest in novel approaches to trial design, such as prag- is one of the smallest compared with other matic clinical trials. These trials are meant to promote research within real world medical subspecialties,7 in spite of the ris- settings to yield clinically relevant results with greater applicability than those of ing prevalence of CKD, the enormously traditional trials, while maintaining many advantages, such as controlling for potential high morbidity and mortality experienced sources of bias. We provide a description of pragmatic clinical trials and a discussion by patients with kidney disease, and the of advantages, disadvantages, and practical challenges inherent to this study design, high costs of CKD and ESRD care.8,9 Pub- in the context of specific scientific questions relevant to patients with CKD. lished nephrology RCTs have often exam- ined intermediate outcomes that are of J Am Soc Nephrol 27: ccc–ccc, 2016. doi: 10.1681/ASN.2015111264 unclear significance to patients, providers, and families, with important exceptions. The complex and heterogeneous nature THE NEED FOR CLINICAL TRIALS After their ubiquitous acceptance as the of CKD has often led to restrictive enroll- IN CKD sine qua non of causal assessment, RCTs ment criteria that limit external validity.7 evolved from small single-center studies Inthesecondhalfofthepastcentury,sig- to large, multicenter trials. With this shift, nificant progress has been made in under- the cost of conducting RCTs increased dra- Published online ahead of print. Publication date standing principles of causal inference and matically, common disease states amena- available at www.jasn.org. in statistical techniques underlying the ble to drug intervention (such as hyper- Correspondence: Dr. Ian de Boer, Division of Ne- proper planning and analysis of interven- cholesterolemia and hypertension) phrology and Kidney Research Institute, University of Washington Medicine/Nephrology, Box 359606, 1 4 tional trials. As a result, the randomized became a dominant focus of inquiry, 325 9th Ave, Seattle, WA 98104. Email: deboer@u. controlled trial (RCT) has become the and commercial entities took on enlarging washington.edu fi 5,6 gold standard tool for proving the ef cacy roles in trial design and conduct. At the Copyright © 2016 by the American Society of and safety of therapeutic interventions.2,3 same time, gaps emerged between Nephrology

J Am Soc Nephrol 27: ccc–ccc, 2016 ISSN : 1046-6673/2710-ccc 1 BRIEF REVIEW www.jasn.org

Insufficient implementation of the few In reality, any single trial is neither fully intervention implementation, safety mon- proven therapies has contributed to in- pragmatic nor fully explanatory.15 The itoring, and outcome ascertainment.18,21 creasedriskofCKDanditscomplications PRagmatic Explanatory Continuum Indi- Interest in PCTs has grown quickly in economically and socially disadvan- cator Summary-2 tool identifies nine do- over the last decade,15 and the nephrol- taged populations.10,11 Moreover, observa- mains to quantitate the extent to which a ogy community has been an early leader tional studies of people with CKD and is pragmatic.20 The most prag- in this surge, as demonstrated by the ex- ESRD have frequently observed seemingly matic trials apply broad eligibility criteria, amples below. PCTs that involve popu- paradoxic associations of traditional risk recruit from clinical settings, conduct pro- lations with kidney disease have been factors with clinical outcomes,12 making cedures in the context of usual care using facilitated and encouraged through spe- it difficult to extrapolate the results of available clinical infrastructure, apply in- cific funding mechanisms from sponsors RCTs conducted in patients with normal terventions with flexible protocols using critical to nephrology research, including kidney function to patients with CKD. usual encouragement, assess outcomes rel- the National Institutes of Health, Depart- It is therefore imperative that the evant to patients using readily obtainable ment of Veterans Affairs, and the Patient nephrology community capitalize on the metrics, and apply intention-to-treat anal- Centered Institute. Of recent wave of interest in novel approaches ysis (Figure 1). Rarely is a clinical trial note, interest in PCTs has grown along to clinical trial design, such as pragmatic pragmatic in all aspects of its design. Of- with interest in another movement that clinical trials (PCTs). This review article ten, some design elements are made less may appear conflicting – the Precision will provide a description of the concept of pragmatic in order to maintain internal Medicine Initiative.22 Indeed, whereas PCTs, a discussion of the advantages, dis- validity or to accommodate practical needs PCTs focus on applicability to popula- advantages, and practical challenges of for implementation. Ultimately, the choice tions, precision medicine focuses on ap- PCTs, and examples of PCTs conducted of design for any clinical trial depends on plicability to individuals. However, there among patients with CKD. the underlying scientific question and the is no reason that trials advancing preci- context in which it is addressed. sion medicine cannot incorporate prag- PCTs are suited for the evaluation of a matic design elements, and PCTs can in- WHAT ARE PCTS? wide variety of medical interventions in- form precision medicine by building in cluding individual medical treatments and personalized components (e.g., inter- PCTs have been recognized as important strategies to deliver those treatments to vention flexibility) or the evaluation of tools for evaluating medical interventions patient populations. Depending on the between-participant heterogeneity in since at least 1967.13 Over time, the term intervention and setting, PCTs may em- response. “pragmatic” has been used to refer to a va- ploy cluster randomization, active com- riety of interrelated elements of trial de- parator groups, and quasi-experimental sign.14–16 Central to all definitions is an designs. The electronic health record POTENTIAL BENEFITS OF PCT emphasis on external validity. PCTs are de- (EHR) plays a central role in most PCTs. DESIGNS IN CKD signed so that their results can be quickly EHR data can efficiently facilitate every and directly applied to relevant clinical step of a clinical trial, from screening to There are several potential benefits to populations. Toward this goal, common enrollment, collection of baseline data, PCTs in CKD (Table 1). First, PCTs may features of PCTs include broad eligibility criteria, comparisons of clinically relevant alternatives, integration of research into clinical practice, and evaluation of a broad range of relevant health outcomes.17,18 The Gruppo Italiano per lo Studio della Strep- tochinasi nell’Infarto Miocardico trial of thrombolysis for acute myocardial infarc- tion is often cited as an example of an early and successful PCT.19 PCTs have been con- trasted with explanatory trials, which strive to determine whether an intervention has beneficial effects in humans, often using conditions that maximize the likelihood of demonstrating a treatment effect while narrowing generalizability.13–16 Some au- thors have compared PCTs to effectiveness evaluation and explanatory trials to efficacy Figure 1. Domains of the explanatory-pragmatic clinical trial design spectrum. evaluation.14,15

2 Journal of the American Society of Nephrology J Am Soc Nephrol 27: ccc–ccc,2016 www.jasn.org BRIEF REVIEW

Table 1. Selected advantages and challenges for PCTs in nephrology unique challenges that have led to debate Advantages Challenges regarding optimal approaches and alter- Enhanced external validity Informed consent and regulatory oversight natives to the traditional written in- 36,37 Ability to enroll socially Achieving large separation between formed consent process. Written, disadvantaged populations comparison groups in person, informed consent may alter Directly applicable to patient care Selection and ascertainment of outcomes the nature of the intervention, introduce Relevant patient-centered outcomes Funding prohibitive logistic barriers, and lead to Reduced cost Lack of experience and training selection bias. These risks are particu- larly apparent for PCTs that employ clus- include a broad range of patients with stakeholders, including patients, care- ter randomization, i.e., assigning groups CKD, enhancing external validity. Pa- givers, and providers, as high-priority of participants rather than individuals to tients with CKD have diverse underlying CKD research areas (Box 1).34,35 Out- treatment arms. With the increasing use and comorbid conditions and often have comes of interest to these stakeholders of EHRs, an electronic consenting pro- 38 complicated medical regimens that may include traditional clinical events such cess can be considered to reduce costs. be more acceptable for PCTs than tradi- as hospitalizations, emergency room However, such a process may introduce tional RCTs with strict eligibility criteria. visits, and mortality as well as patient- selection bias, as target populations In addition, PCTs may include partici- centered outcomes of growing interest, would necessarily be literate and have pants from socially disadvantaged pop- such as symptoms, treatment prefer- access to information technology, and ulations that bear a disproportionate ences, quality of life, and satisfaction recent national surveys have suggested a general preference for an interactive burden of disease and experience large with care. Use of the EHR can effectively written consent process rather than al- translational gaps between research and ascertain clinical events, although fur- ternatives, such as informing pa- practice,10,11,23 who may be excluded ther planning is required to collect pa- tients that a clinical site is participating from efficacy trials owing to a lack of re- tient-reported outcome data, which are 21 in a PCT.39,40 liable transportation or limited English not currently available in most EHRs. language proficiency or may choose not Fourth, pragmatic trials for CKD may fi to enroll in efficacy studies because of be less costly than ef cacy trials. This is Regulatory Issues mistrust in the research process.24,25 So- particularly important when consider- Institutional review boards (IRBs) serve cially disadvantaged patients may be ing the length of follow-up necessary to to protect the safety of study subjects, and more willing to participate in the context observe some clinicaloutcomesof CKD, each has different rules and regulations of their usual site of health care26;re- such as progression to ESRD. to account for local conditions of risk. cruiting and enrollment data from early Multicenter PCTs require harmonization PCTs will help address this possibility. of these local regulations to ensure par- Second,findingsfromPCTscanoftenbe BARRIERS AND OBSTACLES FOR ticipant safety and data integrity. This directly applied to the practice environment PCTS IN CKD process may be facilitated through use of central IRBs with shared review,41 (s) in which the study was conducted. This Informed Consent but confusion regarding divisions of re- helps translate research results to clinical Informed consent is a cornerstone of sponsibility for collaborating central care, lowering well known barriers to the ethical clinical research, but PCTs pose and local IRBs remains a barrier for implementation of existing practice guide- lines for care of patients with CKD.27,28 The majority of CKD patients are cared for by EXAMPLES OF RESEARCH QUESTIONS IDENTIFIED BY DIVERSE primary care providers with or without STAKEHOLDERS AS HIGH-PRIORITY CKD RESEARCH AREAS collaboration with nephrologists,8,29 and What is the optimal screening frequency in populations at high-risk of de- PCTs may be well positioned to affect veloping CKD? CKD care in the primary care setting. Ex- How effective are lifestyle programs for preventing deteriorating kidney amples of studies bridging nephrology and function in early CKD? primary care include trials of CKD care co- What is the effectiveness of strategies to improve provider awareness and ordination with patient navigators30 adherence to guidelines on improving outcomes for patients with CKD? and strategies to implement guideline- What is the effectiveness of patient-safety interventions in reducing compli- concordant CKD care with electronic cations among CKD patients? checklists,31 primary care practice facilita- What is the effectiveness of computer decision support for CKD management in tion,32 and pharmacist-led interventions.33 reducing complications? Third, pragmatic trials offer an op- What strategies will improve family consent for deceased donor kidney portunity to examine clinical questions donation, taking different cultural groups into account? that have been identified by diverse

J Am Soc Nephrol 27: ccc–ccc, 2016 Pragmatic Clinical Trials in CKD 3 BRIEF REVIEW www.jasn.org widespread adoption.18 Moreover, the particularly if study participants have lit- (NCT02059408). In the usual care arm, concepts and design elements of PCTs tle contact with dedicated research staff. no systematic screening for CKD is are new to many IRBs and data and safety PCTs may also increase burden to prac- recommended. The first intervention eval- monitoring boards, who may expect, re- ticing providers, requiring effective strat- uates the efficacy of a screen-and-educate quest, or even demand traditional clini- egies to increase provider engagement strategy that utilizes a “triple-marker” cal trial design elements that reduce and participation without compromising CKD screening approach (creatinine, pragmatic elements of PCTs. Contribut- the integrity and “pragmatism” of the cystatin C, and albuminuria44) coupled ing to this tension, it may be difficult to study. In addition, PCTs in nephrology with provider education, compared define appropriate standards of care for are likely to require partnerships with with usual care. This strategy takes patients with CKD because many guide- large, integrated health systems, and it advantage of the VA EHR to identify eli- lines may not offer clear consensus or be may be difficult to harmonize protocols gible persons, order screening tests, and based on limited evidence.42 PCTs seek- while maintaining local care priorities deliver recommendations to primary ing to challenge traditional care practices when there are substantial differences in care providers. The second intervention may have difficulty defining an appropri- workflow, EHR, personnel, and clinic evaluates whether a screen, educate, and ate control group or gaining IRB or data culture. Nephrology experience with intensify treatment program coled by a and safety monitoring board approval PCTs is nascent, so many implementa- primary care provider and a clinical for a design that involves more flexible tion barriers will be new to investigative pharmacist can improve BP management protocols or less extensive research-spe- teams. among persons with CKD, compared cific monitoring than traditional explan- with the screen-and-educate provider atory trials. Novel Therapeutics strategy and with usual care. The primary Novel therapeutic approaches are sorely outcome of the study is change in BP and Outcome Ascertainment needed to reduce the incidence and pro- secondary outcomes include processes of Ideally, PCTs identify outcomes using gression of CKD and its complications. care and clinician burden. This prag- available clinical data sources, e.g., EHRs, The initial evaluation of new drugs is matic study design can be easily trans- to enhance relevance and promote effi- often best accomplished using more lated to any primary care practice with ciency. Challenges for use of such data traditional RCTs that feature explanatory an EHR that utilizes the chronic care include lack of specificity for coded clin- design elements. For example, extensive model. ical events, potentially leading to mis- eligibility criteria may be used to limit classification of study outcomes, and participation to a subset of patients more The Kidney Awareness Registry and missing or biased outcomes data. For ex- likely to benefit and less likely to experi- Education Study ample, laboratory outcomes, such as ence adverse events than the broader The Kidney Awareness Registry and Ed- changes in slope of eGFR, serum phos- target population, and extra efforts may ucation (KARE) study (NCT01530958) phorus, or hemoglobin can be strongly be used to enhance adherence and main- is a two by two factorial trial that assesses influenced by loss to follow-up, and clin- tain separation of treatment groups. PCTs the effectiveness of a multilevel interven- ical events can be difficult to ascertain if may be more suitable for subsequent tion to improve BP control among low- participants change health care systems. studies in broader populations, for drugs income patients with CKD (Table 2). The Outcomes that are ascertained from na- thathave known efficacy andarerelatively study design relies on two levels of ran- tional databases, such as initiation of safe, and for studies that evaluate imple- domization: primary care practice teams dialysis or transplantation, may not mentation of existing therapies or health are randomized to one of two arms (ac- be affected to the same degree. Patient- care delivery. cess to an electronic CKD registry or reported outcomes, such as health- usual care); patients of each practice related quality of life measures, medication team are subsequently randomized to adherence, and medication side effects, EXAMPLES OF ONGOING participate in a CKD self-management are difficult to track pragmatically, as PRAGMATIC TRIALS IN CKD program or usual care. With four study they are not often ascertained as part arms, KARE allows assessment of the of routine clinical care.21 Screening for CKD individual and additive effect of both in- The effectiveness of screening for CKD terventions. Although the KARE study Implementation Barriers is currently being evaluated in a high design is complex, its pragmatic elements As with traditional RCTs, sufficient sep- risk primary care population of United include: setting and participants, inter- aration in treatment between compari- States veterans (Table 2). Using cluster- vention components (electronic CKD son groups is needed to detect differences randomization by primary care team, registry embedded in the EHR and a pa- in clinical outcomes.43 For PCTs, large this pilot PCT aims to evaluate two tient self-management support program separations between groups may be dif- strategies to improve BP control among that relies on language-concordant ficult to achieve in a clinical setting nondiabetic hypertensive veterans with health coaches), outcome ascertainment with large commonalities in practice, unrecognized CKD in primary care (change in systolic BP and proportion of

4 Journal of the American Society of Nephrology J Am Soc Nephrol 27: ccc–ccc,2016 www.jasn.org BRIEF REVIEW

Table 2. Selected examples of ongoing PCTs in CKD Study Intervention Outcomes(s) Pragmatic Aspects Barriers and Obstacles Screening Usual care versus BP levels, change in VA clinical practice Informed consent for CKD screen-educate BP: proportion versus screen- with BP controlled educate-intensify BP Processes of care, Wide inclusion criteria Ensuring screening after lab orders management by appropriate use of placed clinical pharmacist ACE/ARB Utilization of EHR for patient identification, Database design for study delivery of interventions and outcome management ascertainment Use of ancillary health personnel embedded in primary care KARE Provider intervention: Change in systolic Set in “real world” primary care practices Accurate outcome ascertainment CKD registry versus BP; proportion of usual care patients with BP Patient intervention: control Utilization of EHR for outcome language concordant ascertainment self-management Recruitment of study participants that are program versus usual not well represented in RCTs. care ICD-Pieces CKD care enhanced by Hospitalizations Broad eligibility criteria Data transmission from multiple information Cardiovascular sites technology, clinical events, Delivery of the intervention in clinical Implementation of various support and practice readmissions, ER setting by primary care providers interventions facilitator versus usual visits, deaths Electronic ascertainment of outcomes Changes in standard clinical care Intention-to treat analysis based on practices assigned cluster ECHO-CKD Calcitriol plus usual care Composite of Integration into clinical health care system Efficient informed consent versus usual care clinical Minimal eligibility criteria Recruiting alone cardiovascular Utilization of EHR for recruiting and Retention and adherence events outcome ascertainment Clinical events outcome Accurate outcome ascertainment ACE/ARB, XXX; VA, XXX; ER, XXX. patients with BP control), and relevance hospital readmissions and improve in a pilot stage, designed to test the to practice. By relying on existing infra- clinical outcomes,46,47 in a cluster- cardiovascular effects of calcitriol, a drug structure and recruiting actively empan- randomized design. Study outcomes in- commonly used in CKD for which long- eled patient participants, KARE results clude hospitalizations, readmissions, term clinical risks and benefits remain will be directly applicable to populations cardiovascular events, ER visits, and uncertain (Table 2).48,49 The overall goal that receive CKD care in similar health death. Other pragmatic characteristics of the ECHO-CKD trial is to determine systems. of ICD-Pieces are broad inclusion cri- whether calcitriol plus usual care, com- teria, flexibility of protocols, imple- pared with usual care alone, reduces the Improving Chronic Disease mentation of the interventions within risk of clinical cardiovascular events Management with Pieces the context of available clinical infra- among patients with stage 3–4CKD.Al- Improving Chronic Disease Manage- structures, robust clinical support though this question could be addressed ment with Pieces (ICD-Pieces) is a large tools, electronic ascertainment of out- using a more traditional explanatory PCT spanning several large health care comes, and intention-to-treat analysis trial, a pragmatic approach facilitates systems that will compare EHR-based basedonassignedcluster.Thegoalof application of results and reduces cost. clinical-decision support to primary care this PCT is to evaluate a promising new Pragmatic features of the planned study providers and practice facilitators with model for the care of CKD patients with include integration into a large compre- usual care among patients with coexis- multimorbidity. hensive health care system, utilization of tent CKD, diabetes, and hypertension electronic resources to identify partici- (Table 2).45 The study uses a novel tech- Evaluating Calcitriol Heart pants, the application of minimal eligi- nology platform, Pieces, which has been Outcomes in CKD bility criteria, delivery and monitoring successfully used in other chronic condi- Evaluating Calcitriol Heart Outcomes in of the study intervention using clinical tions, such as heart failure, to reduce CKD (ECHO-CKD) is a PCT, currently pharmacy and laboratory services, and

J Am Soc Nephrol 27: ccc–ccc, 2016 Pragmatic Clinical Trials in CKD 5 BRIEF REVIEW www.jasn.org the ascertainment of clinical outcomes patients with CKD with high external (Ridgefield, CT), and Janssen (Titusville, NJ). S.N. of importance to patients and providers. validity and applicability and low costs. has served on the event adjudication committee of However, informed consent, regulatory clinical trials sponsored by AbbVie (North Chicago, issues, outcome ascertainment, and other IL), Bayer, and Boehringer-Ingelheim. OTHER PCTS IN NEPHROLOGY implementation barriers provide chal- lenges to designing and implementing REFERENCES PCTsmay also be useful to improve care in PCTs.If barriers to PCTscan be surmoun- fi dialysis, glomerular disease, kidney trans- ted and bene ts realized, then further 1. Hill AB: The Environment and Disease: As- plantation, hypertension, and other fields training, infrastructure, and funding will sociation or Causation? Proc R Soc Med 58: within nephrology. As an example, the be needed to facilitate additional studies. 295–300, 1965 TiME trial (NCT02019225) evaluates the Few investigators in nephrology have sub- 2. Organization, review, and administration of effects of a minimum hemodialysis ses- stantial experience in PCTs, and training cooperative studies (Greenberg Report): a report from the Heart Special Project Com- sion duration of 4.25 hours compared will be necessary to advance and imple- mittee to the National Advisory Heart with usual care on mortality, hospitaliza- ment high-quality PCTmethods. Because Council, May 1967. Control Clin Trials 9: tions, and health-related quality of life PCTsarecriticallydependent onthe health 137–148, 1988 among patients with ESRD initiating care setting in which they are based, 3. Concato J, Shah N, Horwitz RI: Randomized, treatment with thrice weekly maintenance investment in EHRs and partnerships of controlled trials, observational studies, and the hierarchy of research designs. NEnglJ hemodialysis. This PCTuses a cluster ran- researchers with health care systems will be Med 342: 1887–1892, 2000 domization design, with the allocation of required to expand bandwidth. For exam- 4. DiMasi JA, Hansen RW, Grabowski HG: The outpatient dialysis units to the two inter- ple, within the Veterans Health Adminis- price of innovation: new estimates of drug vention arms. Another novel aspect of the tration, a national coordinating center has development costs. JHealthEcon22: 151– TiME trial is that it is conducted through a been developed to help design and launch 185, 2003 5. Clancy C, Collins FS: Patient-Centered Out- partnership between academic investiga- PCTs. Currently, the Patient-Centered comes Research Institute: the intersection of tors and two large dialysis provider orga- Outcomes Research Institute and several science and health care. Sci Transl Med 2: nizations in approximately 320 dialysis institutes within the National Institutes of 37cm18, 2010 facilities, thereby combining traditional Health have offered funding mechanisms 6. DeMets DL, Califf RM: A historical perspec- funding mechanisms with existing indus- specifictoPCTs;additionalfundingmech- tive on clinical trials innovation and leader- ship: where have the academics gone? JAMA try resources and expertise. The TiME tri- anisms appropriate for this study design 305: 713–714, 2011 al is planning to enroll approximately will be required. Ongoing studies, includ- 7. Inrig JK, Califf RM, Tasneem A, Vegunta RK, 6432 patients, and it is estimated to be ing those described above, will provide Molina C, Stanifer JW, Chiswell K, Patel UD: completed in September 2017. In glomer- valuable information on the feasibility, The landscape of clinical trials in nephrology: ular disease, NephCure has brought to- strengths, limitations, and hurdles ofPCTs a of Clinicaltrials.gov. Am J Kidney Dis 63: 771–780, 2014 gether patients, families, and researchers in CKD, as well as additional next steps 8.SaranR,LiY,RobinsonB,AyanianJ, to support patient-oriented research well required to move forward this promising Balkrishnan R, Bragg-Gresham J, Chen JT, suited to PCTs. In kidney transplantation, approach to clinical trials. Cope E, Gipson D, He K, Herman W, Heung many interventions are made using pro- M, Hirth RA, Jacobsen SS, Kalantar-Zadeh K, Kovesdy CP, Leichtman AB, Lu Y, Molnar MZ, tocols starting from a clear time zero, and ’ ACKNOWLEDGMENTS Morgenstern H, Nallamothu B, O Hare AM, alternative protocols may be effectively Pisoni R, Plattner B, Port FK, Rao P, Rhee CM, compared using pragmatic trial designs. Schaubel DE, Selewski DT, Shahinian V, Sim Hypertension is a highly prevalent condi- The San Francisco VA Medical Center and JJ, Song P, Streja E, Kurella Tamura M, tion with many available treatments University of California San Francisco Tentori F, Eggers PW, Agodoa LY, Abbott for which BP targets, specific antihyperten- CKD screening study is supported by grant KC. US Renal Data System 2014 Annual Data R34DK102152 from the National Institute Report: Epidemiology of Kidney Disease in sive agents, and antihypertensive combi- the United States. AmJKidneyDis66: S1- nations or approaches could be compared for Diabetes and Digestive and Kidney 305, 2015 using pragmatic designs. The Living Diseases (NIDDK). KARE is supported by 9. Després JP, Nadeau A, Tremblay A, Ferland Textbook of PCTs, sponsored by the grants R34DK093992 and R01DK104130 M, Moorjani S, Lupien PJ, Thériault G, Pinault National Institutes of Health Collabo- from the NIDDK. ICD-Pieces is supported S, Bouchard C: Role of deep abdominal fat by grant UH3DK104655 from the NIDDK. in the association between regional adipose ratory, further describes ongoing PCTs tissue distribution and glucose tolerance in 18 withinandbeyondnephrology. The ECHO-CKD study is supported by obese women. Diabetes 38: 304–309, 1989 UH2HL125122 from the National Heart, 10. Tarver-Carr ME, Powe NR, Eberhardt MS, Lung, and Blood Institute. LaVeist TA, Kington RS, Coresh J, Brancati NEXT STEPS FOR PCTS IN CKD FL: Excess risk of chronic kidney disease among African-American versus white sub- DISCLOSURES jects in the United States: a population- PCTsoffer a valuable approach to increase I.D.B. has received consultancy fees from Amgen, based study of potential explanatory factors. the evidence base for the treatment of Bayer (Whippany, NJ), Boehringer-Ingelheim J Am Soc Nephrol 13: 2363–2370, 2002

6 Journal of the American Society of Nephrology J Am Soc Nephrol 27: ccc–ccc,2016 www.jasn.org BRIEF REVIEW

11. Volkova N, McClellan W, Klein M, Flanders Rocco MV, Choi MJ; National Kidney Foun- 39. Nayak RK, Wendler D, Miller FG, Kim SY: D, Kleinbaum D, Soucie JM, Presley R: dation: Perceptions and use of the national Pragmatic Randomized Trials Without Stan- Neighborhood poverty and racial differ- kidney foundation KDOQI guidelines: a sur- dard Informed Consent?: A National Survey. ences in ESRD incidence. JAmSocNephrol vey of U.S. renal healthcare providers. BMC Ann Intern Med 163: 356–364, 2015 19: 356–364, 2008 Nephrol 14: 230, 2013 40. Cho MK, Magnus D, Constantine M, Lee SS, 12. Kovesdy CP, Anderson JE: Reverse epide- 28. Abdel-Kader K, Greer RC, Boulware LE, Kelley M, Alessi S, Korngiebel D, James C, miology in patients with chronic kidney dis- Unruh ML: Primary care physicians’ familiar- Kuwana E, Gallagher TH, Diekema D, Capron ease who are not yet on dialysis. Semin Dial ity, beliefs, and perceived barriers to practice AM, Joffe S, Wilfond BS: Attitudes Toward 20: 566–569, 2007 guidelines in non-diabetic CKD: a survey Risk and Informed Consent for Research on 13. Schwartz D, Lellouch J: Explanatory and study. BMC Nephrol 15: 64, 2014 Medical Practices: A Cross-sectional Survey. pragmatic attitudes in therapeutical trials. J 29. Diamantidis CJ, Powe NR, Jaar BG, Greer Ann Intern Med 162: 690–696, 2015 Chronic Dis 20: 637–648, 1967 RC, Troll MU, Boulware LE: Primary care- 41. Sugarman J, Califf RM: Ethics and regulatory 14. Macpherson H: Pragmatic clinical trials. specialist collaboration in the care of patients complexities for pragmatic clinical trials. Complement Ther Med 12: 136–140, 2004 with chronic kidney disease. Clin J Am Soc JAMA 311: 2381–2382, 2014 15. Patsopoulos NA: A pragmatic view on prag- Nephrol 6: 334–343, 2011 42. Reisin E, Harris RC, Rahman M: Commentary matic trials. Dialogues Clin Neurosci 13: 30. Jolly SE, Navaneethan SD, Schold JD, Arrigain on the 2014 BP guidelines from the panel 217–224, 2011 S, Konig V, Burrucker YK, Hyland J, Dann P, appointed to the Eighth Joint National 16. GodwinM,RuhlandL,CassonI,MacDonaldS, Tucky BH, Sharp JW, Nally JV: Development Committee (JNC 8). J Am Soc Nephrol 25: Delva D, Birtwhistle R, Lam M, Seguin R: Prag- of a chronic kidney disease patient navigator 2419–2424, 2014 matic controlled clinical trials in primary care: program. BMC Nephrol 16: 69, 2015 43. Parfrey PS, Block GA, Correa-Rotter R, the struggle between external and internal 31. Mendu ML, Schneider LI, Aizer AA, Singh K, Drüeke TB, Floege J, Herzog CA, London validity. BMC Med Res Methodol 3: 28, 2003 Leaf DE, Lee TH, Waikar SS: Implementation GM, Mahaffey KW, Moe SM, Wheeler DC, 17. Tunis SR, Stryer DB, Clancy CM: Practical of a CKD checklist for primary care providers. Chertow GM: Lessons Learned from EVOLVE clinical trials: increasing the value of clinical Clin J Am Soc Nephrol 9: 1526–1535, 2014 for Planning of Future Randomized Trials in research for decision making in clinical and 32. Fox CH, Vest BM, Kahn LS, Dickinson LM, Patients on Dialysis. Clin J Am Soc Nephrol . JAMA 290: 1624–1632, 2003 Fang H, Pace W, Kimminau K, Vassalotti J, 11: 539–546, 2016 18. 2016. Available at: http://sites.duke.edu/ Loskutova N, Peterson K: Improving evi- 44. Peralta CA, Shlipak MG, Judd S, Cushman rethinkingclinicaltrials/. Accessed March 17, dence-based primary care for chronic kidney M, McClellan W, Zakai NA, Safford MM, 2016 disease: study protocol for a cluster ran- Zhang X, Muntner P, Warnock D: Detection 19. Effectiveness of intravenous thrombolytic domized control trial for translating evidence of chronic kidney disease with creatinine, treatment in acute myocardial infarction. into practice (TRANSLATE CKD). Implement cystatin C, and urine albumin-to-creatinine Gruppo Italiano per lo Studio della Strepto- Sci 8: 88, 2013 ratio and association with progression to end- chinasi nell’Infarto Miocardico (GISSI). Lan- 33. Cooney D, Moon H, Liu Y, Miller RT, stage renal disease and mortality. JAMA 305: cet 1: 397–402, 1986 Perzynski A, Watts B, Drawz PE: A pharmacist 1545–1552, 2011 20. Loudon K, Treweek S, Sullivan F, Donnan P, based intervention to improve the care of 45. Fujioka S, Matsuzawa Y, Tokunaga K, Tarui S: Thorpe KE, Zwarenstein M: The PRECIS-2 patients with CKD: a pragmatic, randomized, Contribution of intra-abdominal fat accumu- tool: designing trials that are fit for purpose. controlled trial. BMC Nephrol 16: 56, 2015 lation to the impairment of glucose and lipid BMJ 350: h2147, 2015 34. Crews DC, Greer RC, Fadrowski JJ, Choi MJ, metabolism in human obesity. Metabolism 21. Drawz PE, Archdeacon P, McDonald CJ, Doggett D, Segal JB, Fawole KA, Crawford PR, 36: 54–59, 1987 Powe NR, Smith KA, Norton J, Williams DE, Boulware LE: Setting an agenda for compara- 46. Amarasingham R, Moore BJ, Tabak YP, Patel UD, Narva A: CKD as a Model for Im- tive effectiveness systematic reviews in CKD Drazner MH, Clark CA, Zhang S, Reed WG, proving Chronic Disease Care through Elec- care. BMC Nephrol 13: 74, 2012 Swanson TS, Ma Y, Halm EA: An automated tronic Health Records. Clin J Am Soc 35. Tong A, Crowe S, Chando S, Cass A, model to identify heart failure patients at risk Nephrol 10: 1488–1499, 2015 Chadban SJ, Chapman JR, Gallagher M, for 30-day readmission or death using elec- 22. Collins FS, Varmus H: A new initiative on Hawley CM, Hill S, Howard K, Johnson DW, tronic medical record data. Med Care 48: precision medicine. N Engl J Med 372: 793– Kerr PG, McKenzie A, Parker D, Perkovic V, 981–988, 2010 795, 2015 Polkinghorne KR, Pollock C, Strippoli GF, 47. Amarasingham R, Velasco F, Xie B, Clark C, 23. Schillinger D. Literacy and health communi- Tugwell P, Walker RG, Webster AC, Wong G, Ma Y, Zhang S, Bhat D, Lucena B, Huesch M, cation: reversing the ’inverse care law’. Am J Craig JC: Research Priorities in CKD: Report Halm EA: Electronic medical record-based Bioeth 7:15–18, 2007 of a National Workshop Conducted in Aus- multicondition models to predict the risk of 24. Farmer DF, Jackson SA, Camacho F, Hall MA: tralia. Am J Kidney Dis 66: 212–222, 2015 30 day readmission or death among adult Attitudes of African American and low so- 36. Platt R, Kass NE, McGraw D: Ethics, regula- medicine patients: validation and compari- cioeconomic status white women toward tion, and comparative effectiveness research: son to existing models. BMC Med Inform medical research. JHealthCarePoorUn- time for a change. JAMA 311: 1497–1498, Decis Mak 15: 39, 2015 derserved 18: 85–99, 2007 2014 48. Bosworth C, de Boer IH: Impaired vitamin D 25. Scharff DP, Mathews KJ, Jackson P, 37. Pletcher MJ, Lo B, Grady D: Informed consent metabolism in CKD. Semin Nephrol 33: 158– Hoffsuemmer J, Martin E, Edwards D: More in randomized quality improvement trials: a 168, 2013 than Tuskegee: understanding mistrust about critical barrier for learning health systems. 49. Ketteler M, Elder GJ, Evenepoel P, Ix JH, research participation. J Health Care Poor JAMA Intern Med 174: 668–670, 2014 Jamal SA, Lafage-Proust MH, Shroff R, Underserved 21: 879–897, 2010 38. Chalil Madathil K, Koikkara R, Obeid J, Thadhani RI, Tonelli MA, Kasiske BL, Wheeler 26. UyBico SJ, Pavel S, Gross CP: Recruiting Greenstein JS, Sanderson IC, Fryar K, DC, Leonard MB: Revisiting KDIGO clinical vulnerable populations into research: a sys- Moskowitz J, Gramopadhye AK: An in- practice guideline on chronic kidney disease- tematic review of recruitment interventions. vestigation of the efficacy of electronic con- mineral and bone disorder: a commentary J Gen Intern Med 22: 852–863, 2007 senting interfaces of research permissions from a Kidney Disease: Improving Global 27. Estrella MM, Jaar BG, Cavanaugh KL, Fox management system in a hospital setting. Int Outcomes controversies conference. Kidney CH, Perazella MA, Soman SS, Howell E, J Med Inform 82: 854–863, 2013 Int 87: 502–528, 2015

J Am Soc Nephrol 27: ccc–ccc, 2016 Pragmatic Clinical Trials in CKD 7