Challenges and opportunities in interventions for chronic kidney disease of unknown origin (CKDu): report from the International Society of Nephrology Consortium of Collaborators on CKDu

Chronic kidney disease of unknown origin (CKDu) is a progressive tubulointerstitial nephropathy reported principally in agricultural communities, specifically manual laborers in dry, lowland regions of Central America, Sri Lanka, and Southern India.1Anand S. Caplin B. Gonzalez-Quiroz M. et al.Epidemiology, molecular, and genetic methodologies to evaluate causes of CKDu around the world: report of the Working Group from the ISN International Consortium of Collaborators on CKDu.Kidney Int. 2019; 96: 1254-1260Abstract Full Text Full Text PDF PubMed Scopus (16) Google Scholar,2Caplin B. Yang C.W. Anand S. et al.The International Society of Nephrology's International Consortium of Collaborators on Chronic Kidney Disease of Unknown Etiology: report of the working group on approaches to population-level detection strategies and recommendations for a minimum dataset.Kidney Int. 2019; 95: 4-10Abstract Full Text Full Text PDF PubMed Scopus (40) Google Scholar Further research may reveal whether it is present, but unrecognized, in other regions, including in parts of South East Asia, Africa, and the United States. The disease was first described 2 decades ago, and although scientists largely agree on common clinical characteristics, much of the pathophysiology remains unclear. A leading hypothesis is occupational exposure to recurrent heat stress leading to recurrent acute kidney injury (AKI), the risk for which may increase in parallel with increasing global temperatures.3Glaser J. Lemery J. Rajagopalan B. et al.Climate change and the emergent epidemic of CKD from heat stress in rural communities: the case for heat stress nephropathy.Clin J Am Soc Nephrol. 2016; 11: 1472-1483Crossref PubMed Scopus (246) Google Scholar Other hypothesized factors include environmental toxins, genetic predisposition, and dietary or pharmaceutical exposures, or some combination thereof. Persisting uncertainty has led researchers to focus on pathogenesis and epidemiology. However, ongoing disease burden, especially given a lack of access to effective treatments, and the potential for affected populations to experience research fatigue, demands that efforts to prevent and treat CKDu are undertaken as a priority. Interventional study designs that test preventative strategies and focus on addressing the concerns of affected populations offer a path forward. Although challenging to design when etiological uncertainty persists, interventional studies can be a strong test of a causal hypothesis and such studies may also advance our understanding of CKDu pathophysiology while potentially benefiting affected populations. In 2016, the International Society of Nephrology convened the International Consortium of Collaborators on Chronic Kidney Disease of Unknown Etiology, which has reported on disease detection strategies2Caplin B. Yang C.W. Anand S. et al.The International Society of Nephrology's International Consortium of Collaborators on Chronic Kidney Disease of Unknown Etiology: report of the working group on approaches to population-level detection strategies and recommendations for a minimum dataset.Kidney Int. 2019; 95: 4-10Abstract Full Text Full Text PDF PubMed Scopus (40) Google Scholar and on methods to elucidate the cause of CKDu.1Anand S. Caplin B. Gonzalez-Quiroz M. et al.Epidemiology, molecular, and genetic methodologies to evaluate causes of CKDu around the world: report of the Working Group from the ISN International Consortium of Collaborators on CKDu.Kidney Int. 2019; 96: 1254-1260Abstract Full Text Full Text PDF PubMed Scopus (16) Google Scholar Recognizing the urgent need for preventative and disease-modifying therapies, International Consortium of Collaborators on Chronic Kidney Disease of Unknown Etiology presents this commentary as a starting point for researchers seeking to mitigate the burden of CKDu. We acknowledge the challenges inherent to interventional research in this area, while emphasizing the need for unbiased evidence and the various study designs and other solutions that can be employed. Interventional studies target specific levels of disease prevention: primary prevention studies target hazard(s) to prevent disease onset, secondary prevention studies enroll participants with early evidence of disease (usually by screening) at a stage where intervention(s) can cure or prevent further progression, whereas tertiary prevention studies attempt to reduce the impact of illness or injury and associated disability. The conduct of interventional studies targeting CKDu (at any level of prevention) faces substantial—although not insurmountable—challenges. Affected communities are often marginalized and have minimal available health resources. In addition, the natural history of CKDu is poorly understood, and potentially important exposures could occur early in life or in several different environmental domains, or both. There are no established surrogate markers for disease progression in humans, and the relatively long follow-up times required to observe sustained decline in glomerular filtration rate may be more difficult to sustain in low-resource and potentially itinerant populations. Indeed, short-term changes in serum creatinine may be particularly important given that one purported causal mechanism for CKDu is recurrent episodes of severe AKI, making prevention of AKI a key potential target of intervention. Note that although AKI may offer a shorter-term end point for interventional studies, its detection may require intensive surveillance in a community-based study. Ultimately, researchers' choice of study end point is best determined as the most meaningful measure likely to respond to the proposed intervention within the available follow-up time. The importance of key study outcomes should be supported by existing observational or experimental data and, if possible, explored by long-term follow-up of study participants for incident CKDu. To date, 3 prospective, nonrandomized, interventional trials have been published, finding reductions in decline in kidney function over periods ranging from 3 weeks to 18 months (Table 1).4Krisher L. Butler-Dawson J. Yoder H. et al.Electrolyte beverage intake to promote hydration and maintain kidney function in Guatemalan sugarcane workers laboring in hot conditions.J Occup Environ Med. 2020; 62: e696-e703Crossref PubMed Scopus (10) Google Scholar, 5Glaser J. Hansson E. Weiss I. et al.Preventing kidney injury among sugarcane workers: promising evidence from enhanced workplace interventions.Occup Environ Med. 2020; 77: 527-534Crossref PubMed Scopus (41) Google Scholar, 6Glaser J. Wegman D.H. Arias-Monge E. et al.Workplace intervention for heat stress: essential elements of design, implementation, and assessment.Int J Environ Res Public Health. 2022; 19: 3779Crossref PubMed Scopus (5) Google Scholar, 7Siriwardhana E. Perera P.A.J. Sivakanesan R. et al.Diminished disease progression rate in a chronic kidney disease population following the replacement of dietary water source with quality drinking water: a pilot study.Nephrology (Carlton). 2018; 23: 430-437Crossref PubMed Scopus (7) Google Scholar, 8Selvarajah M. Mendis S. Jayasinghe S. et al.Randomized controlled trial of treatment of chronic kidney disease of uncertain aetiolgy with enalapril.J Clin Toxicol. 2016; 6: 281Google Scholar Two are examples of primary prevention with a package of health and behavioral education (Box 1),4Krisher L. Butler-Dawson J. Yoder H. et al.Electrolyte beverage intake to promote hydration and maintain kidney function in Guatemalan sugarcane workers laboring in hot conditions.J Occup Environ Med. 2020; 62: e696-e703Crossref PubMed Scopus (10) Google Scholar and improved working conditions.5Glaser J. Hansson E. Weiss I. et al.Preventing kidney injury among sugarcane workers: promising evidence from enhanced workplace interventions.Occup Environ Med. 2020; 77: 527-534Crossref PubMed Scopus (41) Google Scholar The other is a tertiary prevention study of replacement of usual water source with bottled water for those with CKDu.7Siriwardhana E. Perera P.A.J. Sivakanesan R. et al.Diminished disease progression rate in a chronic kidney disease population following the replacement of dietary water source with quality drinking water: a pilot study.Nephrology (Carlton). 2018; 23: 430-437Crossref PubMed Scopus (7) Google Scholar In addition, a randomized tertiary prevention trial, comparing enalapril with placebo in proteinuric CKDu, demonstrated reduced proteinuria with no difference in decline in estimated glomerular filtration rate over 12 months.8Selvarajah M. Mendis S. Jayasinghe S. et al.Randomized controlled trial of treatment of chronic kidney disease of uncertain aetiolgy with enalapril.J Clin Toxicol. 2016; 6: 281Google Scholar Seven interventional studies are currently registered as underway (Supplementary Table S1). These studies demonstrate the need to involve workers, employers, health care services, and communities in study design, setup, and implementation. Clearly, with appropriate design to address sources of bias, interventional studies in CKDu can be conducted and may generate evidence to inform changes in workplace, practice, or future research. Furthermore, accumulating epidemiologic data, paired with convincing animal models for a potential pathophysiologic pathway, will generate new candidates for potential interventions. Although they require independent replication, some examples of the latter include a murine model of heat stress and recurrent dehydration, identifying polyol pathway activation, vasopressin, and uric acid generation as possible contributors to kidney injury, and others demonstrating tubular kidney injury following silica nanoparticle or paraquat exposure (Supplementary Table S2).Table 1Published trials in CKDuStudyInterventionParticipantsRandomizedDesignResultsKrisher et al. (2020)4Krisher L. Butler-Dawson J. Yoder H. et al.Electrolyte beverage intake to promote hydration and maintain kidney function in Guatemalan sugarcane workers laboring in hot conditions.J Occup Environ Med. 2020; 62: e696-e703Crossref PubMed Scopus (10) Google ScholarIncreased provision of electrolyte-containing ORS50 Guatemalan sugarcane workersNoPre-post comparison. Increasing amounts of ORS made available over 3 wk.Serum creatine kinase levels were significantly reduced with increasing volumes of ORS. No differences in creatinine or hydration status were identified.Glaser et al. (2020)5Glaser J. Hansson E. Weiss I. et al.Preventing kidney injury among sugarcane workers: promising evidence from enhanced workplace interventions.Occup Environ Med. 2020; 77: 527-534Crossref PubMed Scopus (41) Google Scholar and (2022)6Glaser J. Wegman D.H. Arias-Monge E. et al.Workplace intervention for heat stress: essential elements of design, implementation, and assessment.Int J Environ Res Public Health. 2022; 19: 3779Crossref PubMed Scopus (5) Google ScholarIncreased provision of rest, shade, water, ORS, and field sanitation facilities.915 Nicaraguan sugarcane workersNoPre-post comparison over 2 successive harvestsIn workers performing the highest-intensity labor (burned cane cutters), the decline in eGFR over harvest period was reduced by 6 ml/min per 1.73 m2 and risk of incident kidney injury reduced by 30% (95% CI, 10%–50%).Siriwardhana et al. (2018)7Siriwardhana E. Perera P.A.J. Sivakanesan R. et al.Diminished disease progression rate in a chronic kidney disease population following the replacement of dietary water source with quality drinking water: a pilot study.Nephrology (Carlton). 2018; 23: 430-437Crossref PubMed Scopus (7) Google ScholarSubstitution of participant and family's drinking and cooking water supply with bottled water versus ongoing use of usual village water source.30 People with biopsy-confirmed CKDu in North Central Sri LankaNoControlled trialeGFR was significantly higher in the bottled water group at 18 mo than in those continuing to use their usual water supply (35.6 ± 5.6 vs. 22.3±4.6 ml/min per 1.73 m2; P < 0.001).Selvarajah et al. (2016)8Selvarajah M. Mendis S. Jayasinghe S. et al.Randomized controlled trial of treatment of chronic kidney disease of uncertain aetiolgy with enalapril.J Clin Toxicol. 2016; 6: 281Google ScholarEnalapril versus placebo263 People from North-Central Sri Lanka with albuminuria, eGFR >15 ml/min per 1.73 m2, and a clinical diagnosis of CKDu. People with diabetes or a history of glomerulonephritis, pyelonephritis, calculi, or snake bite were excluded.YesPlacebo-controlled trialSignificant between-group difference in change in UACR at 12 mo of –161.9 mg/g in favor of enalapril. There was no difference in change in eGFR. (Note that inclusion of albuminuric participants contrasts with typical description of no-to-minimal proteinuria in CKDu. May represent a mixed or distinct population from Central American CKDu populations.)CI, confidence interval; CKDu, chronic kidney disease of uncertain etiology; eGFR, estimated glomerular filtration rate; ORS, oral rehydration solution; UACR, urine albumin-to-creatinine ratio. Open table in a new tab Box 1Case studyThe Center for Health, Work and Environment (CHWE), Colorado School of Public Health, partnered with a Guatemala-based agribusiness to assess and improve the health, safety, and well-being of their workforce, in particular sugarcane field workers. During the 2016–2017 sugarcane harvest, a Total Worker Health intervention integrating worker safety with the promotion of health was applied in the form of an education program on the importance of water, electrolytes, rest, and shade along with a "wellness incentive" based on workers' hydration status at the start and end of the work shift. Participants with abnormal kidney function were identified throughout the study and were given additional education and clinical assessments. The researchers observed that dehydration and insufficient electrolyte consumption were risk factors for acute kidney injury across the work shift. In addition, participants identified as having poor kidney health at the start of the study had improvements in markers of kidney health with the intervention, when compared with their trajectories of decline before the intervention over multiple years.On the basis of the findings from the intervention study, the agribusiness, in collaboration with CHWE, conducted a 3-week pragmatic comparative effectiveness trial to evaluate impact of electrolyte supplementation on hydration status and health outcomes. Workers received an electrolyte hydration intervention during the 3-week trial. This trial demonstrated the feasibility of maintaining workers' electrolyte levels under extremely hot and humid conditions. With the involvement of the agribusiness' medical team, the study was able to determine that the intervention was achievable and practical to implement. The success of the intervention trial led to a revised hydration program for all field workers in Guatemala, which was rolled out to the following harvest season.4Krisher L. Butler-Dawson J. Yoder H. et al.Electrolyte beverage intake to promote hydration and maintain kidney function in Guatemalan sugarcane workers laboring in hot conditions.J Occup Environ Med. 2020; 62: e696-e703Crossref PubMed Scopus (10) Google Scholar CI, confidence interval; CKDu, chronic kidney disease of uncertain etiology; eGFR, estimated glomerular filtration rate; ORS, oral rehydration solution; UACR, urine albumin-to-creatinine ratio. The Center for Health, Work and Environment (CHWE), Colorado School of Public Health, partnered with a Guatemala-based agribusiness to assess and improve the health, safety, and well-being of their workforce, in particular sugarcane field workers. During the 2016–2017 sugarcane harvest, a Total Worker Health intervention integrating worker safety with the promotion of health was applied in the form of an education program on the importance of water, electrolytes, rest, and shade along with a "wellness incentive" based on workers' hydration status at the start and end of the work shift. Participants with abnormal kidney function were identified throughout the study and were given additional education and clinical assessments. The researchers observed that dehydration and insufficient electrolyte consumption were risk factors for acute kidney injury across the work shift. In addition, participants identified as having poor kidney health at the start of the study had improvements in markers of kidney health with the intervention, when compared with their trajectories of decline before the intervention over multiple years. On the basis of the findings from the intervention study, the agribusiness, in collaboration with CHWE, conducted a 3-week pragmatic comparative effectiveness trial to evaluate impact of electrolyte supplementation on hydration status and health outcomes. Workers received an electrolyte hydration intervention during the 3-week trial. This trial demonstrated the feasibility of maintaining workers' electrolyte levels under extremely hot and humid conditions. With the involvement of the agribusiness' medical team, the study was able to determine that the intervention was achievable and practical to implement. The success of the intervention trial led to a revised hydration program for all field workers in Guatemala, which was rolled out to the following harvest season.4Krisher L. Butler-Dawson J. Yoder H. et al.Electrolyte beverage intake to promote hydration and maintain kidney function in Guatemalan sugarcane workers laboring in hot conditions.J Occup Environ Med. 2020; 62: e696-e703Crossref PubMed Scopus (10) Google Scholar All interventional study designs have strengths and weakness (Tables 2 and 3). The archetypal parallel group individually randomized controlled trial can be applied to CKDu populations (Box 2), but often is neither feasible nor ethical. The key consideration in the selection of an intervention study design is the nature of the intervention being tested. Many potential primary or secondary prevention interventions for CKDu, such as changes to workplace practice, provision of clean drinking water, environmental protection, personal protective equipment use, and education, are most efficiently applied to communities, and would be difficult to ethically deny a "control" group. Furthermore, applying such interventions to individuals may be impractical, may not be acceptable to the community members, or may be thwarted by sharing of information and behaviors between individuals, resulting in contamination of the original random allocation.Table 2Key randomized trial designsTrial designIndividually randomizedCluster randomizedParallel groupsParallel-groups trial: distinct individuals exposed to different treatmentsCluster trial: distinct groups/sites exposed to different treatmentsCrossoverCrossover trial: same individuals exposed to each treatment in sequenceStepped-wedge trial: same groups/sites exposed to each treatment in sequence, with staggered crossoverSolid arrows indicate intervention (treatment), and dashed arrows indicate comparator (control). Open table in a new tab Table 3Study design considerations in CKDuTrial designStrengthsWeaknessesConsiderations for randomized designs in CKDuRandomized study structuresIndividual randomizationParallel groupStatistically simple and well-established designParticipants allocated to control arm do not receive treatment, which may not be possible ethically, nor acceptable to community and participants.Spread of intervention to control group ("contamination") may be difficult to prevent (especially for behavioral, workplace, or educational interventions).Unsuitable when intervention likely to affect the whole study population, or intervention and control individuals likely to share behaviors, information, or treatment (contamination).Ethical and cultural aspects of randomization need careful consideration in CKDuCrossoverSmaller sample sizes generally required because of ability to compare intervention and control in same study subjectAll participants receive interventionThis design cannot be used where a lasting effect of the intervention on the outcome or disease natural history is expected (carry-over effect).Unlikely to be of use in CKDu because of carry-over effects (e.g., prevention of incident AKI may affect future susceptibility to AKI, resulting in a carry-over effect that would prevent use of a crossover design). Similar considerations would apply to an educational or behavioral intervention.Cluster-randomized designsParallel groupSuitable for whole-community or workplace interventionsMay permit enrollment of a more representative sample of the populationAnalysis must account for intracluster correlationCluster trials are statistically less efficient in terms of number of recruited individuals, but this may be outweighed by more efficient implementation of intervention and reduced risk of bias from contamination.A minimum number of clusters is required (typically ≥6 clusters)aMinimum cluster requirement varies for each study and is dependent on a statistical power analysis. Most studies are likely to require >6 sites.Ideal design for workplace and community interventions in CKDu; however, requirement for multiple sites may present a challenge.Stepped-wedgeGreater study power than parallel-group cluster study when clusters are heterogeneous.Best suited to study relatively short-term change in incidence of an event.Statistically more complex with need to account for intracluster correlation and for effect of time.Risk of bias from underlying temporal changes.As a subtype of crossover design, stepped-wedge studies are not suited to examine long-term disease progression (e.g., CKD), or where recurrent events within the same pool of individuals is likely (e.g., recurrent AKI within a cohort of workers), as such circumstances may create a "carry-over" bias.Where cluster members are relatively stable over time (e.g., many workplaces, most communities), one must be able to assume that end points occurring in the same individual at different times during the study are independent. It is unclear how strong this assumption can be for recurrent episodes of AKI in a static population.Nonrandomized (quasi-experimental) designsPre-postSuitable when few study sites/clusters availableCan be prospective or, given adequate information on the intervention, retrospectiveUseful where randomization not feasible/ethicalLower cost when intervention simpleSubject to bias because of uncontrolled differences between the preintervention and postintervention periods.Subject to effect of underlying temporal changes. Trajectory of change in incidence over time can be analyzed for effect of intervention (time-series analysis).Prospective pre-post studies are preferred. A control period of observation is established before introduction of the intervention. This permits standardized outcome ascertainment and a better understanding of any underlying temporal trends.Nonrandomized intervention and control comparisonsMay be only feasible design when few sites availableGenerate preliminary data, requiring further evaluation.Subject to bias because of uncontrolled differences between groupsAs far as possible, data collection should be done in the same way across groups.Detailed information on each group and their exposures/treatments is important to identify potential sources of bias.AKI, acute kidney injury; CKD, chronic kidney disease; CKDu, chronic kidney disease of uncertain etiology.a Minimum cluster requirement varies for each study and is dependent on a statistical power analysis. Most studies are likely to require >6 sites. Open table in a new tab Box 2Randomized study in Sri LankaIn 2017, a parallel-group, open-label, randomized controlled study commenced at the Renal Clinic, Girandurukotte, Sri Lanka. A total of 376 people with chronic kidney disease of uncertain etiology were randomized to allopurinol (targeting serum urate levels <6 mg/dl in men and <5 mg/dl in women) or usual care alone. The primary outcome of this study is change in serum creatinine, with secondary outcomes including hospitalization, cardiovascular events, and need for dialysis. In addition to clinical staff and research assistants, the study also benefits from an existing network of patients' relatives and field health staff who assist in monitoring of adverse effects, as well as facilitating and encouraging participation. Three-year follow-up was completed in 2021, and results are expected in late 2022. Solid arrows indicate intervention (treatment), and dashed arrows indicate comparator (control). AKI, acute kidney injury; CKD, chronic kidney disease; CKDu, chronic kidney disease of uncertain etiology. In 2017, a parallel-group, open-label, randomized controlled study commenced at the Renal Clinic, Girandurukotte, Sri Lanka. A total of 376 people with chronic kidney disease of uncertain etiology were randomized to allopurinol (targeting serum urate levels <6 mg/dl in men and <5 mg/dl in women) or usual care alone. The primary outcome of this study is change in serum creatinine, with secondary outcomes including hospitalization, cardiovascular events, and need for dialysis. In addition to clinical staff and research assistants, the study also benefits from an existing network of patients' relatives and field health staff who assist in monitoring of adverse effects, as well as facilitating and encouraging participation. Three-year follow-up was completed in 2021, and results are expected in late 2022. Thus, cluster-randomized designs can be a solution for interventions applied to a group or community and are in principle suited to primary or secondary prevention studies in CKDu. A potential limitation of the parallel group cluster design is the number of sites (clusters) required to obtain reasonable balance between treatment or intervention arms (generally at least 6). Stepped-wedge cluster-randomized designs (Table 2) are attractive in that they may require fewer clusters, and all groups receive the intervention in a staged roll out. However, the stepped-wedge design, a variation of the crossover study, is not suitable for studying outcomes that require years to accrue, or where the independence of study outcomes cannot be assumed over the study period. For example, AKI occurring during the control phase may influence the likelihood of AKI occurring in the same individual if the individual remains in the study during the intervention phase. There is, in short, no one-size-fits-all study design in this unique, incompletely understood condition affecting underresourced populations. Given the above challenges, nonrandomized intervention study designs are likely to continue to play an important role in developing the evidence base for CKDu (Box 3). Measures to minimize the potential bias in such studies include a prospective design, randomized recruitment of participants or groups and contemporaneous control group, and collection of sufficient data to permit adjusted analyses that can assess the impact of differences in confounding factors (Box 4). Regardless of intervention study design, careful assessment of the implementation of an intervention is essential. Otherwise, distinguishing between an ineffective intervention and poor implementation can be difficult.6Glaser J. Wegman D.H. Arias-Monge E. et al.Workplace intervention for heat stress: essential elements of design, implementation, and assessment.Int J Environ Res Public Health. 2022; 19: 3779Crossref PubMed Scopus (5) Google Scholar Finally, intervention studies should consider incorporating practical and feasible implementation efforts and interventions, and should include economic evaluations to help all stakeholders assess the costs and benefits in light of their own resources and competing community needs. Although the added cost of such measures may present challenges, the robust evidence produced becomes a powerful argument for future funding and investment in successful interventions, which justifies the initial outlay.Box 3Prospective, nonrandomized, stepped-wedge study in NicaraguaThe Adelante Initiative is preparing a trial at 3 worksites in similar geographic and climactic regions. Cane cutters at each site will be followed over 3 successive harvests, with stepped introduction of a program of hydration and altered work practices designed to reduce the incidence of kidney injury (see Supplementary Figure S1). In year 1, site A will continue usual practice, site B will receive a hydration intervention (provision of water and isotonic beverages), and site C will receive a combined program of hydration and an altered work structure designed to minimize work during the hottest part of the day (with earlier starting times and frequent scheduled breaks in the shade during working hours). In year 2, site A will receive the hydration intervention, whereas sites B and C will receive the combined program. In year 3, all sites will receive the combined program. Concurrent with this, different implementation support strategies will also be tested by offering intervention training sessions only as opposed to training session plus onsite implementation support in a structured manner through successive harvests. The primary outcome will be episodes of acute kidney injury identified during presentations to local clinics. This stepped-wedge design will test the effect of different aspects of a workplace intervention while ensuring that all groups eventually receive the combined program. With only 3 sites available, randomization will be unable to meaningfully balance confounding factors, making adjustment for potential confounders an important part of the final analysis plan. This design also permits further demonstration of the hypothesis that heat (and elevated core temperature) per se may contribute to kidney injury irrespective of hydration, thus adding nuance to the understanding of chronic kidney disease of uncertain etiology while also potentially finding readily implementable solutions.Box 4Key biases in interventional designsPerformance bias: Interventions should be applied similarly (adherence) at each study site and avoid "contamination" of control groups by (often well-meaning) application of interventions to all participants.Detection bias: All participants should have their outcome measured in the same manner to avoid biased assessment of outcomes. This includes standard outcome definitions and blinding of assessors wherever possible.Attrition bias: Participants lost to follow-up are likely to be different to those who remain in the study, creating a risk of bias due to informative events going unobserved. In occupational health literature, attrition bias includes the "Healthy Worker Selection Effect," where injured or ill workers drop out, leaving the remaining cohort healthier overall and so potentially decreasing the impact of an intervention. The Adelante Initiative is preparing a trial at 3 worksites in similar geographic and climactic regions. Cane cutters at each site will be followed over 3 successive harvests, with stepped introduction of a program of hydration and altered work practices designed to reduce the incidence of kidney injury (see Supplementary Figure S1). In year 1, site A will continue usual practice, site B will receive a hydration intervention (provision of water and isotonic beverages), and site C will receive a combined program of hydration and an altered work structure designed to minimize work during the hottest part of the day (with earlier starting times and frequent scheduled breaks in the shade during working hours). In year 2, site A will receive the hydration intervention, whereas sites B and C will receive the combined program. In year 3, all sites will receive the combined program. Concurrent with this, different implementation support strategies will also be tested by offering intervention training sessions only as opposed to training session plus onsite implementation support in a structured manner through successive harvests. The primary outcome will be episodes of acute kidney injury identified during presentations to local clinics. This stepped-wedge design will test the effect of different aspects of a workplace intervention while ensuring that all groups eventually receive the combined program. With only 3 sites available, randomization will be unable to meaningfully balance confounding factors, making adjustment for potential confounders an important part of the final analysis plan. This design also permits further demonstration of the hypothesis that heat (and elevated core temperature) per se may contribute to kidney injury irrespective of hydration, thus adding nuance to the understanding of chronic kidney disease of uncertain etiology while also potentially finding readily implementable solutions. Performance bias: Interventions should be applied similarly (adherence) at each study site and avoid "contamination" of control groups by (often well-meaning) application of interventions to all participants. Detection bias: All participants should have their outcome measured in the same manner to avoid biased assessment of outcomes. This includes standard outcome definitions and blinding of assessors wherever possible. Attrition bias: Participants lost to follow-up are likely to be different to those who remain in the study, creating a risk of bias due to informative events going unobserved. In occupational health literature, attrition bias includes the "Healthy Worker Selection Effect," where injured or ill workers drop out, leaving the remaining cohort healthier overall and so potentially decreasing the impact of an intervention. Community, worker, workplace, health system, and policy-maker engagement is vital for CKDu studies to succeed. Consideration must be given to formalizing stakeholder involvement before, during, and following the study. Involvement might include an independent consultative committee or the addition of independent members to study steering committees. Researchers must also be cognizant of potential consequences to participating workers who may face economic loss or job insecurity if poor kidney health is identified, and the feasibility of sustaining any potential intervention within the local context. A clinical referral plan should be established before the start of the study in case adverse health outcomes are identified during the study. Employers or health policy makers should be engaged to mitigate these consequences. Researchers may convey equipoise for potential for kidney health benefit, but other benefits may nonetheless accrue (e.g., independent of effects on kidney health, safer work practices, and/or greater health literacy may yet improve workplace productivity5Glaser J. Hansson E. Weiss I. et al.Preventing kidney injury among sugarcane workers: promising evidence from enhanced workplace interventions.Occup Environ Med. 2020; 77: 527-534Crossref PubMed Scopus (41) Google Scholar or quality of life). Such engagement may include qualitative data collection with the potential to inform the design of suitable interventions for participating communities.6Glaser J. Wegman D.H. Arias-Monge E. et al.Workplace intervention for heat stress: essential elements of design, implementation, and assessment.Int J Environ Res Public Health. 2022; 19: 3779Crossref PubMed Scopus (5) Google Scholar Studies should incorporate secondary outcomes that could demonstrate that investments in community sanitation, water, and health care may have far-reaching benefits, while still allowing a rigorous evaluation of the intervention for CKDu specifically.6Glaser J. Wegman D.H. Arias-Monge E. et al.Workplace intervention for heat stress: essential elements of design, implementation, and assessment.Int J Environ Res Public Health. 2022; 19: 3779Crossref PubMed Scopus (5) Google Scholar,9Pacheco-Zenteno F. Glaser J. Jakobsson K. et al.The prevention of occupational heat stress in sugarcane workers in Nicaragua-an interpretative phenomenological analysis.Front Public Health. 2021; 9: 713711Crossref PubMed Scopus (4) Google Scholar Finally, in keeping with authentic community engagement, researchers should provide results and feedback to participants and communities, local health agencies, workers, and workplaces during and after the study, as well as discuss means of translating findings into an established practice or program. After 2 decades of investigation, several plausible hypotheses have been offered to explain the pathogenesis of CKDu, none of which have yet been proven. There is clearly much still to be learned. Given the impact of CKDu on affected communities in these low-resource settings, and the potential benefit from practical interventions, it is critical to develop and execute interventional studies that systematically address suspected risk factors, typically associated with problematic occupational or environmental conditions, and which, at the same time, have the power to both contribute to understanding causality of this devastating disease, whilst improving the health of affected or at-risk individuals and their communities.