http://orcid.org/0000-0003-1537-3355
Reader Comments
Post a new comment on this article
Post Your Discussion Comment
Please follow our guidelines for comments and review our competing interests policy. Comments that do not conform to our guidelines will be promptly removed and the user account disabled. The following must be avoided:
- Remarks that could be interpreted as allegations of misconduct
- Unsupported assertions or statements
- Inflammatory or insulting language
Thank You!
Thank you for taking the time to flag this posting; we review flagged postings on a regular basis.
closeMisdirected lessons from studying first time sugarcane harvesters
Posted by kjakobsson on 18 May 2020 at 11:22 GMT
The article Longitudinal trends in renal function among first time sugarcane harvesters in Guatemala by Dally et al, Plos One, 2020, presents findings that appear to contradict much published literature. Hence, it requires close scrutiny. The paper examines new cane cutters with no prior cutting experience, a potentially valuable addition to the literature. However, I am concerned with how the authors have interpreted their data and the potential for a conflict of interest. While acknowledging the approach taken was clever, its execution was hampered by circumstance, selection bias and loss to follow-up. I do not believe the findings support the authors’ conclusion for the following reasons:
The authors relied on routinely collected pre-employment data from the company, assembled 2012-2016. The loss of follow-up was substantial. Two thirds of the baseline population of 534 first job applicants did not return for a 2nd year and only 29 subjects had more than two years of follow-up.
The authors chose an unfortunate way to describe the two groups created by a joint latent class mixed analysis. One group has “non-stable” function although the observed rate of change in that group is neither clinically relevant nor significantly different from the other “stable” group: ”non-stable” annual age-adjusted estimated glomerular filtration rate (eGFR) change of -1.0 ml/min/1.73 m2 (95% CI: -3.4, 1.3) compared to 0.3 (95% CI -0.9, 1.5) in those with “stable” function. Changes at this level likely represent random variations or measurement error.
The authors highlight findings that the “non-stable” kidney function group had i) lower baseline eGFR, ii) mild hypertension at baseline, and iii) higher proportion of local residents as opposed to migrants. Multivariate modeling yielded adjusted odds ratios of very high magnitude (as high as 7) for associations of these risk factors with ‘group’ assignment. In most epidemiologic research ORs of this magnitude lead investigators to question whether, as is often the case, these result from a data error or artifact in the analysis.
Regarding mild baseline hypertension, the authors were unfortunately not in control of data collection and could not verify that baseline blood pressure data were consistently collected in a validated, standardized manner. Concern with measurement accuracy is suggested by the extreme finding of more than 1/3 of workers with "stable" trajectory having SBP <105 mmHg. Also, it should be noted that there was more Stage 1 hypertension in the “non-stable” group, but more Stage 2 hypertension in the “stable” group.
With regard to local residence, the overrepresentation of workers living near the mill in the “non-stable” kidney function group may well represent healthy worker selection. There were stricter criteria regarding eligibility for pre-employment examination of highland workers [1]. Whereas workers living nearby were free to apply for pre-employment examination at the mill, highland workers were preselected for pre-employment screening based on, among other criteria, not having any medical problems that would impact their ability to work.
Finally, I am puzzled that the authors indicate the company that funded part of the study had no role in the evaluation design, data analysis or interpretation. Nonetheless, three of the coauthors are identified as company personnel and the authors conclude that their findings should point researchers to focus on the community setting to identify risks outside of recognized occupational risks for CKDnt. Since the conclusions point away from work as an important component of kidney function in this workforce, the possibility of a conflict of interest is real.
In sum, there is selection bias in the study population that affects baseline health, the loss to follow-up is huge, and the mild baseline hypertension is quite possibly an artifact. Moreover, the epidemic of CKDu, which is the underlying concern for this research, refers to a disease which clinically and epidemiologically is defined by the absence of hypertension. To declare “mild hypertension” at pre-employment as a possible important risk factor for a non-hypertensive disease would require a coherent pathophysiological explanation, which still is missing.
Lower eGFR as a risk factor for subsequent decline has previously been observed [2] and is indeed of importance for prevention. In contrast, the unsupported conclusion that the non-occupational factors of residence and mild hypertension of workers entering the workforce explain kidney function decline during their employment, distracts the attention from the impact of this physically demanding work in hot environments and from addressing the important occupational risks that have been documented [3].
1. Butler-Dawson J, Krisher L, Asensio C, Cruz A, Tenney L, Weitzenkamp D, Dally M, Asturias EJ, Newman LS. Risk factors for declines in kidney function in sugarcane workers in Guatemala. J Occup Environ Med. 2018;60(6):548-58. https://doi:10.1097/JOM.0.... PMID: 29370016
2. Hansson E, Glaser J, Weiss I, Ekström U, Apelqvist J, Hogstedt C, Peraza S, Lucas R, Jakobsson K, Wesseling C, Wegman DH. Workload and cross-harvest kidney injury in a Nicaraguan sugarcane worker cohort. Occup Environ Med. 2019 Nov;76(11):818-826. doi: 10.1136/oemed-2019-105986. PubMed PMID: 31611303; PubMed Central PMCID: PMC6839725.
3. Wesseling C, Glaser J, Rodríguez-Guzmán J, Weiss I, Lucas R, Peraza S, da Silva AS, Hansson E, Johnson RJ, Hogstedt C, Wegman DH, Jakobsson K. Chronic kidney disease of non-traditional origin in Mesoamerica: a disease primarily driven by occupational heat stress. Rev Panam Salud Publica. 2020;44:e15. https://doi: 10.26633/RPSP.2020.15. PMID: 31998376
RE: Misdirected lessons from studying first time sugarcane harvesters
mdally replied to kjakobsson on 03 Jun 2020 at 14:49 GMT
We would like to thank the commenter for opening up a community dialogue around this devastating epidemic. There is general consensus in the research community that the causes of Chronic Kidney Disease of Unknown Cause (CKDu) remain unknown[1]. Performance of intense work under conditions of high heat and dehydration have been shown to be associated with kidney dysfunction[2, 3]. We, along with research groups and public health professionals around the world, continue to investigate other contributors to disease risk in order to implement evidence-based interventions that maximally benefit workers. This is especially important based on evidence that the successful implementation of water, rest, and shade interventions, while beneficial, still leave some workers experiencing reductions in kidney function[4-6]. Disciplined research approaches led by investigators who are open to considering all potential risk factors are needed to fill the knowledge gaps[1].
In this study, we have shown that there may be pre-existing risk factors that predispose young workers to accelerated kidney function decline. This study establishes that there is a subgroup of people for whom their occupation can make them more susceptible to reductions in kidney function. By identifying these groups of individuals, we are allowing for the development of enhanced surveillance and intervention approaches that are in-line with occupational safety and health best practices.
To be fully transparent we have elected to have the editorial history of this manuscript published along with the article. It can be accessed here: https://journals.plos.org.... Along with the commenter, one of the reviewer’s concerns was the potential bias introduced into the analysis due to the loss to follow-up. We addressed this by examining differences between baseline eGFR and other covariates with loss to follow-up. While our previous research has shown that kidney function was related to mid-harvest drop out[7], in the present study those who did not return only had slightly lower baseline eGFR than those who were included in the cohort (117 ml/min per 1.73 m2 compared to 119 ml/min per 1.73 m2, respectively) as shown in Supporting Information Table 1.
The commenter raises an important issue surrounding the linear estimation of yearly eGFR decline. Our joint latent class mixed model posterior classification probabilities ranged from 91% to 96% indicating good discrimination between the groups based on longitudinal profile of eGFR and drop out. It is important to note that we allowed for a quadratic term for time in these models but reported linear eGFR trends for the groups for ease of comparison to other studies. One of the major limitations of longitudinal studies of kidney function decline is that it is assumed to be linear, while the assumption of linearity may not be appropriate[8]. We encourage researchers in the field to take this into consideration when performing their analyses.
High odds ratios from a single study should not serve as conclusive evidence and should be reviewed critically. Studies have shown that lower baseline eGFR[9-11] is a risk factor for rate of kidney function decline and coastal home of residence is one of the only meta-analysis-supported risk factors for CKDu[12]. Additionally, the potential for misclassification of home of residence is low.
The commenter cited one of our previously published works in regard to potential overrepresentation of local workers. We stated in that manuscript that, “highland workers come from the highland regions and are preselected based on whether they cut an average of 5 tons per day and worked at least 90 days the previous harvest. In addition, they must also not have any medical problems that would impact their ability to work. The people who pass the preselection are invited to apply for a position during the pre-employment process. New highland workers can also be invited to apply on-site by Pantaleon.”[9] Of note, all cane cutters go through the pre-employment medical screening and must have an eGFR >90 ml/min/1.73 m2 to work that harvest season. While this may represent a slight selection bias, as shown in Supporting Information Table 1, in this study there was a higher proportion of local workers (versus highlands workers) not seeking subsequent employment.
As the commenter pointed out, one of the main findings of this study was that having a mild elevation of blood pressure (systolic blood pressure ≥ 130 or diastolic blood pressure ≥ 80) was associated with non-stable group classification. The finding itself is not surprising, the observation that mild hypertension is associated with kidney injury and that it may predispose to progression is well known[13]. With regard to data collection, the blood pressure readings were performed by trained clinical personnel employed by the business following a detailed protocol. Per clinical protocol, elevated blood pressure readings were re-checked by a licensed medical doctor. The clinical data were obtained as part of a routine preemployment medical screening process by practitioners who were blind to any hypothesis and, in fact, had no knowledge that the data would be later used for purposes of research. As evidenced in the published peer review, we had originally provided more detailed information about the blood pressure measurement process but removed it at the request of the reviewer. To determine if our results surrounding mild elevations of blood pressure were stable and robust, we randomly re-classified 20% of those with mild elevations to the normal blood pressure category. This sensitivity analysis supported our findings and was published in the Supporting Information.
Three of the co-authors of this manuscript are employed by Pantaleon. These authors, AC, WDP, and CA, all met the requirements for authorship as laid out by the authorship policy and their contributions following the CRediT taxonomy has been included with this manuscript. The researchers from the University of Colorado have been transparent in their relationship with Pantaleon. This relationship has been reviewed during the peer review process of this manuscript, including by reviewers, the editor of this journal, the University of Colorado, as well as a fully independent external advisory group. As experienced occupational safety and health scientists, we are rigorous in our approach to conflict of interest, ethics, and methods of engaging with any commercial or non-commercial collaborator.
Much of the knowledge gained thus far regarding CKDu, including the evaluation of interventions that directly benefit workers, has only been possible when researchers and public health professionals have established working relationships with industry[4-6, 14]. Strict adherence to ethical principles and to transparency, including acknowledgment of any potential conflicts of interest, are essential. We call upon all researchers in this field to fully disclose all relationships, financial and non-financial, when working in the field of occupational health.
We would again like to thank the commenter for opening up this discussion.
References
1. Johnson RJ, Wesseling C, Newman LS. Chronic Kidney Disease of Unknown Cause in Agricultural Communities. New England Journal of Medicine. 2019;380(19):1843-52. doi: 10.1056/nejmra1813869.
2. Sorensen CJ, Butler-Dawson J, Dally M, Krisher L, Griffin BR, Johnson RJ, et al. Risk Factors and Mechanisms Underlying Cross-Shift Decline in Kidney Function in Guatemalan Sugarcane Workers. Journal of Occupational and Environmental Medicine. 2019;61(3):239-50. doi: 10.1097/jom.0000000000001529.
3. Wesseling C, Aragón A, González M, Weiss I, Glaser J, Rivard CJ, et al. Heat stress, hydration and uric acid: a cross-sectional study in workers of three occupations in a hotspot of Mesoamerican nephropathy in Nicaragua. BMJ Open. 2016;6(12):e011034-e. doi: 10.1136/bmjopen-2016-011034. PubMed PMID: 27932336.
4. Wegman DH, Apelqvist J, Bottai M, Ekström U, García-Trabanino R, Glaser J, et al. Intervention to diminish dehydration and kidney damage among sugarcane workers. Scand J Work Environ Health. 2017. Epub 2017/07/07. doi: 10.5271/sjweh.3659. PubMed PMID: 28691728.
5. Butler-Dawson J, Krisher L, Yoder H, Dally M, Sorensen C, Johnson RJ, et al. Evaluation of heat stress and cumulative incidence of acute kidney injury in sugarcane workers in Guatemala. International Archives of Occupational and Environmental Health. 2019. doi: 10.1007/s00420-019-01426-3.
6. Hansson E, Glaser J, Weiss I, Ekström U, Apelqvist J, Hogstedt C, et al. Workload and cross-harvest kidney injury in a Nicaraguan sugarcane worker cohort. Occupational and environmental medicine. 2019;76(11):818-26. doi: 10.1136/oemed-2019-105986. PubMed PMID: 31611303.
7. Dally M, Butler-Dawson J, Krisher L, Monaghan A, Weitzenkamp D, Sorensen C, et al. The impact of heat and impaired kidney function on productivity of Guatemalan sugarcane workers. PLoS One. 2018;13(10):e0205181-e. doi: 10.1371/journal.pone.0205181. PubMed PMID: 30289894.
8. Li L, Astor BC, Lewis J, Hu B, Appel LJ, Lipkowitz MS, et al. Longitudinal progression trajectory of GFR among patients with CKD. American journal of kidney diseases : the official journal of the National Kidney Foundation. 2012;59(4):504-12. Epub 2012/01/26. doi: 10.1053/j.ajkd.2011.12.009. PubMed PMID: 22284441.
9. Butler-Dawson J, Krisher L, Asensio C, Cruz A, Tenney L, Weitzenkamp D, et al. Risk Factors for Declines in Kidney Function in Sugarcane Workers in Guatemala. Journal of occupational and environmental medicine. 2018;60(6):548-58. doi: 10.1097/JOM.0000000000001284. PubMed PMID: 29370016.
10. Wesseling C, Aragón A, González M, Weiss I, Glaser J, Bobadilla NA, et al. Kidney function in sugarcane cutters in Nicaragua – A longitudinal study of workers at risk of Mesoamerican nephropathy. Environmental Research. 2016;147:125-32. doi: https://doi.org/10.1016/j....
11. Gonzalez-Quiroz M, Smpokou E-T, Silverwood RJ, Camacho A, Faber D, Garcia BLR, et al. Decline in Kidney Function among Apparently Healthy Young Adults at Risk of Mesoamerican Nephropathy. J Am Soc Nephrol. 2018;29(8):2200-12. Epub 2018/06/15. doi: 10.1681/ASN.2018020151. PubMed PMID: 29907699.
12. González-Quiroz M, Pearce N, Caplin B, Nitsch D. What do epidemiological studies tell us about chronic kidney disease of undetermined cause in Meso-America? A systematic review and meta-analysis. Clin Kidney J. 2018;11(4):496-506. Epub 2017/12/08. doi: 10.1093/ckj/sfx136. PubMed PMID: 30094014.
13. Johnson RJ, Herrera-Acosta J, Schreiner GF, Rodríguez-Iturbe B. Subtle Acquired Renal Injury as a Mechanism of Salt-Sensitive Hypertension. New England Journal of Medicine. 2002;346(12):913-23. doi: 10.1056/NEJMra011078.
14. Bodin T, Garcia-Trabanino R, Weiss I, Jarquin E, Glaser J, Jakobsson K, et al. Intervention to reduce heat stress and improve efficiency among sugarcane workers in El Salvador: Phase 1. Occup Environ Med. 2016;73(6):409-16. Epub 2016/04/14. doi: 10.1136/oemed-2016-103555. PubMed PMID: 27073211; PubMed Central PMCID: PMCPMC4893112.