Journal Pre-proof

Pesticides and public health in Brazil

Francisco J.R. Paumgartten

PII: S2468-2020(20)30005-X
DOI: https://doi.org/10.1016/j.cotox.2020.01.003
Reference: COTOX 236

To appear in: Current Opinion in Toxicology

Received Date: 21 December 2019

Accepted Date: 7 January 2020

Please cite this article as: F.J.R. Paumgartten, Pesticides and public health in Brazil, Current Opinion in
Toxicology, https://doi.org/10.1016/j.cotox.2020.01.003.

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition
of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of
record. This version will undergo additional copyediting, typesetting and review before it is published
in its final form, but we are providing this version to give early visibility of the article. Please note that,
during the production process, errors may be discovered which could affect the content, and all legal
disclaimers that apply to the journal pertain.

© 2020 Elsevier B.V. All rights reserved.

Francisco J.R. Paumgartten

Addresses

National School of Public Health, Oswaldo Cruz Foundation, Laboratory of

Environmental Toxicology. Av. Brasil 4036, Rio de Janeiro, RJ 21040-361,

Brazil.

paum@ensp.fiocruz.br

1
Abstract

Brazil is one of the world’s top-four agricultural pesticide-consuming countries.

The extensive use of pesticides raises concerns about the consequences for

the human health. This review appraised the evidence provided by

epidemiological studies on the adverse health effects of pesticides conducted in

Brazil within the past 2 to 4 years. Available data come from ecological, cross-

sectional and case-control studies which are relatively easy, quick and

inexpensive to conduct, but of limited usefulness for inferring causation and to

identify hazards in pesticide risk assessment. Inaccurate exposure assessment

is another weakness common to these studies. No longitudinal cohort

investigation of pesticide adverse health effects and no analytical data-based

biomonitoring study were found in the literature. In conclusion, studies

conducted in Brazil failed to generate data relevant for risk assessment and

management of pesticides. Large prospective cohorts studies and good

analytical data on exposures are needed to bridge this public health research

gap

Key words: Pesticides, cancer, mental disorders, acute poisoning, risk

assessment.

2
1. Introduction

In the mid-20th century, the introduction of new, high-yielding varieties of

food crops, and the widespread adoption of mechanization, irrigation techniques

and other agricultural technologies resulted in dramatic gains in the production

of grains (wheat, maize, soybean and others), particularly in developing

countries. This unprecendented increase in cropland productivity (“Green

Revolution”) is thought to have changed dramatically the geography of hunger

averting millions of deaths from famine worldwide. Nonetheless, Green

Revolution’s reliance on a massive use of agrochemicals has also raised

concerns about the consequences of the extensive use of pesticides for the

environment and public health [1].

Brazil is the second largest producer of soybean in the world, and an

important producer and exporter of several other commodity crops as well.

Pesticide use considerably increased in tandem with marked rises in the

production of grains, and Brazil has become one of the world’s top-four

pesticide consumers (together with the US, EU and China) [2].

Recent surveys consistently confirmed that farmers seldom wear

appropriate Personal Protective Equipment (PPE) when handling, mixing and

applying plant protection products [3-10], a finding that adds to current concerns

about the health consequences of country’s consumption of pesticides for

agricultural workers.

Agricultural commodities are one of the mainstays of Brazil’s economy

and pesticide benefits for crop protection must be properly balanced against

their potential risks for the environment and human health. Reliable scientific

3
data are necessary to put the estimated burden of diseases associated with

pesticide use on one of the dishes of the balance.

What is the morbidity and mortality associated with the extensive use of

agricultural pesticides in Brazil? Which active ingredients currently used in the

country’s agriculture are the most dangerous and should be phased out? What

public health interventions could lead to a consistent reduction in the burden of

diseases attributable to pesticides?

These questions have remained largely unanswered so far.

Epidemiological studies focusing on them are needed for evidence-based public

health interventions intended to reduce the burden of disease associated with

the current use of pesticides in Brazil.

This article appraised the evidence provided by recent studies which

addressed directly or indirectly these research gaps.

Epidemiology in health risk assessement of pesticides

Epidemiology is one of the cornerstones of Public Health providing data

on the distribution of disease in a population (descriptive epidemiology) and on

the relationships between exposures and health outcomes within specific

populations (analytic epidemiology). Analytic epidemiology approaches test

hypotheses on determinants of disease (risk factors) by interventional (clinical

trials) and observational investigations. Owing to ethical constraints, hipotheses

on the health harms caused by pesticides are generally investigated by

observational studies, including longitudinal (retrospective or prospective)

cohort studies, case-control comparisons and cross-sectional studies. Analytic

studies measure the strength of associations between exposure and diseases,

or other adverse health outcomes, in a population at a given time.

4
 In pesticide risk assessment (RA), epidemiology contributes to the

identification of potential health hazards, assessment of dose-response

relationships, and exposure assessment [3,11]. As far as hazard identification is

concerned, epidemiological research has the advantage of circumventing

uncertainties involved in extrapolating findings from in vitro assays and animal

experiments to humans. Nonetheless, association (or correlation) does not

necessarily imply causation, and thus non-causal explanations for observed

associations should be excluded before concluding that they are likely to be

causal. In 1965, a seminal article by Bradford Hill [12] listed nine aspects that

should be considered before concluding that causation is a probable

explanation for the association. The strength of the association is at the top of

Hill’s list. According to Doll [13], when relative risks are small (of the order of

2:1 or less) the problems of eliminating bias and confounding are immense and

usually require massive data. Demonstration of a relationship between the

magnitude of exposure (dose) and the health outcome incidence or severity

(response) also contributes to prove causality.

Lack of accurate assessments of pesticide exposure is a pitfall of most

observational studies. In complex exposure scenarios as those of agriculture,

assessment of exposure is a challenging issue. It is particularly challenging,

however, if pesticide exposure is assessed when the health outcome of interest

has already occurred, as in case-control and retrospective cohort studies.

Studies on the health impact of pesticides

Acute pesticide poisonings and pesticide suicides

Pesticide poisonings, including self-poisonings (suicides), have been

considered a serious public health problem in rural communities of developing

5
countries [4,15-21]. The magnitude of the problem in Brazil, however, remains

unclear. Various descriptive studies reported data from surveys performed in

some particular regions, or data retrieved from nationwide databases (Notifiable

Diseases Information System–SINAN, and National System of Toxic-

Pharmacological Information–SINITOX). There are apparent inconsistencies

between reports based on different nationwide databases and one recent study

presented clear indications of a significant underreporting of fatal cases of

pesticide poisoning [20].

Pesticide exposure and health effect biomarkers

A number of cross-sectional studies compared biomarkers of genetic

damage (evaluated by comet and cytokinesis-block micronucleus assays, and

other tests in lymphocytes and buccal mucosal cells) and/or oxidative stress in

Brazilian agricultural workers and individuals not occupationally exposed to

pesticides [9,22-25]. These studies, as a rule, concluded that pesticide

exposures might result in genetic damage and/or enhanced oxidative stress. A

study also described clinical pathology findings in farmers exposed to pesticides

[26].

Two cross-sectional studies reported changes in blood levels of thyroid

hormone which were inconsistent across sexes and active ingredients [27,28].

The foregoing studies have limitations the most notorious of which were

non-random selection of subjects, small sample sizes, inaccurate assessments

of pesticide exposure, multiple comparison pitfalls, and lack of a clear and

plausible test hypothesis.

Pesticide exposure and long term conditions

6
Breast cancer. A case-control study [29] reported that living near of croplands

with pesticide application (mostly soybean and corn crops) was associated with

a nearly 2-fold increase in breast cancer risk (OR: 2.37; CI 95%: 1.78-3.16).

Case–control studies are inexpensive and quick to conduct, and an efficient

method for the investigation of rare outcomes. Nonetheless, they suffer from

serious limitations including their susceptibility to bias (e.g. recall and selection

biases) and reverse causality. In this particular study, the evidence for causality

is additionally weakened by an inaccurate assessment of pesticide exposure.

The proximity (residing within 500 m) of a pesticide application area was used

as a proxy for a quantitative measurement exposure. No data and no previous

study, however, were presented to corroborate the underlying hypothesis that

dwellers within 500 m of pesticide application areas would be in fact more

exposed to this or that pesticide, or to a mixture of active ingredients, than those

people who lived outside these arbitrary boundaries. Moreover, the exposure

status of case and control women was inferred from responses to

questionnaires, and analyzed as a dichotomous variable (exposed/not-

exposed), yet pesticide exposure (internal doses received by members of a

population) is likely to be a continuous variable. In the case of binary variables,

nondifferential misclassification might eventually favor the null hypothesis, i.e., it

may underestimate hypothetical associations of exposure with health outcomes.

Colon cancer (CC). An ecological study raised the hypothesis that pesticide

exposure could be a potential risk factor behind a steadily rising incidence of

CC in the Southern and Southeast regions of Brazil [30,31]. Ecological studies

are descriptive epidemiological approaches that take populations, or groups, as

the unit of observation. Being cheap, easy and quick to conduct, they may be

7
useful when individual-level data are either difficult or impossible to collect. A

notorious weakness of ecological approaches, however, is that aggregate data

are used to assess exposure. Extrapolating conclusions drawn from group

results (average in the population) to the individual level may result in

interpretation errors (ecological fallacy); i.e., associations observed between

group-level variables (e.g., pesticide consumption and CC) do not necessarily

exist for particular individuals of the group (i.e., people with CC may not be

those who were exposed to pesticides). Owing to its weaknesses, an ecological

study provides, at its best, hypothesis-generating information. Conclusions from

ecological studies are thus of limited value, if any, for hazard idenfication in RA.

Non-Hodgkin lymphoma (NHL). An ecological study reported a correlation

between per capita consumption of pesticides and standardized mortality rate

by NHL [32]. An additional drawback of this study was that it estimated risks for

NHL as an aetiologically related group of malignancies. Nonetheless, NHL

encompasses distinct subtypes, and associations of pesticides with NHL seem

to be subtype- and compound-specific [33].

Skin melanoma. A pooled analysis of two case-control studies conducted in

Italy and Brazil found a higher risk of cutaneous melanoma in individuals

occupationnaly exposed to pesticides and sunlight [34]. A previous case-control

study in Southern Brazil detected associations between occupational (OR;

CI95%: 3.2; 1.2–6.8) and residential (indoors for >10 years; 1.9; 1.2–8.2)

exposure to pesticides and cutaneous melanoma [35]. In both studies pesticide

exposure was evaluated by a questionnaire.

Depressive disorders. A cross-sectional study found an association of

depressive symptoms (assessed by the Beck Depression Inventory-BDI-II) with

8
pesticide exposure in male farmers (220 coffee growers) from Southeast Brazil

[36]. Information about demographic data, life style habits (tobacco, alcohol)

and pesticide exposure (yes or no, for any type of pesticide) was assessed by a

structured questionnaire applied by trained students and university staff. The

multivariate analysis found that pesticide exposure (OR: 5.52; CI95%: 1.18,

25.88), tobacco use (2.81; 1.11, 7.11), poor self-perceived health (2.61;1.33,

5.11) and chronic disease (2.38;1.16, 4.87) were associated with depressive

symptoms. Lack of control for possible selection (sampling by convenience

instead of random sampling) and information biases seem to be major

drawbacks of this study.

Another cross-sectional investigation [37] in a rural region of Southern

Brazil found associations (OR; CI 50%) between self-reported “pesticide

poisoning” and “common mental disorders” (2.63; 1.62-4.25) and “pesticide

poisoining” and “depression” (2.62; 1.63-4.21). Shortcomings of this study

include multiple comparison pitfalls, inaccurate assessments of exposure and

health outcomes (self-reported), and lack of control of potential biases and

confoundings.

A limitation of cross-sectional approaches for this type of inference is

that, unless it can be safely assumed that pesticide exposure was stable over

time and was not influenced by the outcome, cross-sectional data are not

suitable for drawing conclusions about determinants of chronic conditions, such

as psychiatric disorders, cancer, cardiovascular illness, and others, that may

have started a long time before associations between exposure and health

outcomes were assessed.

9
Male fertility. A cross-sectional investigation in Southern Brazil [38], compared

reproductive hormone levels and sperm quality parameters in random samples

of young men from rural (n=99) and urban (n=36) areas. Exposure was

assessed by questionnaire application. Exposure (lifetime) to pesticides was

associated with abnormal sperm morphology and reduced LH and prolactin,

while prenatal exposure (reported maternal farming during pregnancy) was

associated with larger anogenital distance and testicular volume (measured in

adulthood) [38]. Small sample sizes, multiple comparisons and inaccurate

assessment of pesticide exposure are major limitations of this cross-sectional

study.

Concluding remarks.

In summary, health effects of occupational exposure to pesticides in

Brazil were investigated by ecological, cross-sectional and case-control studies

with inaccurate assessments of exposure. Although being relatively easy, quick

and inexpensive to conduct, the methodological limitations inherent to these

studies hamper causal inferences. To the best of our knowledge, no longitudinal

retrospective or prospective cohort study was conducted in Brazil to investigate

the health effects of pesticides. Reliable biomonitoring studies and analytical

data on pesticide exposures are missing as well. Biomonitoring data are not

only an integral part of RA but also needed for a posteriori evaluations of the

effectiveness of public health interventions to reduce exposure and risks [39].

In conclusion, epidemiological studies conducted in Brazil – a world’s top

pesticide-consuming country – largely failed to generate data relevant for RA of

pesticides. The excess of morbidity and mortality attributable to pesticides

remains elusive and so does the burden of disease associated with their

10
extensive use in the country’s agriculture. Large prospective cohort

investigations are needed to brigde this public health research gap. Obviously,

enforcement of strict regulations on the use of PPE and improvement of

agricultural and rural extension programs are effective health risk reducing

interventions that do not rely on further epidemiological data.

11
Conflict of Interest.

The author declares no conflict of interest.

References

1. Pimentel D: Green Revolution agriculture and chemical hazards. Sci.

Total Environ. 1996, 188 (Suppl.1): S-86-S98.

2. Donley N: The USA lags behind other agricultural nations in banning

harmful pesticides. Environmental Health 2019, 18:44.

3. Lermen J, Bernieri T, Rodrigues IS, Suyenaga ES, Ardenghi PG:

Pesticide exposure and health conditions among orange growers in

Southern Brazil. J Environ Sci Health B. 2018; 53(4):215-221.

4. Magalhães AFA, Caldas ED. Occupational exposure and poisoning

by chemical products in the Federal District. Rev Bras Enferm. 2019;

72 (suppl 1):32-40.

5. Marcelino AF, Wachtel CC, Ghisi NC. Are Our Farm Workers in

Danger? Genetic Damage in Farmers Exposed to Pesticides. Int J

Environ Res Public Health. 2019;16(3). pii: E358.

6. Ribeiro MG, Colasso CG, Monteiro PP, Pedreira Filho WR, Yonamine M.

Occupational safety and health practices among flower

greenhouses workers from Alto Tietê region (Brazil). Sci Total

Environ. 2012; 416:121-6.

7. Chaves TVS, Islam MT, de Moraes MO, de AlenDeccar MVOB, Gomes

DCV, de Carvalho RM, Maluf SW, de Moura do Amaral FP, Paz MFCJ,

Cerqueira GS, Rolim HML, de Castro E Sousa JM, de Carvalho Melo-

Cavalcante AA, de Moraes MEA. Occupational and life-style factors-

12
 acquired mutagenicity in agric-workers of northeastern Brazil.

Environ Sci Pollut Res Int. 2017; 24(18):15454-15461.

8. Brust RS, Oliveira LPM, Silva ACSSD, Regazzi ICR, Aguiar GS, Knupp

VMAO. Epidemiological profile of farmworkers from the state of Rio

de Janeiro. Rev Bras Enferm. 2019; 72(suppl 1):122-128.

9. Marcelino AF, Wachtel CC, Ghisi NC. Are Our Farm Workers in

Danger? Genetic Damage in Farmers Exposed to Pesticides. Int J

Environ Res Public Health. 2019;16(3).

10. Cazé AMB, Lacerda ABM, Lüders D, Conto J, Marques J, Leroux T.

Perception of the Quality of Life of Tobacco Growers Exposed to

Pesticides: Emphasis on Health, Hearing, and Working Conditions.

Int Arch Otorhinolaryngol. 2019;23(1):50-59.

11. Samet JM, Schnatter R, Gibb H. Epidemiology and risk assessment.

Am J Epidemiol. 1998;148(10):929-36.

12. Christensen K, Christensen CH, Wright JM, Galizia A, Glenn BS, Scott

CS, Mall JK, Bateson TF, Murphy PA, Cooper GS: The use of

epidemiology in Risk Assessment: Challenges and opportunities.

Hum Ecol Risk Assess 2015; 21(6): 1644-1663.

13. Hill AB. The environment and disease: Association or causation?

Proc R Soc Med. 1965; 58:295-300.

14. Doll R. Proof of causality: deduction from epidemiological

observation. Perspect Biol Med. 2002; 45(4):499-515.

15. Hendges C, Schiller ADP, Manfrin J, Macedo EK Jr, Gonçalves AC Jr,

Stangarlin JR: Human intoxication by agrochemicals in the region of

13
 South Brazil between 1999 and 2014. J Environ Sci Health B. 2019;

54(4):219-225.

16. Queiroz PR, Lima KC, Oliveira TC, Santos MMD, Jacob JF, Oliveira

AMBM. Notifiable Diseases Information System and human

poisoning by pesticides in Brazil. Rev Bras Epidemiol.

2019;22:e190033.

17. Karunarathne A, Gunnell D, Konradsen F, Eddleston M. How many

premature deaths from pesticide suicide have occurred since the

agricultural Green Revolution? Clin Toxicol (Phila). 2019 Sep 9:1-6.

18. Buendía JA, Chavarriaga GJR, Zuluaga AF. Burden of paraquat

poisoning in the department of Antioquia, Colombia. BMC

Pharmacol Toxicol. 2019; 20(1):11.

19. Gondim AP, Nogueira RR, Lima JG, Lima RA, Albuquerque PL, Veras

MD, Ferreira MA. Suicide attempts by exposure to toxic agents

registered in a Toxicological Information and Assistance Center in

Fortaleza, Ceará, Brazil, 2013. Epidemiol Serv Saude. 2017;26(1):109-

119.

20. * Magalhães AFA, Caldas ED. Underreporting of fatal poisonings in

Brazil – A descriptive study using data from four information

systems. Forensic Sci Int. 2018; 287:136-141.

This comparative analysis of notifiable diseases and toxic pharmacological

information systems reveals inconsistences between databases and suggests

that fatal poisonings are likely to be underreported.

21. Magalhães AFA, Caldas ED. Two health information systems to

characterize poisoning in Brazil-a descriptive study. J Public Health

(Oxf). 2019; 41(1):203-211.

14
22. Benedetti D, Lopes Alderete B, de Souza CT, Ferraz Dias J,

Niekraszewicz L, Cappetta M, Martínez-López W, Da Silva J. DNA

damage and epigenetic alteration in soybean farmers exposed to

complex mixture of pesticides. Mutagenesis. 2018;33(1):87-95.

23. Claudio SR, Simas JMM, Souza ACF, DO Carmo Baracho DE Alencar

M, Yamauchi LY, Ribeiro DA. Genomic Instability and Cytotoxicity in

Buccal Mucosal Cells of Workers in Banana Farming Evaluated by

Micronucleus Test. Anticancer Res. 2019;39(3):1283-1286

24. Tomiazzi JS, Judai MA, Nai GA, Pereira DR, Antunes PA, Favareto APA.

Evaluation of genotoxic effects in Brazilian agricultural workers

exposed to pesticides and cigarette smoke using machine-learning

algorithms. Environ Sci Pollut Res Int. 2018;25(2):1259-1269.

25. Hilgert Jacobsen-Pereira C, Dos Santos CR, Troina Maraslis F, Pimentel

L, Feijó AJL, Iomara Silva C, de Medeiros GDS, Costa Zeferino R, Curi

Pedrosa R, Weidner Maluf S. Markers of genotoxicity and oxidative

stress in farmers exposed to pesticides. Ecotoxicol Environ Saf.

2018;148:177-183.

26. Dalbó J, Filgueiras LA, Mendes AN. Effects of pesticides on rural

workers: haematological parameters and symptomalogical reports.

Cien Saude Colet. 2019;24(7):2569-2582.

27. Piccoli C, Cremonese C, Koifman RJ, Koifman S, Freire C. Pesticide

exposure and thyroid function in an agricultural population in

Brazil. Environ Res. 2016; 151:389-398

15
28. Bernieri T, Rodrigues D, Barbosa IR, Ardenghi PG, Basso da Silva L.

Occupational exposure to pesticides and thyroid function in

Brazilian soybean farmers. Chemosphere. 2019;218:425-429.

29. * Silva AMC, Campos PHN, Mattos IE, Hajat S, Lacerda EM, Ferreira

MJM. Environmental Exposure to Pesticides and Breast Cancer in a

Region of Intensive Agribusiness Activity in Brazil: A Case-Control

Study. Int J Environ Res Public Health. 2019;16(20). pii: E3951.

This is a typical case control control with an inaccurate retrospective

assessment of pesticide exposure.

30. Uyemura SA, Stopper H, Martin FL, Kannen V. A Perspective

Discussion on Rising Pesticide Levels and Colon Cancer Burden in

Brazil. Front Public Health. 2017;5:273

31. Martin FL, Martinez EZ, Stopper H, Garcia SB, Uyemura SA, Kannen V.

Increased exposure to pesticides and colon cancer: Early evidence

in Brazil. Chemosphere. 2018;209:623-631.

32. Boccolini Pde M, Boccolini CS, Chrisman Jde R, Markowitz SB, Koifman

S, Koifman RJ, Meyer A. Pesticide use and non-Hodgkin's lymphoma

mortality in Brazil. Int J Hyg Environ Health. 2013;216(4):461-6.

33. * Leon ME, Schinasi LH, Lebailly P, Beane Freeman LE, Nordby KC,

Ferro G, Monnereau A, Brouwer M, Tual S, Baldi I, Kjaerheim K,

Hofmann JN, Kristensen P, Koutros S, Straif K, Kromhout H, Schüz J.

Pesticide use and risk of non-Hodgkin lymphoid malignancies in

agricultural cohorts from France, Norway and the USA: a pooled

analysis from the AGRICOH consortium. Int J Epidemiol. 2019;

48(5):1519-1535.

16
 This large study presents evidence suggestive that associations of pesticide

exposure with NHL may be NHL-subtype and chemical specific.

34. Fortes C, Mastroeni S, Segatto M M, Hohmann C, Miligi L, Bakos L,

Bonamigo R. Occupational Exposure to Pesticides With

Occupational Sun Exposure Increases the Risk for Cutaneous

Melanoma. J Occup Environ Med. 2016;58(4):370-5.

35. Segatto MM, Bonamigo RR, Hohmann CB, Müller KR, Bakos L,

Mastroeni S, Fortes C. Residential and occupational exposure to

pesticides may increase risk for cutaneous melanoma: a case-

control study conducted in the south of Brazil. Int J Dermatol.

2015;54(12):e527-38

36. * Conti CL, Barbosa WM, Simão JBP, Álvares-da-Silva AM. Pesticide

exposure, tobacco use, poor self-perceived health and presence of

chronic disease are determinants of depressive symptoms among

coffee growers from Southeast Brazil. Psychiatry Res. 2018; 260:187-

192.

Illustrate the use a cross-sectional survey to study risk factors of a long-term

disorder (depression).

37. Campos E, Dos Santos Pinto da Silva V, Sarpa Campos de Mello M,

Barros Otero U. Exposure to pesticides and mental disorders in a

rural population of Southern Brazil. Neurotoxicology. 2016;56:7-16.

38. Cremonese C, Piccoli C, Pasqualotto F, Clapauch R, Koifman RJ,

Koifman S, Freire C. Occupational exposure to pesticides,

reproductive hormone levels and sperm quality in young Brazilian

men. Reprod Toxicol. 2017;67:174-185

17
39. ** Aylward LL. Integration of biomonitoring data into risk

assessment. Current Opinion in Toxicology.2018, 9:14-20.

Interesting analysis of the integration of biomonitoring data into risk assessment

and management. The example (figure) of changes in contine concentrations in

the US population in 2 time Windows (1988-1991 and 2013-2014) is particularly

illustrative of the use of biomonitoring data to assess the effectiveness of public

health interventions.

LEGEND TO FIGURE

Figure. Analytic epidemiological approaches test hypotheses on the

determinants (risk factors) of adverse health outcomes or diseases. Owing to

methodological limitations inherent to the study design, it is generally difficult to

rule out noncausal explanations for associations dectected by cross-sectional (a

snapshot of characteristics of study subjects in a single point in time) and case-

control studies. Prospective cohort studies (groups based on exposure status

followed over time), on the other hand, provide the strongest evidence of

causality of observed associations.

18
19
Evidence hierarchy of epidemiological study designs for
hazard identification in pesticide risk assessment

Prospective cohort studies

Retrospective cohort studies

Case-control studies

Cross-sectional studies
(with comparison) Increasing weight
of evidence
Cross-sectional studies
Descriptive

(with no comparison)

Ecological studies

Case series

Case study / report

Please check the following as appropriate:

o All authors have participated in (a) conception and design, or analysis and
interpretation of the data; (b) drafting the article or revising it critically for
important intellectual content; and (c) approval of the final version.

o This manuscript has not been submitted to, nor is under review at, another
journal or other publishing venue.

o The authors have no affiliation with any organization with a direct or indirect
financial interest in the subject matter discussed in the manuscript

o The following authors have affiliations with organizations with direct or

Author’s name Affiliation