Table 2 Characteristics of the included prospective studies (by decreasing year of publication)

Sandoval-Insausti et al., 2022 Environ. Int. [17]

United-States

137,378 W (NHS, 1998–2019, and NHSII, 1999–2019) and 23,502 M (HPFS, 1998–2020) free of CVD, cancer, or diabetes at baseline.

27,026 deaths, 4,318 from CVD, 6,426 from cancer, 3,081,360 person-years of follow-up.

Cumulative exposure over time of F&V (FFQ) classified as high and low-pesticide residues using the PRBS, a validated score system (contamination data from USDA PDP).

F&V with PRBS ≥ 4 classified as having high-pesticide-residue status, F&V with PRBS < 4 as low-pesticide-residue status. F&V without matching PDP data as having undetermined -pesticide residue status.

Two different exposures (i.e. F&V intake, considering pesticide residue status): high-pesticide and low-pesticide-residue F&V intakes.

All-cause mortality.

Death recorded from state vital records, National Death Index, and reports from family members or the postal authorities. Classification of causes according to ICD-8.

Multivariable Cox model (HR, 95%CI). Confounding factors: age (time scale), BMI, race, physical activity, family history of cancer and CVD, smoking (package/y), baseline HTA, hypercholesterolemia, total energy intake, alcohol intake, modified aHEI (excluding F&V and alcohol) and hormone use for menopaused W.

Intakes of high-, low- and undetermined-pesticide residue F&V were simultaneously included in all models.

Cote et al., 2022 Am J Epidemiol [18]

United-States

121,701 W from NHS nurses and 116,686 W from NHSII and 51,529 M from HPFS.

A total of 275 glioma cases / 2,745,862 person-years.

Cumulative exposure over time of F&V (FFQ) classified as high and low-pesticide residues using the PRBS, a validated score system (contamination data from USDA PDP).

F&V with PRBS ≥ 4 classified as having high-pesticide-residue status, F&V with PRBS < 4 as low-pesticide-residue status. F&V without matching PDP data as having undetermined -pesticide residue status.

Two different exposures (i.e. F&V intake, considering pesticide residue status): high-pesticide and low-pesticide-residue F&V intakes.

Self-report primary brain malignancy cases confirmed by medical record review by study investigators as diagnosis of code 191 ICD 9^th

Multivariable Cox model (HR, 95%CI).

Age as time scale stratified by calendar time. Confounding factors: age (time scale), BMI, smoking status, alcohol intake, race, physical activity, and a modified aHEI score that did not include the fruit, vegetable, and alcohol intake components, energy intake.

In a supplementary model, adjustment for total flavonoid intake.

Intakes of high-, low- and undetermined-pesticide residue F&V were simultaneously included in all models

Rebouillat et al., 2022 Env Health [19]

France

33,013 participants, mean age 53 years old, including 76% of women.

340 incident T2D cases, 169,904.28 person-years, median follow-up = 5.95 years.

Combination of baseline dietary intakes (organic/conventional consumption) with pesticide residue concentration values in plant foods.

Contamination data from the European Union reference laboratory for pesticides, CVUAS.

Use of WHO scenario for exposure estimation.

25 pesticides and NMF used for extraction of pesticide exposure profiles: 4 NMF-factors.

Self-report of T2D and validation using medico-administrative database, drug use and official database (CePIDC) for vital status.

Multivariable Cox model for T2D incidence (HR, 95%CI).

Confounding factors: age (time scale), gender, physical activity, smoking status, educational level, occupation, household income, marital status, alcohol-free energy intake, family history of diabetes, weight, height, overall quality of the diet (measured by the PNNS-GS2 score).

Additional models adjusted for provegetarian score and three scores assessing quality of foods, cDQI, pDQI and aDQI.

Test for interactions between gender and sPNNS-GS2 (i.e. potential modulating factors) and NMF components.

Sandoval-Insausti et al., 2021 Environ Int [20]

United-States

150,830 W (NHS, 1998–2016, and NHSII, 1999–2017) and 29,486 M (HPFS, 1998–2016) without a history of cancer.

23,678 incident cancer cases, 2,862,118 person-years of follow-up. 14 years of follow-up.

Cumulative exposure over time of F&V (FFQ) classified as high and low-pesticide residues using the PRBS, a validated score system (contamination data from USDA PDP).

F&V with PRBS ≥ 4 classified as having high-pesticide-residue status, F&V with PRBS < 4 as low-pesticide-residue status. F&V without matching PDP data as having undetermined -pesticide residue status.

Two different exposures (i.e. F&V intake, considering pesticide residue status): high-pesticide and low-pesticide-residue F&V intakes.

Self-report of cancer confirmed by review of medical and pathology records (histology, anatomic location, and stage) by study physicians. For prostate cancer, only advanced prostate cancer was considered. Primary outcome = total cancer incidence.

Selected specific sites as secondary outcomes.

Multivariable Cox model (HR, 95%CI). Confounding factors: age, height, BMI, ethnicity, physical activity, family history of cancer, physical examination in the past 2 years, history of colonoscopy or sigmoidoscopy, smoking in packyears, current multivitamin use, regular aspirin use, total energy intake, alcohol intake, and aHEI score, hormone use for menopaused W and mammography in the past 2 years in NHS and NHSII and for prostate-specific antigen testing in the past 2 years in HPFS. Intakes of high-, low- and undetermined-pesticide residue F&V were simultaneously included in all models.

Rebouillat et al., 2021 Int J Epidemiol [21]

France

13,149 women mean age at baseline was 60.5 years (SD = 7.39).

169 incident breast cancer cases, 57 203.70 person-years, median follow-up = 4.83 years.

Combination of baseline dietary intakes (organic/conventional consumption) with pesticide residue concentration values in plant foods.

Contamination data from the European Union reference laboratory for pesticides, CVUAS.

Use of WHO scenario for exposure estimation.

25 pesticides and NMF used for extraction of pesticide exposure profiles: 4 NMF-factors.

Self-report of postmenopausal breast cancer and validation using medical records and official database (CePIDC) for vital status.

Multivariable Cox for cancer incidence (HR, 95%CI).

Confounding factors: age (time scale), smoking status, educational level, alcohol intake, alcohol-free energy intake, physical activity, BMI, height, family history of cancer, menopausal treatment, parity, overall quality of the diet (measured by the PNNS-GS2 score).

Additional models with adjustments for residing currently in an agricultural area, the level of ultra-processed foods in the diet and the provegetarian score.

Test for interactions between BMI, sPNNS-GS2 and the level of plant-based (i.e. modulating factors) and NMF components.

Chiu et al., 2019 Environ Int [22]

United-States

145,789 W and 24,353 M free of CVD and cancer from the NHS (NHS: 1998–2012), the NHS-II (1999–2013), and the HPFS (HPFS: 1998–2012).

3,707 incident CHD cases, 2,241,977 person-years.

Cumulative exposure over time of F&V (FFQ) classified as high and low-pesticide residues using the PRBS, a validated score system (contamination data from USDA PDP).

F&V with PRBS ≥ 4 classified as having high-pesticide-residue status, F&V with PRBS < 4 as low-pesticide-residue status. F&Vs without matching PDP data as having undetermined -pesticide residue status.

Two different exposures (i.e. F&V intake, considering pesticide residue status): high-pesticide and low-pesticide-residue F&V intakes.

Validation of cases based on medical records or telephone interviews. Non-fatal myocardial infarction confirmed using WHO criteria on the basis of symptoms plus elevated specific cardiac enzymes or electrocardiogram changes indicative of new ischemia.

Multivariable Cox model (HR, 95%CI). Confounding factors: age (time scale), BMI, race, physical activity, family history of myocardial infarction/diabetes, smoking status, baseline HTA, hypercholesterolemia, and diabetes, current multivitamin use, current aspirin use, total energy intake, alcohol intake, aHEI (excluding F&V and alcohol), hormone use for menopaused W (in NHS and NHSII), oral contraceptive (in NHSII). Intakes of high-, low- and undetermined-pesticide residue F&V were simultaneously included in all models.