Translate this page into:
Pesticide exposure associated with mild cognitive impairment & depression among agricultural workers: Case control study in rural India
Present address: †Department of Allied Health Sciences, Brainware University, Kolkata, West Bengal, India
For correspondence: Prof Amit Chakrabarti, Department of Mental Health, ICMR-Centre for Ageing and Mental Health, Kolkata 700 091, West Bengal, India e-mail: amit.chakrabarti@gov.in
-
Received: ,
Accepted: ,
Abstract
Background & objectives
This study was carried out to identify association of chronic pesticide exposure with cognitive impairment, depression, and movement disorder by utilising a comprehensive paradigm of symptom-based validated screening; and estimation of RBC acetylcholinesterase (AChE), plasma butyrylcholinesterase (BChE) and paraoxonase 1 (PON1) levels among high risk rural population, particularly agricultural workers in West Bengal, India.
Methods
Galsi II block, Purba Bardhaman district, West Bengal was selected as the study site. Samples were drawn among residents aged 50 yr and above of either sex, continuously living in the area for at least last five years. Primary screening was applied to enquire about target neuropsychiatric disorders. On getting an affirmative reply to any of the questions, participants were further screened for cognitive impairment and depression. Any participant screening ‘positive’ for either of the three conditions fulfilled criteria for recruitment as a ‘case’. Blood samples were collected from participants for estimation of AChE, BChE and PON1.
Results
Of 808 participants, 180 (22.3%) were screened as ‘Cases’. Pesticide exposure was a significant risk for detection as a ‘Case’ [Odds ratio (OR)=2.9], which supports premise of this study. Additionally, male gender (OR=1.98), more years of pesticide exposure (>30 yr) (OR=1.8) and more frequent use of pesticides (1/week) (OR=2.5) were significant risks of screening as a ‘Case’. Agricultural workers, who were exclusively engaged with agriculture as occupation, had the lowest cognition score (77.5±19.6) (P<0.001), highest impairment in daily living (2.1±1.7) (P=0.01) and highest BDI (Beck Depression Inventory) score (15±7) (P=0.003), i.e., a higher risk of depression. Paraoxonase 1 (PON1) expression was significantly increased among workers engaged in agricultural activity for more than 8 h in a day (82.3±42) (P=0.03) and who applied pesticides at least once a week (80.3±40) (P=0.02).
Interpretation & conclusions
This population-based study among rural agricultural population in West Bengal has identified a prevalence of 18.9 per cent of cognitive impairment with or without depression; 8.3 per cent depression with or without cognitive impairment and 1.5 per cent possible movement disorder. We also identified that pesticide exposure is a risk for development of these neuropsychiatric disorders of neuroinflammatory origin. Among biomarkers, PON1 was identified to be significantly higher among participants who spent more hours in agricultural work and applied pesticides more frequently.
Keywords
Case control
cognitive impairment
depression
pesticide
paraoxonase1 (PON1)
organophosphorus
neuroinflammation
Indian population is at risk of chronic environmental pesticide exposure from agricultural sources. India is a developing economy with 54.6 per cent of workforce engaged in agriculture sector; total number of agricultural workers increasing from 234.1 million in 2001 to 263.1 million in 20111. Globally of all the chemicals used in agriculture, insecticides constitute only 44 per cent, whereas, in India insecticides are the major chemicals used in agriculture and constitute almost 76 per cent2. For example, chlorpyriphos (1,872) and quinalphos (1,474), both organophosphorus compounds, were among the top pesticides in demand (in metric ton) in India during 2009 - 20103. This difference in trend of pesticide usage with more emphasis on use of easily available and less expensive organophosphorus (OP) pesticides creates the possibility of population level harmful pesticide exposure with rural population, particularly agricultural workers at higher risk.
An association between pesticide exposure and common neuropsychiatric disorders of neuroinflammatory origin can be hypothesised among high risk Indian population. Levels of RBC acetylcholinesterase (AChE) and plasma butyrylcholinesterase (BChE), the enzymes catalysing acetylcholine synthesis have been established as effective biomarkers of pesticide exposure at population level4. Paraoxonase 1 (PON1) hydrolyzes OP pesticides and its plasma level is another useful biomarker of pesticide exposure5.
West Bengal is the fourth most populous State in India. With nearly 69 per cent (as per 2011 census) population living in rural areas, agriculture is the chief occupation in West Bengal.
The present case-control study aimed to predict association of chronic environmental pesticide exposure with common neuropsychiatric disorders of neuroinflammatory origin by utilising a comprehensive paradigm of symptom-based validated screening for cognitive impairment, depression and movement disorders; and estimation of AChE, BChE, PON1 levels among high risk rural population, particularly agricultural workers in West Bengal, India.
Materials & Methods
This case control study was undertaken by the department of Mental Health, ICMR-Centre for Ageing and Mental Health (erstwhile Regional Occupational Health Centre Eastern), Kolkata, West Bengal, India after obtaining the ethical approval by the Institutional Ethics Committee (IEC). All issues related to protection of participants were ensured according to the guidelines of Indian Council of Medical Research (ICMR), New Delhi, India. All participants underwent informed consent process before consenting to participate in the study. The study was conducted in the Purba Bardhaman district of West Bengal. Purba Bardhaman with a population size of 77,17,563 with 60 per cent rural population (2011 census) and 31 blocks with 2,438 villages was chosen as the study site as it is one of the most agriculturally productive districts6 in West Bengal.
Study site
All blocks in Purba Bardhaman district were classified according to pesticide use in kg per hectare during preceding five years. A high pesticide use block was chosen according to convenience. Further, according to secondary information, during 2008-2009 Galsi II, Galsi I, Ausgram II, Ketugram I, Khandaghosh, Bhatar, Mangalkote, Raina I and Raina II blocks were agriculturally most active, based on total land use and total agricultural produce7. Secondary evidence suggested that blocks most active in agriculture were also expected to be high-pesticide consumption blocks.
Galsi II block, Purba Bardhaman district, West Bengal was selected as the study site. This block was identified as one of the most agriculturally active blocks in this district. Galsi II has around 1,000 villages with ⁓1,50,000 population7. The site of the study was villages around the Galsi II Block Primary Health Centre (BPHC) at Adrahati, where approximately 400 to 500 people from different parts of the block assemble each day for treatment services. The nearby villages where household sampling was carried out were Belan, Goromba, Adrahati and Serorai. During the household visits and sampling procedure the research staffs were assisted by village peers as well as Accredited Social Health Activists (ASHAs), which was coordinated at the Adrahati BPHC in discussion with health care staff at the BPHC.
Recruitment of participants and assessments was initiated in the Galsi II block area from end of September 2018 and was continued till February 2020, when the field work was called off due to the COVID-19 pandemic.
Sampling
Samples were drawn from residents in the sampling areas aged 50 yr and above of either sex, continuously living in the area for at least last five years. Sampling locations were households and sampling strategy consisted of systematic random sampling, e.g., visiting every 3rd or 5th household. Sampling was further balanced across all locations to recruit up to 40 per cent as women participants.
Procedure
During random screening at the sampling locations, possible participants were asked about their age and duration of stay in the area. The age was determined according to the official documents. When there was no available document, the age was determined by asking a historical event relating to the participant’s birth. A set of general questionnaire-based assessment (described below in ‘instruments’) as primary screening was applied to enquire about possible neurodegenerative diseases to participants above 50 yr of age. These answers were also verified from his/her family member/s. Broadly, the questions included change in a person’s memory, mood and motor problem observed over last six months or more by the participant or by his/her family member/s.
On getting an affirmative reply to any of the above questions the participants were administered further screening instruments for cognitive impairment and depression. In this study we used the Kolkata Cognitive Screening Battery for subjects who were detected as ‘Cases’ after primary screening. We also used the Bengali version of the Beck Depression Inventory (BDI) to look for possible depression. The Kolkata Cognitive Screening Battery is a culturally validated tool developed and used for epidemiological study in West Bengal8.
Any participant screening ‘positive’ for either of three diseases fulfilled criteria for recruitment as a ‘case’. Any participant screening ‘negative’ for all three diseases was recruited as a ‘control’. The selected participants were offered to go through the informed consent process and participate in the study. The research staff ensured to recruit a maximum of two persons qualifying as ‘control’ from each household. However, no ‘control’ was recruited from the household of a ‘case’.
Following this procedure, the socio-demographic instrument including exposure status, i.e., ‘direct’ or ‘indirect’ was completed. The Knowledge attitudes and practices (KAP) instrument was administered to both ‘cases’ and “controls” reporting any ‘direct’ exposure to pesticides. Next step was collection of blood by venipuncture in an EDTA vial (3 mL) and a serum vial (3 mL), which was carried out by a trained para-medical staff with full aseptic precautions. Following completion of study procedure, the participants were briefed about health issues related to pesticide exposure; occupationally exposed participants were briefed about prevention issues; and any person with any positive disease signal was referred to the nearest treatment facility. All these processes were incorporated in the informed consent procedure. Locator information of all participants was securely maintained for follow up surveillance, if any, in future.
Instruments
Following instruments were used after recruitment on completion of the informed consent procedure. All instruments were administered in Bengali language.
Screening for neurodegenerative diseases
Included validated history/symptom based screening tool. If found positive, the participant was assessed further using following instruments.
(i) Kolkata cognitive screening battery for participants with suspected dementia
This instrument incorporates components on verbal fluency, naming test, elements from the Hindi version of the Mini Mental State Examination (MMSE), delayed word list memory, and delayed recognition word8-10.
(ii) Beck depression inventory for depression
A set of 21 questions describing recent feelings of the participants. A cumulative score of 11 and above suggests mood disturbance to severe depression11.
(a) Socio-demographic instrument
Semi-structured and previously used during other studies; was used after further modification suited to the present study. Population indirectly exposed to pesticides was defined by self-report of lifetime occupational non-exposure. Population directly exposed to pesticides was defined by self-report of occupational exposure at least past five years.
(b) KAP instrument
This instrument regarding use of pesticides was adapted from a Brazilian study12 for administration to only directly exposed ‘case’ and ‘control’ participants.
Biomarkers
Following biomarkers of organophosphorus exposure were analysed among ‘Cases’ and ‘Controls’.
1. Acetylcholinesterase (AChE) and butyrylcholinesterase (BChE) estimation
Acetylcholinesterase and butyrylcholinesterase were estimated by spectrophotometric method based on modified Ellman method13; AChE and BChE were expressed as U/L.
2. Paraoxonase 1 (PON1) enzyme estimation
PON1 activity was determined by using 4-nitrophenyl acetate14. The result was expressed in terms of IU/L.
Sample size
Considering the existing population size of Purba Bardhaman district; its distribution among 31 blocks; one randomly selected rural block is expected to have ⁓1,50,000 population. Available evidence suggests a minimum 10 per cent combined prevalence of depression, dementia and Parkinson’s disease among Indian population aged 50 and above. Therefore, to detect a minimum 10 per cent prevalence of common neuropsychiatric diseases at 95 per cent confidence interval (CI), we needed to screen at least 138 persons. For the case control study, we assumed a minimum 10 per cent difference of predictor variables (epidemiological and biomarkers) between two study groups. To detect this difference at a 95 per cent two-sided confidence interval, 80 per cent power and a recruitment ratio of 1:1 for controls to cases with adequate power the study needed to recruit at least n=199 ‘controls’ and n=199 ‘cases’. This sample size is also expected to generate acceptable risk estimation (OR=2.25). The recruitment target, therefore, was 225 ‘controls’ and 225 cases from the selected block. To recruit 225 cases, the population to be screened shall be ∼3,500 individuals from the block. At an average rate of 15 screenings per day; ∼230 days were required to complete the study processes. Therefore, an allocation of 12 months was considered adequate for the field work. In case of higher prevalence, recruitment target was likely to be reached by screening less population. This was also expected to provide an idea about the magnitude of the problem (estimate of prevalence) in the selected block.
Data management and statistical analysis
Frequency distribution was used to describe demographic and KAP variables and any significant difference among the study groups was identified using Chi-Square test. Relationship between outcome variables, i.e., screening as case vs. control as well as identification of separate categories of neurological conditions, viz., cognitive impairment, depression, both cognitive impairment and depression and possible movement disorder with predictor variables identifying risks related to agricultural activity and pesticide use was expressed as OR with 95 per cent CI. For binary variables case was coded as 1 and control as 0. Categorical variables were coded numerically assigning lower number for lower risk and higher number for higher risk, e.g., exposure (1=no, 2=yes) and likewise. Further, relationship between outcome variables, i.e., screening as case vs. control with continuous variables, was analysed by one way analysis of variance (ANOVA). Logistic regression was used to identify significant predictor variables (related to agricultural activity and pesticide use) of identification as a case. Regression analysis was controlled for age, gender and ethnicity (Bengali and Tribal). In regression model variables were entered stepwise. Regarding other possible covariates we could not get validated information on nutritional status and comorbidities. Further, on preliminary enquiry we understood that education level and economic status of agricultural workers in the given area are almost uniform. Thus, these variables were not considered in regression analysis. Correlation matrix showed that the pairwise correlation coefficients of predictor variables were not close to +1 or -1; thus ruling out multicollinearity. Further, a non-significant Hosmer-Lemeshow test suggested adequate goodness-of-fit of the regression model. The events-per-variable (EPV) was 13; thus meeting the minimum criteria of 10, suggesting model stability. Acceptable level of significance was set at P≤0.05. Data were analysed by IBM SPSS Statistics 22.0 (IBM, New York, USA).
Results
The recruitment and assessments had to be called off from February 2020 due to the COVID-19 pandemic. Therefore, up to suspension of field work 808 participants could be. Of these, 623 (77.1%) were males and n=185 (22.9%) were females. Following screening 180 (22.3%) participants were screened positive for cognitive impairment, depression, both cognitive impairment and depression and possible movement disorder (Cases). Of these, 101 (56.1%) screened positive cognitive impairment, 15 (8.3%) were positively screened for depression, 12 (6.7%) with possible movement disorder and 52 (28.9%) with both cognitive impairment and depression. Another 181 participants were included as controls. The mean age of the cases was 57.7±6.8 yr and that of the control was 57.1±6.7 yr (P=0.35).
Table I highlights the important demographic characteristics along with occupational and pesticide use behaviour among participants. It may be noted that agricultural workers (n=268) include farmers who cultivate their own land (n=85) as well as sharecroppers (n=183). Use of combination pesticides signifies use of more than one pesticide for agriculture. Participants in this study using combination pesticides were using combinations of two to four types of pesticides. Organophosphorus compounds were the commonest pesticide used by the workers.
| Variables | Frequency, n (%) |
|---|---|
| Gender (n=361) | |
|
Male Female |
251 (69.5) 110 (30.5) |
| Education (n=361) | |
|
School & above Illiterate |
123 (34.1) 238 (65.9) |
| Occupation (n=361) | |
|
Worker Owner |
268 (74.2) 93 (25.3) |
| Type of agriculture (n=329) | |
|
Grain Both |
160 (48.6) 169 (51.4) |
| Ethnicity (n=361) | |
|
Bengali Tribal |
239 (66.2) 122 (33.8) |
| Pesticide exposure history (n=361) | |
|
Yes No |
318 (11.9) 43 (88.1) |
| Duration of agricultural activity (n=346) | |
|
<30 yr > 30 yr |
189 (54.6) 157 (45.4) |
| Hours of agricultural work (n=343) | |
|
<8 h > 8 h |
112 (32.7) 231 (67.3) |
| Frequency of pesticide application (n=250) | |
|
1-2/month 1/wk |
204 (81.6) 46 (18.4) |
| Use of combination pesticides (n=318) | |
|
No No idea Yes |
49 (15.4) 93 (29.2) 176 (55.3) |
Table II demonstrates that pesticide exposure was a significant risk for cases. Male gender, more years of pesticide exposure (>30 yr) and more frequent use of pesticides (1/wk) were all significant risks for screening as a case.
| Participant characteristic | Cases (n=180), n (%) | Control (n=181), n (%) | OR (95% CI) | P value |
|---|---|---|---|---|
| Male gender | 138 (76.7) | 113 (62.4) |
1.98 (1.25-3.12) |
0.003 |
| Exposed to pesticides | 168 (93.3) | 150 (82.8) |
2.89 (1.43-5.84) |
0.002 |
| Exposed as worker | 134 (50) | 134 (50) |
1.02 (0.64-1.64) |
0.80 |
| Exposure duration >30 yr | 90 (52,6) | 67 (38.3)) |
1.79 (1.17-2.75) |
0.007 |
| Agricultural work >8 h/day | 115 (67.6) | 116 (67) |
1.03 (0.65-1.61) |
0.91 |
| Pesticide use frequency at least once/wk | 32 (24.8) | 14 (11.6) |
2.52 (1.27-5) |
0.007 |
OR, odds ratio; CI, confidence interval
The relationship between predictor variables related to agricultural activity and pesticide use with screening outcome by disease category, i.e., cognitive impairment, depression, movement disorder and both cognitive impairment and depression, was also examined. Significant differences were observed between screened disease categories and controls by gender (P<0.01), longer duration of pesticide exposure (>30 yr) (P<0.001), more hours of agricultural work per day (>8 h) (0.05) and higher frequency of pesticide use (P<0.02). Lifetime alcohol use also has a significant relation with screened neurological conditions (P=0.002). Regarding ethnicity tribal population are likely to have a greater chance of being screened with cognitive impairment.
The Kolkata Cognitive Screening Battery, BDI and EASI were administered to only the ‘case’ population (n=180). The distribution of scores of these three tools among the case population is given below.
For this rural population participating in this study, cognitive ability scores between 50th and 25th percentile, i.e., between 93 and 74.5 can be considered as mild cognitive impairment (MCI) and scores below 25th percentile can be regarded as significant cognitive impairment. For BDI where higher scores signify higher risk of clinically significant depression the scores between 50th and 25th percentile, i.e., between 12 and 7 can be considered as ‘normal’; scores between 75th and 50th percentile, i.e., between 16 and 12 can be considered as ‘borderline’ risk for depression. As EASI is rated in a scale of 0 to 1, a cumulative score of 0 suggests no impairment of daily activities and higher scores suggest more impairment.
Significantly low cognition score (between 50th to 25th percentile, qualifying as MCI) was observed among participants with only cognitive impairment, both cognitive impairment and possible depression and possible movement disorder (P=0.02). Impairment of daily activities was maximum among participants with only cognitive impairment and both cognitive impairment and possible depression (P<0.001). BDI score was highest among participants with both cognitive impairment and possible depression (P<0.001).
Further analysis showed cognition score was significantly less among women ‘cases’ (82.4±20.1) in comparison to male participants (91.1±19.7) (P=0.01). Cognition scores also showed an ethnic difference with tribal population (80.9±20.8) having significantly lower scores than Bengali participants (93.4±18.3) (P<0.001). It was interesting to observe that agricultural workers, who are exclusively engaged with agriculture as occupation (n=85) have the lowest cognition score (77.5±19.6) (P<0.001), highest impairment in daily living (2.1±1) (P=0.01) and highest BDI score (15±7) (P=0.003), i.e., a higher risk of depression.
Biomarkers of organophosphorus pesticide exposure, viz., AChE, BChE and PON1 were estimated in a sub sample of the participants (Table III). No significant difference was observed between ‘cases and controls’ for all these biomarkers. No significant difference was observed among ‘control’ participants and participants screened positive for any of the individual neurological conditions. It was also observed that PON1 expression was significantly increased among workers who were engaged in agricultural activity for more than 8 h in a day (82.3±42) (P=0.03) as well as higher frequency of pesticide use (114.8±41.2) (P=0.02).
| Screening category of cases | F | P | ||||||
|---|---|---|---|---|---|---|---|---|
| Control | Cognition | Depression | Movement | Both | ||||
| BChE | Mean | 5772.4 | 5910.4 | 6481.7 | 5885.1 | 5109 | 0.58 | 0.68 |
| SD | 1340.6 | 1850.9 | 1726.8 | 1890.5 | 1335.8 | |||
| n | n=52 | n=36 | n=8 | n=6 | n=4 | |||
| AChE | Mean | 9148.3 | 8295.9 | 6740.9 | 9175.5 | 6623.9 | 0.80 | 0.53 |
| SD | 5011.6 | 3071.9 | 2810.1 | 2915.7 | 3118.7 | |||
| n | n=52 | n=32 | n=6 | n=6 | n=4 | |||
| PON1 | Mean | 74.1 | 86.7 | 57.7 | 41.5 | 67 | 1.74 | 0.15 |
| SD | 41.4 | 41.5 | 19.3 | 26.4 | 8.9 | |||
| n | n=39 | n=30 | n=6 | n=4 | n=4 | |||
SD, standard deviation
In logistic regression controlled for age, gender, ethnicity and lifetime alcohol use (Table IV) higher frequency of pesticide use (at least once per wk) was the most consistent predictor of screening as a ‘case’, i.e., likely to have any one of the common neurological conditions.
| Baseline Predictors | Screening as ‘Case’ | ||
|---|---|---|---|
| Standardised beta | P | OR (95% CI) | |
| Gender | 0.80 | 0.01 | 1.98 (1.25-3.12) |
| Ethnicity | -0.38 | 0.27 | 1.17 (0.76-1.81) |
| Lifetime alcohol use | 0.19 | 0.56 | 0.67 (0.41-1.10) |
| Exposed to pesticides | 21.89 | 0.10 | 2.89 (1.43-5.84) |
| Nature of occupation | -0.52 | 0.16 | 1.02 (0.64-1.64) |
| Duration of exposure | 0.49 | 0.20 | 1.79 (1.17-2.75) |
| Hours of work / day | -0.54 | 0.14 | 1.03 (0.65-1.61) |
| Frequency of pesticide use | 0.75 | 0.05 | 2.52 (1.27-5) |
OR, odds ratio; CI, confidence interval
Discussion
Although our study could screen only 808 participants we still have an approximation of magnitude of the problem of common neurological conditions of neuroinflammatory origin in rural Indian population. Our results show that 12.5 per cent were identified with only cognitive impairment after primary and secondary screening. The prevalence of cognitive impairment in Indian urban elderly has been identified in the range of 8.8 to 10 per cent15,16. The reported population based studies were carried out in both southern15 and northern16 India. Among community based urban elderly of 50 yr and above in eastern India using the Kolkata Cognitive Screening Battery, i.e., the tool used in our study, the estimated prevalence of mild cognitive impairment was 14.9 per cent17, which is similar to our observations, although our study was conducted in a rural setting. The presence of depression along with cognitive impairment also needs to be taken in to consideration as an independent entity. Systematic review and meta-analysis have identified that the co-existence of depression along with mild cognitive impairment is high and could be up to 32 per cent among patients with mild cognitive impairment in community and clinic-based settings18. Our study reported that 33 per cent of participants with cognitive impairment were identified to be having problems of depression, which is very similar to available evidence. The prevalence depression in our study of 1.85 per cent is much lower than 34.4 per cent reported earlier in Indian elderly aged 60 yr and above with higher risks among females, rural populations and in eastern India19. Although our observations point to a much lower prevalence, if we consider the sample having both depression and cognitive impairment the prevalence can be up to 8.3 per cent (n=67). Our study has come up with a prevalence of 1.5 per cent for possible movement disorder in rural elderly. Although the screened participants cannot be conclusively diagnosed; the prevalence of Parkinson’s disease internationally is in the range of 0.6 to 2.6 per cent20 in elderly with higher risk with increasing age.
Our study has identified that pesticide exposure, more year of exposure, more frequency of exposure are risks for cognitive impairment and/or depression with males at higher risk. Globally an increasing trend in dementia can be linked with increasing use of pesticides starting decades ago raising the possibility that such exposure may, partly, be associated in pathogenesis of cognitive impairment21. Similar evidence is also available from different countries of the world. Among Korean farmers22; Greek elderly population23; older Mexican Americans24; and in a study from South America (Chile)25 pesticide exposure has been associated with poor cognitive function. Our study also identifies that lifetime alcohol use is likely to have a relation with only cognitive impairment whereas the risk is attenuated among participants screened positive for both cognitive impairment and co-existing depression. Internationally systematic scoping review does show that heavy alcohol use is a risk for cognitive impairment at the population level26.
Our study has shown that the risk of these neurological conditions is higher among men but cognition score is significantly lower in females, possibly suggesting greater severity, although within the range of MCI. Available evidence suggests that possibility of higher severity of cognitive impairment among women could be related to differences in socioeconomic status between two genders27. Cognition score was again significantly lower among tribal population subgroup compared to Bengali subgroup, although within the range of MCI. Such ethnic differences have been observed in Chinese population among Han and Hui ethnic groups28. Interestingly in our study agricultural workers, who are exclusively engaged only in agricultural occupation have the lowest cognition score and highest BDI score compared to owners (indirectly exposed) and sharecroppers (also engaged in other occupations apart from agriculture). This highlights the issue that agricultural workers directly exposed to organophosphorus pesticides have highest risk of cognitive impairment as well as possibility of depression. Impairment in daily living is also significantly higher in this subgroup compared to owners and sharecroppers. Evidence also suggests that directly exposed agricultural workers have the highest risk of cognitive impairment compared to indirectly exposed populations25.
AChE and BChE inhibition signifies exposure to organophosphorus pesticides and has also been associated with poor neuropsychological functioning in population-based studies29. Suppression of PON1 is a measure of organophosphorus pesticide exposure and PON1 suppression along with suppression of AChE and BChE frequently co-exist among population exposed directly or indirectly to organophosphorus pesticides5,30,31. Decline in cognitive scores in different forms of cognitive impairment including mild cognitive impairment and dementia has also been documented to have an association with declining PON1 levels32,33. Available evidence also suggests that lower PON1 levels can be associated with major depression and could be a trait marker for the same34,35. In our study we did not find a significant difference between AChE, BChE and PON1 levels between case and controls. This observation is likely from the statistical underpowering of the study due to incomplete sampling arising out of the impasse due to the COVID-19 outbreak. Our original estimate was to collect 500 samples for AChE, BChE and PON1 from the total sample. However, we could only sample 100 for AChE, 106 for BChE and 83 for PON1.
Therefore, summarily this population-based study among rural agricultural population in West Bengal has identified a prevalence of 18.9 per cent of cognitive impairment with or without depression; 8.3 per cent depression with or without cognitive impairment and 1.5 per cent possible movement disorder. We have also identified that pesticide exposure is a risk for development of neurological disorders of neuroinflammatory origin. Among biomarkers PON1 is likely to predict higher levels of pesticide use.
Due to the COVID-19 outbreak sampling for symptom-based assessments as well as biological samples needed to be called off prematurely. Whereas our original target was to recruit n=275 cases and n=225 controls, we could end up with n=180 cases and n=181 controls. This under-sampling has statistically underpowered our study and possibly influenced the final outcome and interpretations.
Following primary and secondary screening of cases population identified with cognitive impairment or depression, or both, or with movement disorder; our plan was to carry out a clinical diagnosis at study site on designated dates by a trained neurologist so that the participants can receive proper clinical care. This was planned for the end phase of the study, which again could not be carried out due to the COVID-19 outbreak.
During conduction of the study we have identified a possible design error. All our participants reside within the same agricultural areas. Thus, the controls are also exposed to some indirect risk of pesticide exposure. Possibly we needed a third control group not living within agricultural area to rule out indirect agricultural exposure to organophosphorus pesticides. In subsequent study we shall rectify this error.
Finally to establish a causal relationship the most appropriate design needed to be a longitudinal approach with assessments of neurological status as well as biomarkers carried out over time in participants. However, the scope of the present study did not permit this approach.
Acknowledgment
Authors acknowledge Director of Health Services (DHS) and Director of Medical Education (DME), Department of Health & Family Welfare, Government of West Bengal for support in implementation of this study; Dr. Pranab Kumar Roy, Chief Medical Officer of Health (CMOH), Purba Bardhaman for assistance at the block level; Block Medical Officers (BMOs) at the Adrahati BPHC, Galsi II, particularly Dr. Raju Sana for his consistent support; all the health care staff at the Adrahati BPHC including the Accredited Social Health Activists (ASHAs) who made the sampling possible; Debasis, Didi (Binadi), Jethu (Ayodhya Ghosh), Kakima and Biswanath Babu in Adrahati and Galsi II for assistance; and villagers of Belan, Goromba, Adrahati and Serorai in Galsi II block for their support.
Financial support & sponsorship
The study received financial support from the Department of Higher Education, Science & Technology and Biotechnology, Government of West Bengal, India (grant no. ST/P/S&T/9G-21/2017).
Conflicts of Interest
None.
Use of Artificial Intelligence (AI)-Assisted Technology for manuscript preparation
The authors confirm that there was no use of AI-assisted technology for assisting in the writing of the manuscript and no images were manipulated using AI.
References
- Census of India. Office of the Registrar General & Census Commissioner India. Available from: https://censusindiagovin/censuswebsite/, accessed on June 7, 2025.
- Future of Indian pesticides industry in next millennium. Pesticide Information. 1999;24:9-23.
- [Google Scholar]
- Directorate of Plant Protection, Quarantine & Storage; Department of Agriculture Cooperation, Ministry of Agriculture, Government of India. Pesticide and Documentation Unit. Statistics of Pesticides. Available from: http://ppqsgovin/IpmPesticideshtm, accessed on June 27, 2025.
- Cholinesterase activity, pesticide exposure and health impact in a population exposed to organophosphates. Int Arch Occup Environ Health. 2009;82:833-42.
- [CrossRef] [PubMed] [Google Scholar]
- Paraoxonase 1 and its relationship with pesticide biomarkers in indigenous Mexican farmworkers. J Occup Environ Med. 2014;56:281-90.
- [CrossRef] [PubMed] [Google Scholar]
- NABARD Consultancy Services (NABCONS). Department of Agriculture, Government of West Bengal. State agriculture plan for West Bengal (2017-18 to 2019-20). Available from: http://rkvynicin/sap/wb/wbpdf, accessed on June 27, 2025.
- Krishi Vigyan Kendra Burdwan Central Research Institute for Jute and Allied Fibres (Indian Council of Agricultural Research) Bud, Burdwan. Comprehensive district agriculture plan for Burdwan District (XIIth plan period). Available from: http://rkvynicin/dap/wb/CDAP-Burdwan%20Districtpdf, accessed on June 27, 2025.
- An urban community-based study of cognitive function among non-demented elderly population in India. Neurology Asia. 2006;11:37-48.
- [Google Scholar]
- “Mini-mental state”. A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res. 1975;12:189-98.
- [CrossRef] [PubMed] [Google Scholar]
- Development of an activities of daily living scale to screen for dementia in an illiterate rural older population in India. Age Ageing. 1999;28:161-8.
- [CrossRef] [PubMed] [Google Scholar]
- An inventory for measuring depression. Arch Gen Psychiatry. 1961;4:561-71.
- [CrossRef] [PubMed] [Google Scholar]
- Knowledge, attitudes, practices and biomonitoring of farmers and residents exposed to pesticides in Brazil. Int J Environ Res Public Health. 2012;9:3051-68.
- [CrossRef] [PubMed] [PubMed Central] [Google Scholar]
- A new and rapid colorimetric determination of acetylcholinesterase activity. Biochem Pharmacol. 1961;7:88-95.
- [CrossRef] [PubMed] [Google Scholar]
- A new automated method for phenotyping arylesterase (EC 3.1.1.2) based upon inhibition of enzymatic hydrolysis of 4-nitrophenyl acetate by phenyl acetate. Eur J Clin Chem Clin Biochem. 1992;30:391-5.
- [CrossRef] [PubMed] [Google Scholar]
- Prevalence of dementia in India: National and state estimates from a nationwide study. Alzheimers Dement. 2023;19:2898-912.
- [CrossRef] [PubMed] [PubMed Central] [Google Scholar]
- Estimation of the global prevalence of dementia in 2019 and forecasted prevalence in 2050: an analysis for the Global Burden of Disease Study 2019. Lancet Public Health. 2022;7:e105-e25.
- [CrossRef] [PubMed] [PubMed Central] [Google Scholar]
- An epidemiologic study of mild cognitive impairment in Kolkata, India. Neurology. 2007;68:2019-26.
- [CrossRef] [PubMed] [Google Scholar]
- Prevalence of depression in patients with mild cognitive impairment: a systematic review and meta-analysis. JAMA Psychiatry. 2017;74:58-67.
- [CrossRef] [PubMed] [Google Scholar]
- Prevalence of depression among the elderly (60 years and above) population in India, 1997-2016: a systematic review and meta-analysis. BMC Public Health. 2019;19:832.
- [CrossRef] [PubMed] [PubMed Central] [Google Scholar]
- Prevalence of Parkinson’s disease in Europe: a collaborative study of population-based cohorts. Neurologic Diseases in the Elderly Research Group. Neurology. 2000;54:S21-3.
- [PubMed] [Google Scholar]
- Linking pesticide exposure and dementia: what is the evidence? Toxicology. 2013;307:3-11.
- [CrossRef] [PubMed] [Google Scholar]
- Pesticide exposure and cognitive decline in a rural South Korean population. PloS one. 2019;14:e0213738.
- [CrossRef] [PubMed] [PubMed Central] [Google Scholar]
- Pesticide exposure and cognitive function: Results from the Hellenic Longitudinal Investigation of Aging and Diet (HELIAD) Environ Res. 2019;177:108632.
- [CrossRef] [PubMed] [Google Scholar]
- Cognitive decline, mortality, and organophosphorus exposure in aging Mexican Americans. Environ Res. 2018;160:132-9.
- [CrossRef] [PubMed] [PubMed Central] [Google Scholar]
- Cognitive impairment in agricultural workers and nearby residents exposed to pesticides in the Coquimbo Region of Chile. Neurotoxicol Teratol. 2017;62:13-9.
- [CrossRef] [PubMed] [Google Scholar]
- Alcohol use and dementia: a systematic scoping review. Alzheimers Res Ther. 2019;11:1.
- [CrossRef] [PubMed] [PubMed Central] [Google Scholar]
- Gender differences in cognitive impairment among the old and the oldest-old in China. Geriatr Gerontol Int. 2019;19:586-92.
- [CrossRef] [PubMed] [Google Scholar]
- Neuropsychiatric Symptoms and cognitive impairment in chinese patients with parkinson’s disease in Han and Hui ethnicity. Curr Med Sci. 2019;39:122-6.
- [CrossRef] [PubMed] [Google Scholar]
- Association between cholinesterase’s inhibition and cognitive impairment: a basis for prevention policies of environmental pollution by organophosphate and carbamate pesticides in Chile. Environ Res. 2020;186:109539.
- [CrossRef] [PubMed] [Google Scholar]
- Serum cholinesterase inhibition in relation to paraoxonase-1 (PON1) status among organophosphate-exposed agricultural pesticide handlers. Environ Health Perspect. 2009;117:1402-8.
- [CrossRef] [PubMed] [PubMed Central] [Google Scholar]
- Pesticide exposure and related health problems among family members of farmworkers in southeast Iran. A case-control study. Environ Pollut. 2020;267:115424.
- [CrossRef] [PubMed] [Google Scholar]
- PON-1 and ferroxidase activities in older patients with mild cognitive impairment, late onset Alzheimer’s disease or vascular dementia. Clin Chem Lab Med. 2015;53:1049-56.
- [CrossRef] [PubMed] [Google Scholar]
- Paraoxonase activity and dementia. J Neurol Sci. 2009;283:107-8.
- [CrossRef] [PubMed] [Google Scholar]
- Lowered plasma paraoxonase (PON)1 activity is a trait marker of major depression and PON1 Q192R gene polymorphism-smoking interactions differentially predict the odds of major depression and bipolar disorder. J Affect Disord. 2014;159:23-30.
- [CrossRef] [PubMed] [Google Scholar]
- Lowered PON1 activities are strongly associated with depression and bipolar disorder, recurrence of (hypo) mania and depression, increased disability and lowered quality of life. World J Biol Psychiatry. 2019;20:368-80.
- [CrossRef] [PubMed] [Google Scholar]