Pesticide exposure and rhinitis: A cross-sectional study among farmers in Pitsanulok, Thailand

Background Pesticide exposure has been suspected as a cause of rhinitis, a common disease that affects the health and well-being of millions of people around the world. This cross-sectional study aimed to examine the association between pesticide use and rhinitis prevalence among farmers in Phitsanulok province, Thailand. Methods Data on historical pesticide use and rhinitis were collected by an in-person interview questionnaire. Data from 9,649 participants were included in the analysis. The association between pesticide exposure and rhinitis was determined by multiple variable logistic regression, adjusted for potential confounding factors. Results The study found an association between pesticide exposure and the prevalence of rhinitis. The association was consistent across various types of pesticides (insecticides, herbicides, fungicides, rodenticides, and molluscicides) and individual pesticides. Some of the relationships were in a dose-response pattern. This finding was new as previous studies often reported the association of only a few specific pesticides. Conclusions The results from this large cross-sectional study supports existing literature on the potential effects of pesticides on rhinitis. In addition, the analysis showed that the rhinitis effect might be in fact related to the properties of the types of pesticides rather than individual chemical toxicity. The impact of pesticides on rhinitis should receive more attention from public health and other organizations responsible for the farmers’ health.


Results
The study found an association between pesticide exposure and the prevalence of rhinitis.The association was consistent across various types of pesticides (insecticides, herbicides, fungicides, rodenticides, and molluscicides) and individual pesticides.Some of the relationships were in a dose-response pattern.This finding was new as previous studies often reported the association of only a few specific pesticides.Any reports and responses or comments on the article can be found at the end of the article.

Background
Rhinitis is a common disease that affects general health, and quality of life.Approximately, 10% to 30% of the different populations worldwide suffer from this disease 1 .One study in Bangkok, Thailand, has reported a prevalence of chronic rhinitis to be approximately 13% 2 .In clinical terms, rhinitis refers to the inflammatory disease of the nasal mucosa, which can cause the following symptoms: nasal congestion, rhinorrhea, and sneezing 3 .Although rhinitis does not have a strict classification criterion, it can be classified into allergic rhinitis (AR), and nonallergic rhinitis (NAR).Both have the same nasal symptoms, with the difference that AR is triggered by allergens.Several factors can trigger NAR, including cold air, climate change, cooking smells, chemical odour, cigarette smoke, volatile organic chemicals, exercise, alcohol ingestion 4 , and cooking fumes 5 .NAR can be further classified by their pathological mechanisms into several subtypes, including occupation rhinitis, hormonal rhinitis, drug-induced rhinitis, food-induced rhinitis, emotion-induced rhinitis, etc 6 .Approximately 43% of all rhinitis cases are AR, and 23% are NAR, while 34% of the cases are a mixture of both 4 .
Studies found pesticide exposure to increase the risk of several respiratory problems, e.g., asthma, chronic bronchitis 7 , and rhinitis 8 .A study among grape farmers in Greece reported a higher prevalence of AR among those who used pesticides 9 .
Another study in France found that children living in areas surrounding vineyards had a higher rate of rhinitis symptoms (OR=3.56;95% CI 1.04-12.12) 10.In an occupational setting, a study found number of hours working in the greenhouse per day to be associated with rhinitis (OR, 1.85; 95% CI, 1.05-3.23) 11.A large survey of farmworkers in the United States of America (U.S.A) found that insecticide and herbicide use has significantly increased the risk of allergic rhinitis and asthma 12 .
A recent systematic review by Rodrigues et al. 13 found pesticides exposure to associate with only asthma in children and adolescents but not allergic rhinitis.In addition, so far there were only a few individual pesticides identified as potential risk factors for rhinitis.Pesticides that were found to have a positive association with rhinitis were bipyridyl herbicides such as paraquat, and the broad-spectrum herbicides 2,4-D, glyphosate, dithiocarbamate fungicides including benomyl, and insecticide diazinon 9,14,15 .It has been suggested that exposure to OP insecticides might exaggerated nasal glandular response resulting in increased rhinitis 16 .
Currently, evidence on the effects of pesticides on rhinitis is limited.Rhinitis might be caused by allergens or irritants and there is no logic that all pesticides will equally affect the disease.The main objective of this cross-sectional study was to determine the association between pesticide use and rhinitis among farmers in Phitsanulok, Thailand.The use of a large sample size in this study has provided the opportunity to assess the risks that different groups and subgroups of pesticide exposure have had on these individuals.The study results would be useful for disease prevention, and comparison with other studies.

Study design and setting
This study was a cross-sectional study.Participants were farmers in Phitsanulok province, located about 370 km north of Bangkok, Thailand.In 2019, the province had approximately 865,368 people (342,787 households) from nine districts.The major crops in the province are rice, sugarcane, and maize 17,18 .

Study participants and sampling procedure
Multistage sampling was used for the random selection of participants.The three districts were randomly selected from the nine districts of Phitsanulok province.From all the selected districts, 18 out of 26 (69.2%) sub-districts were further selected.In each sub-district, all local hospitals participated in the study and provided support for data collection.In each sub-district, farmers were selected by village health volunteers (VHV), who were working in the hospitals.In Thailand, VHVs are trained volunteers who provide primary healthcare service, including data survey, and they are familiar with villagers.By using the data from the local authority and personal contacts, the VHV selected the farmers and set up the interviewing plan.To collect the data, they visited the target household and conducted the interview.After finishing, they moved on to the next one, until the target number of participants was reached.By using this snowball sampling technique, the study collected data from 9,649 participants.In each family, only one adult aged 20 years or older who does agricultural work, was interviewed.The interview mostly took place in the participant's home.However, sometimes it was done at a local temple or hospital.In these situations, an interview was conducted in a privated circumstance.In the data collection protocol of the study, a group interview was not allowed.Data was collected from October 2020 to February 2021, by 210 VHV.Before data collection, these volunteers had to attend a one-day training program to be informed on the purpose of the study and to learn how to properly interview and collect data by using an online questionnaire.

Amendments from Version 2
In the Introduction, a statement on rational of study was added to justify the study objective.
The Methods section is updated with more detailed information on participants selection and data collection.
The discussion section is updated with more information on study limitations and strengths.Potential bias due to multiple testing, and common method bias has been discussed.In addition, statements on study strength and importance of study rhinitis were also added.
The conclusion statement was revised to precisely reflect the actual results of the study.
There were no changes made in tables, title, author list and order.

Any further responses from the reviewers can be found at the end of the article
The minimum sample size was calculated to be 10,002, based on the following assumptions: significance level = 95%; power of detection = 80%; ratio of unexposed/exposed = 1; percent of unexposed with outcome = 10% 2 ; odds ratio = 1.2 9 .
Study questionnaire and data collection Data were collected by using an in-person interview questionnaire (provided as Extended data in English and Thai) 19 .Data on rhinitis was collected by using a modified form of SFAR (Score For Allergic Rhinitis) questionnaire which is recommended for population studies, where medical diagnosis and objective measurements were absent or difficult to obtain 20 .
The SFAR encompasses questions regarding the eight features of AR.Each of the items can be quantitatively scored and yield a maximum score of 16.AR refers to those with a score of 7 or more.Self-reported rhinitis was defined as the participant answering "yes" to the question: "during the last 12 months, have you ever had symptoms such as sneezing, or a runny, or blocked nose when you did not have a cold or the flu?".
For pesticide exposure, the data was collected by a questionnaire used in our previous study 21 .Data on the long-term use of pesticides, either by types of pesticides (insecticide, herbicide, fungicide, rodenticide, and molluscicide), or by specific individual pesticides, were collected.A list of 39 individual pesticides were chosen from those that were commonly used in Thailand and were reported to cause adverse health effects 15,21 .
Participants were asked whether they have ever used pesticides, defined as a mixture, or spray pesticides, in their lifetime.To help them recall better and distinguish the individual pesticide, either chemical name, common name, or trade name of individual pesticide were included in the question, and the interviewer had to read all of them to the participants when asking the question.Participants were also asked to provide data on the duration (days/year and total years) of pesticide use.This information was used to calculate total days, and quartiles of days using each pesticide in the farmers' lifetime.
To control a potential confounding effect of the use of other pesticides, correlation matrices of types and individual pesticides were developed.Those pesticides with highly correlation (Spearman coefficient ≥0.30) were identified and included the regression model.A list of potential confounding pesticides was presented in Table S1 and Table S2.The dose-response relationship was analysed using chi-squared tests for trend using an ordinal term for the quartile days as a category.This statistical analysis was available only when there was sufficient frequency of exposure and disease.Data analysis was performed using IBM SPSS version 26, and OpenEpi (online version 3.01).All statistical values were two-tailed, and a p-value <0.05, was considered as statistically significant.

Ethical considerations
The study was approved by the Ethical Committee of Naresuan University (COA No. 657/2019) and written informed consent from the participants was obtained before the interview process.

Results
It was found that the proportion of female participants was slightly higher than that of the male participants (Table 1 and the underlying data 22 ).Most of these individuals were aged 40 years and older with an average age of 55 (±12 years), married (78.0%), finished primary school or with lower education (77.2), and had an average family income of 10,000 THB or less.A total of 8.7% of these females are cigarette smokers, and 13.7% consume alcohol.
The months with the highest frequency of symptoms were March to July (summer season in Thailand), and November to February (winter season).The prevalence of rhinitis was significantly associated with marital status, education, family income, cigarette smoking, and alcohol consumption (Table 3).
All five types of pesticides included in the study were found to be significantly related to rhinitis prevalence.Fungicides were significant until adjustments for using other pesticides were made, then it was shown to be not significant (Table 4).
b Model2: Adjusting for all the factors in the model1 and using of other pesticides.
significant after the adjustment for using other potential confounding pesticides.The association of some pesticides were in a dose-response pattern (Table 6).

Discussion
The study found an association between the history of pesticide exposure and the prevalence of rhinitis.However, the results must be interpreted with caution due to serveral methodological weaknesses.The association was consistent across various types of pesticides, including insecticides, herbicides, fungicides, rodenticides, and molluscicide (Table 4).For individual pesticides, 32 out of 39 showed a significant relation with the disease and some with a dose-response pattern (Table 5, Table 6).The relationship exists after adjustment by potential confounding factors including the use of other pesticides.This finding was new as previous studies often reported the association of only a few specific pesticides.
The difference might be explained by the fact that most of the rhinitis cases were NAR and the exposure was chronic.The study results also implied that the rhinitic effect might be related to the general characteristics of pesticide types, rather than individual properties.Pesticides can affect AR and NAR by several biomechanisms, involving immune and nonimmune pathways, after acute and chronic exposure.For acute high dose exposure, pesticides and the solvent composition of the mixture that contains them can directly irritate the nose and throat 23 .Organophosphate and carbamate can cause cholinergic stimulation of the nasal mucosa and inhibit acetylcholinesterase thus causing more secretion of mucus in the nose and airway system 24 .Organochlorines and pyrethroids are another type of insecticides designed to target the nervous system of insects.Exposure to the compounds can cause hyperexcitability of nerve cells by interacting with the sodium channel and keeping it open longer 25 .Pyrethroid like other insecticides can irritate the respiratory tract, nose, and throat 26 .
For herbicides, evidence from both animal and human studies showed that it can cause airway inflammation 27 .An experimental study found that acute exposure to herbicide 2,4-D increases the mast cells in the nasal epithelium of mice 28 .
Paraquat was another herbicide with high toxicity to the respiratory system, and previous studies reported a positive association of it and several respiratory illnesses, including rhinitis 9 and wheezing 29 .In an experimental study, it was found that fungicides have cytotoxic effects on bronchial epithelial cells 30 .
For chronic exposure, pesticide exposure can lead to exaggerated responses that increase the risk of rhinitis 16 .This continual response had been involved in the development of other chronic diseases, e.g., asthma, and bronchitis 31 .In a respiratory health study, OPs affect the bronchial lining and increase susceptibility to allergens or other stimuli 32 .OP can also induce airway hyperreactivity at doses lower than those required to cause acetylcholinesterase inhibition 33,34 .
Although it is hard to differentiate between AR and NAR, by using the SFAR questionnaire and its scoring method 20 , the study found that most of rhinitis cases in this study (64.2%) were NAR (Table 2).Further analysis by comparing the association of pesticides and types of rhinitis revealed that NAR had a stronger association (Table S3).This piece of information was often missed in the literature as most previous studies on rhinitis did not have enough data or focus on allergic types.The information helps explain why the results OR found in this study were higher than those previously reported.
This study has some limitations.The study did not have information and control for other potential confounding pollutants, such as grain and hay handling, and maintenance activities, e.g., repairing engines and pesticide equipment 15 .It was also very likely that participants were exposed to environmental pesticides.However, it is more likely that both the control and study groups will have a similar risk to those factors and the problems would bias toward the null.The information on pesticide exposure was dependent solely on the questionnaire method.However, as the study is focusing on long-term exposure, the questionnaire method was the best option 35 .This type of study could also be subject to recall bias as the case groups might be more aware of pesticide use than the control.However, the problem was less likely to occur because the information on the rhinitic effect of pesticides was not yet available in Thailand.In addition, the bias would only minimize but not increase the association 36 .Also, it was necessary to note that the questionnaire was first developed in France which has a different cultural background to Thailand, and this might affect the outcome.However, as most of the questions are of the "yes" or "no" type and ask about common symptoms, such as sneezing, runny, and blocked nose, application in a different cultural setting should not pose a serious problem.By using a cross-sectional design, the causeeffect association cannot be directly determined.In addition, the study collected data on pesticides used and rhinitis by using the same questionnaire method.This might cause common method bias which will affect the results and therefore the validity of the associations found.However, the bias was minimized to some extent by using a face-to-face interview instead of a self-administer questionnaire.The survey questionnaire was also designed so that exposure to information was collected before some of the health effects and rhinitis was noted.In addition, as the information is rather new, the workers were unlikely to be aware of the relationship between pesticide exposure and rhinitis.Lastly, as many statistical tests were performed to explore the association between various pesticides and rhinitis, the results were subjected to bias from multiple testing.The study results should be interpreted with caution.
The study is particularly strong in its design using a large sample size and many of the participants experienced using several pesticides in their lifetime.The survey also collected data regarding several individual chemicals from various types of pesticides.This posts a unique opportunity to study the effects of pesticides on rhinitis.The research is particularly important as rhinitis is an important public health issue and its prevalence is rising 1 .The disease not only affects quality of life and causes economic burden but also connected to other serious health problems, such as asthma and chronic bronchitis 7 .
Understanding the effects of pesticides is crucial as the chemicals are widely used and almost everyone is at risk of exposure.The information is crucial for the development of effective prevention and control measurements of rhinitis.

Conclusion
The results from this large cross-sectional study supports existing literature on the potential effects of chronic exposure to pesticides on rhinitis.The effects of pesticides on rhinitis should receive more public attention, and the information be incorporated into the disease prevention program.
This project contains the following underlying data: • Dataset pesticide and rhinitis-Thailand (SAV and CSV).
(Study questionnaire in English.) • Questionnaire-pesticide and rhinitis-Thai Is an Interesting study large based on a cross-sectional study aimed to determine the association between pesticide use and rhinitis prevalence among farmers.
Data were obtained form data bases of pesticide use and correlated with rhinitis reported cases by interview questionnaire.Data from 9,649 participants were included in the analysis.The association between pesticide exposure and rhinitis was determined by multiple variable logistic regression, adjusted for potential confounding factors.
Results are interesting, as well as the associations founded is a Local study that can be applied for other latitudes.

Is the work clearly and accurately presented and does it cite the current literature? Yes
Is the study design appropriate and is the work technically sound?Yes

If applicable, is the statistical analysis and its interpretation appropriate? Yes
Are all the source data underlying the results available to ensure full reproducibility?Yes

Are the conclusions drawn adequately supported by the results? Yes
Competing Interests: No competing interests were disclosed.
Reviewer Expertise: Asthma, COPD, Air pollution and health effects, Organophosphates and Asthma, Respiratory Toxicology I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard.The manuscript is well written and easy to understand.However, there are several methodological challenges in the study: There are no clear hypotheses in the study.Pesticides consist of many substances, and they are from different chemical classes.There is no logic in all of these to be causing rhinitis.
Rhinitis might be caused by allergens or irritants, and the authors should have commented on this in the introduction and tried to make specific hypotheses related to specific pesticides.When this is not done, it should be added as a discussion point in the Discussion.

○
The selection of workers is not totally clear.How were the farmers addressed?How did the researchers meet them, and how was the circumstances when the interviews were conducted?Did they answer in a group of people, or in rooms where no one heard the answer.At work or at home?This should be described and the influence of the situation should be discussed in the Discussion.
The snowball sampling must be described in detail, how did this happen?

○
The sample calculations described must have references to where the figures came from.

○
The interview was based on a questionnaire that originally was French, and the different cultures in France and Thailand should be mentioned as a challenge in the Discussion.

○
Workers seldom know a lot about pesticides they used.How could the workers know exactly which pesticides they had been exposed to?This part of the interview should have been described better.

○
In the Discussion, the authors must add text on discussion of 2 important weaknesses: 1. Common method bias: This study collected data on the outcome rhinitis at the same time as exposure information (if understood correctly, it is not totally clear described).This might cause a bias, and make workers report what they think is the situation, more than what is true.At least this should be clearly addressed in the discussion.
When all these methodological weaknesses exist in a study, the authors need to revise the conclusions.The results are far from clear due to these weaknesses.
At the start of the discussion, the text is: The study found a strong association between the history of pesticide exposure and the prevalence of rhinitis.
This must be changed to: The study shows an association between the history of pesticide exposure and rhinitis.However, the results must be interpreted with caution due to several methodological weaknesses.
In conclusion: -remove the word ' strongly' in the first sentence.The next sentence should be deleted: The study was the first to confirm that most individual pesticides from various groups are associated with rhinitis, especially the NAR type.
These details are not likely to be true.

○
The discussion should add more about why it is important to study rhinitis.This symptom/disease might be the start of a more serious health problem, asthma for instance, and by observing the existence of rhinitis, one can prevent more serious health problems.This is the most important message from this study and must be highlighted and not forgotten.Reviewer Expertise: Occupational health and neurotoxicology I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have significant reservations, as outlined above.
Author Response 01 Dec 2023

Comment:
This is an interesting study about rhinitis and pesticide exposure.It is a cross-sectional study among 9 649 Thailand farmers, interviewed about rhinitis and pesticide exposure.Regression analyses show associations between the exposure to different pesticides and the rhinitis.
The manuscript is well written and easy to understand.However, there are several methodological challenges in the study: There are no clear hypotheses in the study.Pesticides consist of many substances, and they are from different chemical classes.There is no logic in all of these to be causing rhinitis.Rhinitis might be caused by allergens or irritants, and the authors should have commented on this in the introduction and tried to make specific hypotheses related to specific pesticides.When this is not done, it should be added as a discussion point in the Discussion.

Response:
We appreciate your thoughtful comments and suggestions.They are very useful, and we hope we do enough to address the issues and to improve the paper's quality.
Yes, I agree that rhinitis might be caused by any allergens or irritants, and there is no logic to believe that all pesticides will equally affect rhinitis.As you know, with limited literature, there was not enough data to hypothesize on the association between specific pesticide and rhinitis.Therefore, our main objective was to affirm the results of previous studies and to explore some more specific potential predictive pesticides.Another point was that the effects of allergens and irritants as potential confounders has been further discussed, and more information has been added to the Introduction as suggested.

Comment:
The selection of workers is not totally clear.How were the farmers addressed?How did the researchers meet them, and how were the circumstances when the interviews were conducted?Did they answer in a group of people, or in rooms where no one heard the answer.At work or at home?This should be described, and the influence of the situation should be discussed in the Discussion.
The snowball sampling must be described in detail, how did this happen?

Response:
In this study, we use village health volunteers (VHV) to collect the data.The VHVs work as public health officers and they normally know every household in their community.The process begins with the VHV contacting the target household and visiting the household to conduct the interview.After finishing, they moved on to the next house and repeated this process until the target number was reached.We did not allow group interviews so there would not be a problem where interviewees influence each other.Sorry for not being aware of the issue and not clearly presenting the information.More information on sampling has been added in the Methods section.Since a large number of villagers participated in the study, so there were no concerns about the representation of the group.

Comment:
The sample calculations described must have references to where the figures came from.

Response:
The relevant references have been added.

Comment:
The interview was based on a questionnaire that originally was French, and the different cultures in France and Thailand should be mentioned as a challenge in the Discussion.

Response:
Although, the original was in French, most of the questions are simple yes or no questions, and asking a common symptom such as, sneezing, runny, and blocked nose.The statement has been added to the discussion as suggested.

Comment:
Workers seldom know a lot about pesticides they use.How could the workers know exactly which pesticides they had been exposed to?This part of the interview should have been described better.

Response:
Thank you for mentioning the issue.That was exactly what happened and we tried to minimize the problem by including all the names of each specific pesticide, especially commercial names in the question and read all of them to the workers when interviewed.We expected that this action could minimize recall bias to some extent.The information has been added to the Methods.

Comment:
In the Discussion, the authors must add text on discussion of 2 important weaknesses: 1. Common method bias: This study collected data on the outcome of rhinitis at the same time as exposure to the information (if understood correctly, it is not totally clear as described).This might cause a bias, and make workers report what they think is the situation, more than what is true.

Response:
Thank you for the insightful suggestions and the two biases have been acknowledged in the study limitations.
In this study, data on pesticides used in the past and the data on rhinitis were collected at the same visit and using the same questionnaire method for self-report data collection.This could cause common method bias and affect the validity of the association of the two variables.To minimize the bias, first we use a face-to-face interview instead of a selfadminister questionnaire.The survey questionnaire was designed so that exposure information was collected before some health effects and rhinitis.In addition, as the information is rather new, the workers were unlikely to be aware of the relationship between pesticide exposure and rhinitis.

Comment:
2. Multiple testing.Very many statistical tests have been performed, and this makes the likelihood of having significant results high.Testing should have been done with corrections, adjusting p-values derived from these tests, to avoid false positive results.
At least this should be clearly addressed in the discussion.

Response:
Yes, we agree that the results might subject to the problem of multiple testing as several individual pesticides were tested against the disease.To acknowledge it, a statement has been added to the study limitations.

Comment:
When all these methodological weaknesses exist in a study, the authors need to revise the conclusions.The results are far from clear due to these weaknesses.
At the start of the discussion, the text is: The study found a strong association between the history of pesticide exposure and the prevalence of rhinitis.
This must be changed to: The study shows an association between the history of pesticide exposure and rhinitis.However, the results must be interpreted with caution due to several methodological weaknesses.
In conclusion: -remove the word 'strongly' in the first sentence.The next sentence should be deleted: The study was the first to confirm that most individual pesticides from various groups are associated with rhinitis, especially the NAR type.
These details are not likely to be true.

Response:
Thank you very much for your kind suggestions.All of them are accepted and the statements have been revised as suggested.

Comment:
The discussion should add more about why it is important to study rhinitis.This symptom/disease might be the start of a more serious health problem such as asthma, and by observing the existence of rhinitis, one can prevent more serious health problems.This is (DOCX).(Study questionnaire in Thai.)Data are available under the terms of the Creative Commons Zero "No right reserved" data waiver (CC0 1.0 Public domain dedication).

Reviewer Report 13
October 2023 https://doi.org/10.5256/f1000research.148334.r212014© 2023 Moen B. This is an open access peer review report distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.Bente Elisabeth Moen University of Bergen, Bergen, NorwayThis is an interesting study about rhinitis and pesticide exposure.It is a cross-sectional study among 9 649 Thailand farmers, interviewed about rhinitis and pesticide exposure.Regression analyses show associations between the exposure to different pesticides and the rhinitis.

○ 2 .
Multiple testing.Very many statistical tests have been performed, and this makes the likelihood of having significant results high.Testing should have been done with corrections, adjusting p-values derived from these tests, to avoid false positive results.

○
Is the work clearly and accurately presented and does it cite the current literature?YesIs the study design appropriate and is the work technically sound?PartlyAre sufficient details of methods and analysis provided to allow replication by others?PartlyIf applicable, is the statistical analysis and its interpretation appropriate?PartlyAre all the source data underlying the results available to ensure full reproducibility?PartlyAre the conclusions drawn adequately supported by the results?PartlyCompeting Interests: No competing interests were disclosed.

Table 2 . Prevalence of rhinitis and its characteristics.
* Allergic rhinitis refers to those with SFAR core ≥7 CI 1.09-2.94).The lowest OR value was 1.67 (95% CI 1.41-1.99)for fungicide and 7.19 (95% CI 4.67-11.06)for insecticide.Many of the associations were in a dose-response pattern.The association remained significant after adjusting for use of other types of pesticides.

Table 3 . Association between demographic data and rhinitis prevalence. Not rhinitis Rhinitis P-value a
a Chi-square test.*Statistically significant difference with p value <0.05
a b Model2: Adjusted for all factors in the model1 and using other types of pesticide.c Significant OR were indicated in bold numbers.