Notes
Slide Show
Outline
1
National Health Interview Survey (NHIS)
Mortality & Disability
among
US Farmers & Pesticide Applicators
  • Gomez-Marin G, Zheng D, LeBlanc W, Lee D, Fangchao Ma , Lee D, Fleming LE,
  • Ma F, Pitman T, Jane D
  • Dept of Epidemiology & Public Health
  • University of Miami School of Medicine
  • Miami, FL
2
Outline
  • Background
  • Objectives
  • Methods
  • Results
  • Conclusions
  • Recommendations
3
 
4
Background: Pesticides & Health
  • Pesticides & Human Health
    • Acute Disease
    • Chronic Disease??
      • Disability?
      • Mortality?
  • Occupational Exposed Groups
    • Manufacturers
    • Farmers
    • Pesticide Applicators
5
 
6
Background: 
Morbidity/Mortality Risks
  • Farmers & Pesticide Applicators
  • ê Tobacco & Ethanol Diseases
    • Cardiovascular disease
    • COPD
  • é Accidents
  • é Neurologic Diseases
  • é Respiratory & Infectious Diseases


7
 
8
Background: é Cancer Risk
  • Farmers
  • Skin
  • Stomach
  • Brain
  • Prostate
  • Multiple Myeloma
  • Leukemia
  • Pesticide Applicators
  • Skin
  • Stomach
  • Brain
  • Testicular
  • Soft Tissue Sarcoma
  • Non Hodgkins Lymphoma



9
Background: Methodologic Issues
  • Farmers & Pesticide Applicators
  • Bias
    • Healthy Worker Effect
  • Population
    • Small sample sizes
    • Variable populations by geography
    • Inadequate follow up
  • Exposure
    • Inadequate exposure assessment
    • Variable exposures
10
Background
  • National Health Interview Survey (NHIS)
  • Annual national estimates on disease, injury, impairment, disability, and related issues on a uniform basis for the entire US population
11
Background
  • National Health Interview Survey (NHIS)
  • Multi-purpose household survey
  • Probability sampling
  • US civilian non-institutionalized population
  • Conducted Annually since 1957
  • 95-98% Response rate
12
Background
  • National Health Interview Survey (NHIS)
  • Periodic Supplements on subpopulations
  • Aggregate Morbidity Data
    • Annual Cross-Sectional
  • Mortality Follow Up through 1997
    • 1986-1994
    • Longitudinal Mortality Data

13
 
14
Objectives
  • 1) NHIS Cross Sectional:  Acute & Chronic Disability & Health Assessment
  • 2) NHIS Mortality Follow Up: Overall & Cause-Specific Mortality
    • Farmers & Pesticide Applicators vs Other US Workers
    • Farmers vs Pesticide Applicators

15
 
16
Methods: National Health Interview Survey (NHIS)
  • Probability sampling US Population
    • Self Reported Data
    • Individual Identifier & Weights
  • Household Survey
    • Proxy information
  • Annual Sub Population Supplements
    • Longest Held Job (1986)
    • Occupational Health (1988)
    • Smoking & other Confounders
17
National Health Interview Survey (NHIS)
  • 1986-1994: > 450,000 US workers > 18 yrs
  • Employment status during 2 weeks prior to interview
  • Standardized Occupational & Industry Codes
      •  All, 46 & 16 SOC & SIC re-Codings
      • Farmers and Pesticide Applicators

18
National Health Interview Survey (NHIS)
  • Demographic Variables
    • Age
    • Gender
    • Race
    • Ethnicity
    • Socio-Economic
      • Education
      • Family Income
    • Geographic Region
      • Urban/Rural
19
National Health Interview Survey (NHIS)
  • Morbidity & Mortality (ICD 9)
    • Acute & Chronic Conditions
    •  6 Lists of Chronic Conditions
    •  Various Recodes
    •  Injury
      • Job related
  • Self Assessed Health
    • Rating
    • Reporting Health Condition

20
National Health Interview Survey (NHIS)
  • Acute Disability (<2 Weeks)
    • Days Restricted Activity
    • Days Bed
    • Days Lost Work
  • Chronic Disability (<12 Months)
    • Doctor Visits
    • Hospitalizations
21
National Health Interview Survey (NHIS)
  • Mortality Follow Up
    • National Death Index (NDI) linkage
    • January 1, 1986 - December 31, 1997
    • Computerized Linkage algorithm
      • Match class & score
      • 97% complete
      • > 18 yrs
22
 
23
 
24
 
25
Example of Hypothesis
  • Pesticide Applicators & Farmers:
  • é Prostate Cancer Risk
    •   Detect Risk Ratio = 1.54
      • Alpha level of 0.05 with 90% Power
26
Statistical Methods
  • Create Database
    • Exclude: Unemployed, <18 yrs, inadequate data
    • Create Pooled Dataset
    • Link Annual Survey & Mortality Follow up (1986-97)
  • Multi-stage sampling design
    • Adjustments for sample weights & design effects
    • SUDAAN Statistical Package
    • Further adjustment of sample weights for analysis of data from combined survey years (Botman 1995)
  • Biases
    • Proxy
    • Interaction
  • Morbidity/Mortality analyses
    • Cox Regression models (SUDAAN Proc Survival)
27
 
28
Results
  • Farmers
  •   N 9576
  •   Deaths 503
  •   Women 1772 (18.5%)
  •   Mean Age 44.1 ± 16.2* ± SD
  • Pest All Other Work
  • 180 453,219
  • 9 8078
  • 7 (3.9%) 214,063 (47.2%) 38.6 ± 12.7* 38.8 ±12.8*



  • * p<0.001
29
Results
  • Farmers
  •   Hispanic 1175 (12.3%)
  •   Black 376 (3.9%)
  •   >12 Grade 2322 (24.5%)
  • Pest All Other Work
  • 21 (11.7%) 33,066 (7.3%)*
  • 27 (15.0%) 56,157 (12.4%)*
  • 59 (32.9%) 211,384 (47.1%)*



  • * p<0.001


30
Results: Acute Disability
  • Farmer Pest


  • Restricted 0.78 (0.70-0.87) 0.83 (0.45-1.53)
  • Bed 0.71 (0.61-0.82) 0.81 (0.35-1.85)
  • Lost Work 0.73 (0.65-0.83) 0.90 (0.45-1.80)



    • Model = Occupation, Age, Gender, Race-Ethnicity, Education (compared to All Other Workers)
31
Results: Chronic Disability
  • Farmer Pest


  • MD Visit 0.82 (0.76-0.88) 1.21 (0.80-1.83)
  • Hospital 1.04 (0.92-1.16) 0.88 (0.42-1.86)




    • Model = Occupation, Age, Gender, Race-Ethnicity, Education (compared to All Other Workers)
32
Results: Health Assessment
  • Farmer Pest


  • Self   1.08 (0.96-1.23) 1.35 (0.80-2.30)
  • Condition 0.92 (0.86-0.98) 1.34 (0.97-1.86)



      • Model = Occupation, Age, Gender, Race-Ethnicity, Education (compared to All Other Workers)

33
Results: High Risk Reported Disability & Health
  • Age Sex Race/Eth Educ


  • Acute Y F WNH Low
  • Chronic O F WNH Low
  • Health O F H/B Low


  • OVERALL: Non Hispanic Older Woman Worker of Lower Education
34
Results:
Overall & Cause-specific Number of Deaths by Occupation & Gender
  • Pesticide Exposed Other
  • M F M F
  • All Deaths 512 59 8078 3914
  • Infectious 9 0 454 116
  • Heart 211 20 2729 1084
  • Respiratory 16 2 394 198
  • Nervous 5 1 83 46
  • MVA/Accidents 37 7 630 223
35
Results:
Cancer-specific Number of Deaths by Occupation & Gender
  • Pesticide Other
  • M F M F
  • Overall cancer 161 23 2515 1699
  • Nervous system 9 2 86 55
  • Digestive 37 4 682 333
  • Respiratory 45 1 884 391
  • Breast 0 2 7 382
  • Prostate 22 -- 177 --
  • Genital 22 1 182 183
  • Urinary 5 1 112 35
  • Lymph/Hemat 20 7 226 150
36
General Mortality Results
Pesticide Exposed vs Other
  • Age Adjusted RR  (95% CI)
  • Disease Male Female Total
  • Overall Mortality 1.2 (1.1-1.3) 1.1 (0.8-1.4) 1.3 (1.2-1.5)
  • Infectious 0.5 (0.3-1.0) 0.6 (0.3-1.2)
  • Heart 1.2 (1.0-1.4) 1.1 (0.7-1.8) 1.4 (1.2-1.7)
  • Respiratory 0.6 (0.4-1.0) 0.5 (0.1-2.1) 0.7 (0.4-1.1)
  • Nervous System 1.4 (0.6-3.3) 1.1 (0.4-2.9) 1.7 (0.2-13.3)
  • MVA/Accidents 1.6 (1.2-2.2) 3.2 (1.5-6.9) 2.1 (1.6-2.8)



37
Cancer Mortality Results:
Pesticide Exposed vs Other
  • Age Adjusted RR  (95% CI)
  • Disease Male Female Total
  • Overall Cancer 1.2 (1.0-1.4) 1.0 (0.7-1.5) 1.2 (1.1-1.4)
  • Nervous 2.1 (1.0-4.5) 2.4 (0.6-8.8) 1.2 (1.1-1.4)
  • Digestive 0.9 (0.3-2.8) 0.7 (0.3-1.7) 1.0 (0.7-1.4)
  • Respiratory 1.0 (0.7-1.3) 0.2 (0.02-1.3) 1.0 (0.8-1.4)
  • Breast 0.4 (0.1-1.5)
  • Prostate 1.3 (0.8-2.2)
  • Genital 1.3 (0.8-2.2) 0.4 (0.1-2.7) 1.3 (0.8-2.1)
  • Urinary 1.2 (0.4-3.6) 1.6 (0.2-10.5) 1.6 (0.6-4.1)
  • Lymphatic/Hemat 1.7 (1.1-2.7) 3.6 (1.6-8.2) 2.2 (1.5-3.2)
38
 
39
NHIS Limitations
  • Self Report & Proxy
  • Access to Care (Disability)
  • Self Employment (Disability)
  • Lack of individual exposure measures
    • Confounding
    • Pesticide exposure
  • Relatively small numbers of deaths


  • Probability Sample of US Worker Population as the major comparison population is appropriate for issues of the healthy worker effect and other biases
40
Conclusions:
Reported Disability Risk
  • NHIS population-based sample US workers
  • Pesticide Exposed vs Other Workers
    • é Aging Farmers
    • ê Acute Disability
    • ê Chronic Disability
    •  Poor Health Assessment
    • ê Conditions
  • é Non Hispanic Older Woman Worker of Lower Education
41
Conclusions: Mortality Risk
  • NHIS population-based sample US workers
  • Pesticide Exposed vs Other Workers
    • é Overall Mortality
    • é Cardiovascular disease
    • ê Tobacco-related diseases
    • ê Infectious diseases
    • é Nervous system diseases
    • é Motor vehicles & accidents
    • é Overall Cancer
      • é Nervous System Cancer
      • é Lymphocytic/Hematopoietic Cancer

42
 
43
Future Work
  • National Health Interview Survey (NHIS): Research Potential
  • Probability Sample ALL US Workers
    • Supplemental Surveys
      • Confounding information
    • Prevalence of Acute & Chronic Morbidity
    • Cross Sectional Trends over Time
    • Mortality Follow Up
      • Longitudinal Cohort Study
      • PMR, SMR
  • Study Website
    • http://www.rsmas.miami.edu/groups/niehs/niosh/
44
Acknowledgments
  • Data
    • Inter-University Consortium for Political & Social Research
    • US Dept of Health & Human Services
    • National Center for Health Statistics
  • Funding
    • National Institute of Occupational Safety & Health (NIOSH)-funded R01 0H03915-01
    • National Institute of Occupational Safety & Health (NIOSH)-funded Deep South Agricultural Center (University of South Florida)
45