Prevalence of people reporting sensitivities to chemicals in a population-based survey

Am J Epidemiol. 1999 Jul 1;150(1):1-12. doi: 10.1093/oxfordjournals.aje.a009908.


To describe the prevalence and correlates of reports about sensitivities to chemicals, questions about chemical sensitivities were added to the 1995 California Behavior Risk Factor Survey (BRFS). The survey was administered by telephone to 4,046 subjects. Of all respondents, 253 (6.3%) reported doctor-diagnosed "environmental illness" or "multiple chemical sensitivity" (MCS) and 643 (15.9%) reported being "allergic or unusually sensitive to everyday chemicals." Sensitivity to more than one type of chemical was described by 11.9% of the total sample population. Logistic regression models were constructed. Hispanic ethnicity was associated with physician-diagnosed MCS (adjusted odds ratio (OR) = 1.82, 95% confidence interval (CI) 1.21-2.73). Female gender was associated with individual self-reports of sensitivity (adjusted OR = 1.63, 95% CI 1.23-2.17). Marital status, employment, education, geographic location, and income were not predictive of reported chemical sensitivities or reported doctor diagnosis. Surprising numbers of people believed they were sensitive to chemicals and made sick by common chemical exposures. The homogeneity of responses across race-ethnicity, geography, education, and marital status is compatible with a physiologic response or with widespread societal apprehensions in regard to chemical exposure.

MeSH terms

  • Adolescent
  • Adult
  • Age of Onset
  • California / epidemiology
  • Community Health Planning
  • Female
  • Health Surveys
  • Humans
  • Incidence
  • Income / statistics & numerical data
  • Logistic Models
  • Male
  • Multiple Chemical Sensitivity / diagnosis
  • Multiple Chemical Sensitivity / epidemiology*
  • Multiple Chemical Sensitivity / etiology
  • Predictive Value of Tests
  • Prevalence
  • Racial Groups
  • Risk Factors
  • Sex Distribution
  • Surveys and Questionnaires