Evidence of recall bias in volunteered vs. prompted responses about occupational exposures

BACKGROUND: Recall bias remains a concern in case-control studies, although few investigations have found evidence of differential recall. This study examined whether differences in occupational exposure reporting occur in volunteered vs. prompted questionnaire responses. METHODS: In a large, population-based, case-control study of a childhood cancer, neuroblastoma, we calculated odds ratios for broad occupational exposure groups on the assumption that in the absence of recall bias, risk estimates for such broad groupings should be close to the null value. RESULTS: Prompted exposures and work activities showed little evidence of differential recall by parents of cases and controls (all OR < 1.2), but case parents were more likely to volunteer information about other exposures or activities (ORs: 1.35-1.71). Case mothers were also more likely than control mothers to report activities involving indirect exposure (OR = 1.41). CONCLUSIONS: These findings suggest that prompted exposure questions are less likely to be subject to recall bias than open-ended questions.     

Neuropsychiatric symptoms among welders exposed to neurotoxic metals.

Neuropsychiatric symptoms in 65 welders exposed to aluminium and 217 railroad track welders were studied with the aid of a previously validated questionnaire. Semiquantitative data on exposure to the metals aluminium, chromium, lead, manganese, and nickel were also recorded by questionnaire. Logistic regression was employed to study the relation between exposure and the prevalence of symptoms. Welders exposed to aluminium, lead, or manganese for a long period had significantly more neuropsychiatric symptoms than welders not exposed to these metals. The results indicate that detailed psychometric studies should be performed on welders exposed long term to specific metals as such exposures might affect their nervous system.     

Comparison of Open and Closed Questionnaire Formats in Obtaining Demographic Information From Canadian General Internists

The objective of this study was to compare the impact of closed- versus open-ended question formats on the completeness and accuracy of demographic data collected in a mailed survey questionnaire. We surveyed general internists in five Canadian provinces to determine their career satisfaction. We randomized respondents to receive versions of the questionnaire in which 16 demographic questions were presented in a closed-ended or open-ended format. Two questions required respondents to make a relatively simple computation (ensuring that three or four categories of response added to 100%). The response rate was 1007/1192 physicians (80.0%). The proportion of respondents with no missing data for all 16 questions was 44.7% for open-ended and 67.0% for closed-ended formats (P < 0.001). The odds of having missing items remained higher for open-ended response options after adjusting for a number of respondent characteristics (2.67, 95% confidence interval 2.01 to 3.55). For the two questions requiring computations focused on professional activity and income, there were more missing data (P = 0.02, 0.02, respectively) but fewer inaccurate responses (P = 0.009, 0.20, respectively) for the open-ended compared to the closed-ended format. Investigators can achieve higher response rates for demographic items using closed format response options, but at the risk of increasing inaccuracy in response to questions requiring computation.     

The everyday concerns of mothers of young children and the motivation to seek ongoing parenting support from experts

Despite a significant body of work that documents the importance of mother–child relationships, the studies on programmes that support these relationships are few, particularly with respect to subclinical populations. In this report, we coded the responses of 114 mothers of infants and toddlers to open-ended questions in order to identify the nature and frequency of the concerns that prompted them to participate in mother–child groups. The responses were coded by two independent raters who achieved a strong inter-rater reliability on the 21-item coding system that was developed. Mothers who reported concerns about their confidence/ambivalence about being a mother and those who indicated a desire for support and guidance from others were significantly more likely to continue to seek ongoing assistance with parenting and enrol for more semesters than mothers who did not express such concerns at intake.     

Lead poisoning in precious metals refinery assayers: A failure to notify workers at high risk

Lead poisoning in a precious metals refinery fire assayer and a routine OSHA inspection prompted an investigation of the index facility, a survey of the industry, and efforts to notify assayers of this previously unrecognized hazard. Air and blood samples were obtained at the index facility. Management personnel from all fire assay laboratories in Rhode Island and southern Massachusetts were interviewed. The industry's trade association, OSHA, NIOSH, trade unions, and the media were asked to assist in a nationwide notification effort. Assayers at the index facility had excessive exposures to lead due to an age-old, lead-based assaying method that remains the industry gold standard. Blood lead levels of the three assayers (mean 61.3 μ/dl, range 48–86 μg/dl) were considerably higher than those of 16 other refinery workers (mean 27.4 μg/dl, range 13–49 μg/dl). The industry survey revealed inadequate knowledge of both the lead hazard and the applicability of the OSHA lead standard. Notification efforts failed in large part due to economic obstacles. The notification of workers at high risk of lead exposure and the eradication of occupational lead poisoning will require greater attention to economic forces. © 1994 Wiley-Liss, Inc.     

Assessment of occupational exposure in a population based case-control study: comparing postal questionnaires with personal interviews.

BACKGROUND: In case-control studies, data collection on occupational exposures by means of personal interviews is usually costly and time consuming. As detailed semiquantitative information on exposure from these interviews often has to be dichotomised in the analyses due to the small numbers of exposed subjects, the question is raised whether simple postal questionnaires yield the same results for occupational exposure in epidemiological studies as job specific personal interviews. METHODS: Data on occupational exposures during pregnancy were compared from 121 women who both completed a checklist with 17 occupational exposure categories in a postal questionnaire and were personally interviewed with specific questions on exposure with details of job and task. kappa Coefficients were calculated as measures of agreement corrected for chance, and sensitivity and positive predictive values as measures of validity and usefulness, with the exposure assessment based on information from the interview as the gold standard. RESULTS: Values of kappa varied from 0.09 for domestic cleaning agents to 0.70 for pesticides, indicating only low to moderate agreement between the questionnaire and the interview. Sensitivity ranged from 38% to 100%, with the highest values for agents used by healthcare workers. Positive predictive values were lower, between 9% and 63%, which indicates that overreporting was more common than underreporting in the questionnaire. CONCLUSIONS: These results underline the high potential for misclassification of occupational exposure in studies based on questionnaires. Therefore, postal questionnaires are not considered an alternative to job and task specific personal interviews in epidemiological studies.     

Current chemical exposures among Ontario construction workers

Current occupational exposures to chemical agents were assessed as part of an epidemiological study pertaining to the cancer and mortality patterns of Ontario construction workers. The task-based exposure assessment involved members from nine construction trade unions. Air samples were taken using personal sampling pumps and collection media. A DustTrak direct-reading particulate monitor was also employed. Exposure assessments included measurements of airborne respirable, inhalable, total, and silica dust; solvents; metals; asbestos; diesel exhaust and man-made mineral fibers (MMMF). In total, 396 single- or multi-component (filter/tube), 798 direct-reading, and 71 bulk samples were collected. The results showed that Ontario construction workers are exposed to potentially hazardous levels of chemical agents. The findings are similar to those reported by other researchers, except for silica exposure. In our study, silica exposure is much lower than reported elsewhere. The difficulty associated with assessing construction workers' exposures is highlighted.     

Agricultural Work Safety Efforts by Wisconsin Extension Agricultural Agents

This study investigated the agricultural work-related safety and health programming of county-level cooperative extension agents who work through land grant universities to provide a range of educational programs to agricultural producers. A questionnaire was designed and administered to all 89 Wisconsin agriculture and agribusiness extension county faculty. The questionnaire obtained valid responses from 98.9 percent of the agents. Ninety percent of all agents conducted some occupational safety and health promotion programming in the last year. These activities occupied an average of 4.8 days per agent per year. Most of the reported activities were group programs for the agricultural labor force that involved other extension agents and included the use of videotapes. The greatest barrier to more programming was lack of time on the part of both the agricultural work force and the agents. Most extension agents placed greater emphasis on training in how to work safely around hazards than on how to recognize and permanently correct hazards. For future programs agents requested more short format materials to use in programming, such as fact sheets, videotapes, and farm hazard inspection checklists. Agents are important training delivery resources for controlling farm-related injury and disease. Agents could be more effective with more time, better materials, and with more emphasis on hazard correction in workplace safety programs.     

An industrial hygiene methodology developed in support of a retrospective morbidity case-control study.

An industrial hygiene methodology that was developed in support of an epidemiologic case-control study is described. This study was conducted to investigate a potentially increased incidence of colorectal cancer among employees who worked at a unit that manufactured polypropylene by a heavy diluent process. Retrospective epidemiologic case-control studies typically have exposure assessment problems because industrial hygiene monitoring data are often not available. Misleading job titles are another problem that can provide a poor framework for estimating exposures. In addition, a job title-based exposure assessment assumes individual work patterns are not important in assessing worker exposures, although this is contradictory to industrial hygiene experience. The design of the case-control study provided an innovative industrial hygiene approach to circumvent these typical exposure estimating problems. The industrial hygiene methodology of the case-control study included assessing historical exposures, developing an exposure estimating matrix, selecting candidate etiologic agents based on a joint toxicologic and industrial hygiene review, administering a work activity questionnaire, calculating exposure scores, and conducting a statistical analysis. The study design also provided an additional exposure measurement independent of the toxicologic and industrial hygiene review. This provided an opportunity to compare the likelihood of exposure misclassification and, as expected, showed that a more detailed exposure estimate resulted in less misclassification.     

Comparison of a computer-administered and paper-and-pencil-administered questionnaire on health and lifestyle behaviors

PurposeThe Health Behavior in School-aged Children (HBSC) study is a cross-national survey examining health and lifestyle behaviors of adolescents, traditionally administered by paper and pencil in a school setting. The purpose of the present study was to investigate if a computer format of the questionnaire would affect the responses of the adolescents in comparison with the paper and pencil format.MethodsIn 2000, the Belgian (Flemish) HBSC survey was administered in 81 secondary schools by paper and pencil (n = 4083) and in 19 schools by computer (n = 1008). Eight hundred four computer participants (aged 12–20 years) were matched with paper and pencil participants on the basis of gender, age, educational level, socioeconomic status, and type of school authority. The questionnaire included items on lifestyle behaviors (e.g., nutrition, substance use, physical inactivity, sexual behavior), psychosomatic health, and social relations.ResultsFor the majority of lifestyle behaviors, we found that mode of administration had no significant effect on adolescents’ responses, whereas for several questions about feelings/affective states, more socially desirable responses were found in the paper and pencil format than in the computer format.ConclusionThe present findings suggest a need for further exploration of potential mode effects on questions about feelings and affective states before transition from the paper and pencil HBSC-survey to the computer format.     

A job-exposure matrix for use in population based studies in England and Wales

The job-exposure matrix described has been developed for use in population based studies of occupational morbidity and mortality in England and Wales. The job axis of the matrix is based on the Registrar General's 1966 classification of occupations and 1968 classification of industries, and comprises 669 job categories. The exposure axis is made up of 49 chemical, physical, and biological agents, most of which are known or suspected causes of occupational disease. In the body of the matrix associations between jobs and exposures are graded to four levels. The matrix has been applied to data from a case-control study of lung cancer in which occupational histories were elicited by means of a postal questionnaire. Estimates of exposure to five known or suspected carcinogens (asbestos, chromates, cutting oils, formaldehyde, and inhaled polycyclic aromatic hydrocarbons were compared with those obtained by detailed review of individual occupational histories. When the matrix was used exposures were attributed to jobs more frequently than on the basis of individual histories. Lung cancer was significantly more common among subjects classed by the matrix as having potential exposure to chromates, but neither method of assigning exposures produced statistically significant associations with asbestos or polycyclic aromatic hydrocarbons. Possible explanations for the failure to show a clear effect of these known carcinogens are discussed. The greater accuracy of exposures inferred directly from individual histories was reflected in steeper dose response curves for asbestos, chromates, and polycyclic aromatic hydrocarbons. The improvement over results obtained with the matrix, however, was not great. For occupational data of the type examined in this study, direct exposure estimates offer little advantage over those provided at lower cost by a matrix.     

A job-exposure matrix for use in population based studies in England and Wales.

The job-exposure matrix described has been developed for use in population based studies of occupational morbidity and mortality in England and Wales. The job axis of the matrix is based on the Registrar General's 1966 classification of occupations and 1968 classification of industries, and comprises 669 job categories. The exposure axis is made up of 49 chemical, physical, and biological agents, most of which are known or suspected causes of occupational disease. In the body of the matrix associations between jobs and exposures are graded to four levels. The matrix has been applied to data from a case-control study of lung cancer in which occupational histories were elicited by means of a postal questionnaire. Estimates of exposure to five known or suspected carcinogens (asbestos, chromates, cutting oils, formaldehyde, and inhaled polycyclic aromatic hydrocarbons were compared with those obtained by detailed review of individual occupational histories. When the matrix was used exposures were attributed to jobs more frequently than on the basis of individual histories. Lung cancer was significantly more common among subjects classed by the matrix as having potential exposure to chromates, but neither method of assigning exposures produced statistically significant associations with asbestos or polycyclic aromatic hydrocarbons. Possible explanations for the failure to show a clear effect of these known carcinogens are discussed. The greater accuracy of exposures inferred directly from individual histories was reflected in steeper dose response curves for asbestos, chromates, and polycyclic aromatic hydrocarbons. The improvement over results obtained with the matrix, however, was not great. For occupational data of the type examined in this study, direct exposure estimates offer little advantage over those provided at lower cost by a matrix.     

Validation of a questionnaire and microenvironmental model for estimating past exposures to ozone

Epidemiology studies of the health effects associated with chronic exposure to ozone are constrained by a lack of validated exposure assessment methods that account for activity patterns and building ventilation, two potentially important predictors of ozone exposure. A pilot study was conducted to evaluate the impact of these parameters on an exposure model, to evaluate how reliably they can be reconstructed, and how this reliability affects estimates of past exposures to ozone. Fourteen subjects completed time-activity diaries and wore personal passive monitors during summer 1996 and reconstructed their activities 1 year later in a questionnaire. These data were used in a microenvironmental exposure model and the results were compared to measured personal exposure. A model using activity patterns, building ventilation, and location-specific ambient ozone data produced mean exposure estimates 18.9 +/- 1.5 ppb ozone greater than personal measurements. A simpler model that omitted variation in indoor ventilation produced exposure estimates 5.0 +/- 1.3 ppb ozone greater than mean measured exposures. The questionnaire was a reliable survey method for reconstructing past activity patterns. However, detailed histories of building ventilation may not be required to reconstruct reliably past exposures to ozone. The simple model was more precise and more practical for implementation in large epidemiological studies of health effects related to chronic exposure to ambient ozone.     

Exposure to mineral oil mist and respiratory symptoms in marine engineers

The purpose of the study was to see if marine engineers have an increased prevalence of respiratory symptoms, and if so whether it can be related to occupational exposures. A self-administered questionnaire was sent to 700 male seamen from three Norwegian ferry companies. Of the 492 respondents, 169 were currently working as marine engineers and 295 had never worked as marine engineers. The outcomes of cough and wheezing, chronic bronchitis, severe dyspnea, any dyspnea, and mucous membrane irritation (MMI) were defined from the questionnaire. Age and smoking-adjusted prevalences of these respiratory conditions were compared between the groups. Logistic regression was used to further elucidate the explanatory variables. The exposure assessment indicated an exposure (TWAC) to oil mist for marine engineers in the range from 0.12 to 0.74 mg/m³ (mean 0.45 mg/m³). When comparing current marine engineers with those who had never worked as marine engineers, the prevalence ratios were 1.38 (95% CI 1.0–1.9) for MMI, 1.53 (95% CI 1.2–1.9) for any dyspnea, and 1.63 (95% CI 1.0–2.6) for severe dyspnea. The differences remained for some of the symptoms after controlling for self-reported former asbestos exposure in the regression analysis. The increased prevalence of respiratory symptoms found among marine engineers in this investigation may partly be explained by oil-mist exposure, or more probably by a combination of past asbestos exposure and past and present oil-mist exposure. Am. J. Ind. Med. 32:84-89, 1997. © 1997 Wiley-Liss, Inc.     

Exposure assessment implications for the design and implementation of the National Children's Study

Examining the influence of environmental exposures on various health indices is a critical component of the planned National Children's Study (NCS). An ideal strategy for the exposure monitoring component of the NCS is to measure indoor and outdoor concentrations and personal exposures of children to a variety of pollutants, including ambient particulate and gaseous pollutants, biologic agents, persistent organics, nonpersistent organics (e.g., pesticides), inorganic chemicals (e.g., metals), and others. However, because of the large sample size of the study (approximately 100,000 children), it is not feasible to assess every possible exposure of each child. We envision that cost-effective strategies for gathering the necessary exposure-related information with minimum burden to participants, such as broad administration of product-use questionnaires and diaries, would likely be considered in designing the exposure component of the NCS. In general a biologic (e.g., blood, urine, hair, saliva) measure could be the dosimeter of choice for many of the persistent and for some of the nonpersistent organic pollutants. Biologic specimens, such as blood, can also indicate long-term internal dose to various metals, including lead and mercury. Environmental measures, on the other hand, provide pathway/source-specific exposure estimates to many of the environmental agents, including those where biologic measurements are not currently feasible (e.g., for particulate matter and for some gaseous criteria pollutants). However, these may be burdensome and costly to either collect or analyze and may not actually indicate the absorbed dose. Thus, an important technical and logistical challenge for the NCS is to develop an appropriate study design with adequate statistical power that will permit detection of exposure-related health effects, based on an optimum set of exposure measurement methods. We anticipate that low-cost, low-burden methods such as questionnaires and screening type assessments of environmental and biologic samples could be employed, when exposures at different critical life stages of vulnerability can be reliably estimated by these simpler methods. However, when reliability and statistical power considerations dictate the need for collecting more specific exposure information, more extensive environmental, biologic, and personal exposure measurements should be obtained from various "validation" subsets of the NCS population that include children who are in different life stages. This strategy of differential exposure measurement design may allow the exposure-response relationships to be tested on the whole cohort by incorporating the information on the relationship between different types of exposure measures (i.e., ranging from simple to more complex) derived from the detailed validation subsamples.     

Exposure to potential occupational asthmogens: Prevalence data from the National Occupational Exposure Survey

Few data are available about the prevalence of occupational exposures to agents which can cause occupational asthma or aggravate preexisting asthma (asthmogens). Using potential occupational exposure data from the National Occupational Exposure Survey (NOES) of 1980–1983, we investigated the number of asthmogen exposures, asthmogen-exposure(s) per production worker, and unprotected occupational asthmogen exposures in different industries and occupations. Data for the entire United States were used to generate estimates of occupational exposure at two selected state and local levels. It was estimated that 7,864,000 workers in the surveyed industries were potentially exposed to one or more occupational asthmogen(s) in the United States. The average number of observed potential exposures per asthmogen-exposed worker was 4.4, and varied from 11.9, in the Water Transportation industry, to 1.2 in Local and Suburban transportation. The largest number of observed potential exposures was recorded in the Apparel and Other Finished Products (garment) industry. This work and further analyses using this approach are expected to contribute to a better understanding of the epidemiology of occupational asthma, and to serve as a guide to target future occupational asthma surveillance efforts. Am. J. Ind. Med. 31:195–201 © 1997 Wiley-Liss, Inc.