xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 001925231
007 cr mnu|||uuuuu
008 080130s2007 flu sbm 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0002166
Williamson, Stephen E.
A new questionnaire to determine the frequency and severity of symptoms caused by inhaled odors, chemicals and irritants in normal subjects and their relation to health-related quality of life
h [electronic resource] /
by Stephen E. Williamson.
[Tampa, Fla.] :
b University of South Florida,
ABSTRACT: Individuals may develop symptoms in response to inhaled odors, chemicals, and irritants. This may affect their quality of life. Little is known about the prevalence and severity of symptoms that result from exposure to odors, chemicals and irritants. This study demonstrates the development of a new respiratory questionnaire to detect the prevalence and severity of symptoms experienced upon exposure to chemicals, odors, and irritants, and relates these symptoms to quality of life. This questionnaire was submitted to 96 volunteers at the University of South Florida College of Public Health who responded to items regarding symptoms developed in response to exposure to automobile exhaust, cigarette smoke, strong smells, cologne, perfumes or scented candles, or fresh paint vapors or fumes. Health-related quality of life was assessed using a subscale included with the questionnaire. The number and severity of symptoms developed in response to exposure to odors, chemicals, and irritants showed a strong negative correlation with health-related quality of life, consistent with intuitive estimates of the direction of this relationship. Also, it was shown that in normal populations, males and females develop statistically similar prevalence and severity of symptoms in response to exposure to odors, chemicals, and irritants.
Thesis (M.S.)--University of South Florida, 2007.
Includes bibliographical references.
Text (Electronic thesis) in PDF format.
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
Title from PDF of title page.
Document formatted into pages; contains 34 pages.
Advisor: Stuart M. Brooks, M.D.
x Public Health
t USF Electronic Theses and Dissertations.
A New Questionnaire to Determine the Frequenc y and Severity of Symptoms Caused by Inhaled Odors, Chemicals and Irritants in No rmal Subjects and Their Relation to HealthRelated Quality of Life by Stephen E. Williamson, M.D. A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science in Public Health Department of Environmenta l and Occupational Health College of Public Health University of South Florida Major Professor: Stuart M. Brooks, M.D. Thomas Truncale, D.O. Thomas E. Bernard, Ph.D. Date of Approval: July 16, 2007 Keywords: survey, exposed, cacosmia, sensitivity, self-administered Copyright 2007, Stephen E. Williamson, M.D.
Acknowledgements Grateful appreciation is expressed for the help given by the members of the Industrial/Organizational Psychol ogy Department at the Univ ersity of South Florida, including Paul Spector, Ph.D., Nicole Jagus ztyn, Ashley Nixon, and Kristen Shockley. Thank go to Dr. Tao Wang, Ph.D., of the Depart ment of Epidemiology and Biostatistics at the University of South Florida, for his guida nce on statistical proces ses. This study was supported by a NIOSH Grant for Educational Programs in Safety and Health, number T420H008438-01.
i Table of Contents List of Tables................................................................................................................. ..........ii List of Figures................................................................................................................ .........iii Abstract....................................................................................................................... ............iv Hypothesis and Specific Aims.................................................................................................1 Introduction................................................................................................................... ...........2 Methods........................................................................................................................ ............6 Results........................................................................................................................ ..............9 Discussion..................................................................................................................... .........17 The Final Questionnaire........................................................................................................ .19 List of References............................................................................................................. .....21 Appendices..................................................................................................................... ........24 Appendix A: Flyer............................................................................................................2 5 Appendix B: Cover Letter.................................................................................................26 Appendix C: Medical and Demographic Information......................................................27 Appendix D: Symptom and Qua lity of Life Questionnaire..............................................28 Appendix E: Proposed Revised Medical and Demographic Information.........................31 Appendix F: Proposed Revised Symptom and Quality of Life Questionnaire.................32
ii List of Tables Table 1: Demographic Summ ary for All Respondents...........................................................9 Table 2: Demographic Summar y for Normal Respondents....................................................9 Table 3: Correlation Coe fficients of Symptom Group with Adjusted Symptom..................14 Table 4: Correlation Coefficients of Sy mptom Group with Likert Styled Quality..............14 Table 5: Correlation Coefficients of Symptom Group with CDC Styled.............................14 Table 6: Probability of Difference from Inclusion Group....................................................15 Table 7: Correlation Coefficients of Substances with Patient Groups.................................15 Table 8: Probability of Different Di stribution from Inclusion Group Symptom..................16 Table 9: Probability of Different Distribution from Inclusion Group..................................16 Table 10: Correlation coefficients between Item 24 and Symptom Group in Various........16
iii List of Figures Figure 1: Histogram of Total Sympto m Scores in Normal Respondents.............................10 Figure 2: Histogram of Likert Quality of Life Scores in the Inclusion Group.....................10 Figure 3: Histogram of CDC Styl ed Quality of Life Scores.................................................11 Figure 4: Scatter Plot of Likert versus CDC Styled Responses............................................11 Figure 5: Scatter Plot of Symptom Score versus Likert Quality of Life Scores...................12 Figure 6: Scatter Plot of Sy mptom Score versus CDC Styled Quality of Life Score...........13
iv A New Questionnaire to Determine the Frequenc y and Severity of Symptoms Caused by Inhaled Odors, Chemicals and Irritants in No rmal Subjects and Their Relation to HealthRelated Quality of Life Stephen E. Williamson, M.D. ABSTRACT Abstract Individuals may develop symptoms in response to inhaled odors, chemicals, and irritants. This may affect their quality of life. Little is known about the prevalence and severity of symptoms that result from exposure to odor s, chemicals and irritants. This study demonstrates the development of a new respirat ory questionnaire to de tect the prevalence and severity of symptoms expe rienced upon exposure to chemical s, odors, and irritants, and relates these symptoms to quality of life. This questionnaire was submitted to 96 volunteers at the University of South Florida College of Public Health who responded to items regarding symptoms developed in response to exposure to automobile exhaust, cigarette smoke, strong smells, cologne, perfumes or scente d candles, or fresh pain t vapors or fumes. Health-related quality of life was assessed usi ng a subscale included with the questionnaire. The number and severity of symptoms deve loped in response to exposure to odors, chemicals, and irritants showed a strong negative corre lation with health-related quality of life, consistent with intuitive estimates of the direction of this relationship. Also, it was shown that in normal populations, males and fema les develop statistically similar prevalence and severity of symptoms in response to e xposure to odors, chemicals, and irritants.
1 Hypothesis and Specific Aims Hypothesis The frequency and severity of symptoms re ported by those who are exposed to inhaled odors, chemicals and irritants can be determined by a self-administered questionnaire, and these symptoms affect their reported quality of life. Specific Aims I: To develop a self-administered questionn aire and demonstrate that it detects the frequency and severity of symp toms reported by normal individuals on their experience with odors, chemicals, and irritants. II: To determine the relationship between se lf-reported quality of life and severity of reported symptoms in normal individuals' expe rience with inhaled odor s, chemicals, and irritants.
2 Introduction Questionnaires are an important and powerful tool to explore occupational health problems. They are easy to apply, inexpensive to administ er, and readily interpreta ble. The proper use of a questionnaire is dependent on careful desi gn, subsequent verifi cation of validity and reproducibility, and the close m onitoring of its application.1 Here, we will demonstrate the creation of a questionnaire and its use to dete rmine the effects of inhaled substances on normal individuals with regard to symptoms they experience and the effect of their symptoms on quality of life. Many airborne chemical substances with an odor may produce sensory ir ritation in the eyes, nose and throat, as well as acute neurotoxic sy mptoms. Such symptoms are dose related and all persons experience them. Some persons who sense the odor from certain chemical substances may feel ill, although this reaction is not experienced by others with the same exposure. Irritant symptoms may arise in envi ronments in which inhaled substances are at levels lower than regulation levels,2 and some say that chemical sensitivity is due to inadequate exposure limits for chemicals.3 Only a few epidemiologic studies from different work places have been performed to study this phenomenon.4 It is important to understand this phenomenon to protect workers properly. Individuals may incur great pers onal costs in order to ease their symptoms, whatever the cause or the current term for th eir symptoms or whether they in cur the symptoms at home or work. Studies have shown that workers have had to change jobs4 or are perhaps drawn or pushed into other jobs5 because they could not tolerate the symptoms they sustain when exposed to odors and irritants at work. The field of odor and irritant sensitivity requ ires understanding the contributions of many authors in the fields of anatomy, phy siology, pathophysiology, and psychology. The literature covering this phenomenon is huge. A basic problem facing the study of sensitivity to inhaled substances is determining if the sy mptoms occur due to psychological factors or physiological factors. Broadly, areas of investigation include olfactory physiology, respiratory physiology, and psychology. People are exposed to airborne substances at home and at work. Many of these substances not only produce an odor but can evoke ocular nasal and throat irritation, as well as neurotoxic symptoms. It is usef ul to define the airborne subs tances of interest as odors or irritants. Briefly, odors stimul ate the olfactory nerve (cranial nerve I). Irritants cause inflammatory effects.6 Lateralization testing demonstrat es this difference. Single nostril stimulation with an irritant cau ses lateralization. That is, one can tell into which nostril the
3 stimulus is applied in the case an irritant, su ch as ammonia. An odor, however, infused into one nostril is sensed bilaterally.7 Many substances have odorous and irritant properties, but not all substances that ar e irritating are odorous (NO2),8 and not all odorous substances are irritating (phenyl ethyl alcohol).2 Irritant substances may be a simple molecule, such as acrolein and ozone or a complex mixture, such as tobacco smoke.9 However, inhaled odors and irritants can elicit complaints of eye, nose, and throat irritation, headache, nausea, diarrhea, hoarseness, sore throat, cough, ch est tightness, nasal c ongestion, palpitations, shortness of breath, stress, drow siness, and alterations in mood.8 The number of individuals who suffer from these symptoms is not known accu rately, as estimates of the prevalence of this chemical sensitivity range from about 11 million10 to 90 million11 individuals. The difference in prevalence is likely due to different case definitions. The field studying sensitivity to inhaled substances suffers from the lack of a consistent case definition or the name for phenomenon of sensitivity to inhaled substances,4 therefore, the results of studies are difficult to compare. O dor perception is highly id iosyncratic, and this variability affects subjective responses. Numerous factors, including adaptation, medications, aging, nutritional stat us, pregnancy, and a variety of diseases and disorders can affect odor perception. Also, disorders of the olfactory system affect over 2.7 million Americans.7 Some of the symptoms of diseases such as asthma and allergic rhinitis overlap those symptoms developed after exposure to odo rs and irritants. Furthermore, allergic symptoms can be exacerbated by no n-allergenic inhaled substances.10, 12 Although it may appear to be straightforward to expose individuals to a chemical and ask them to report on the level of irritation they experience, the potentia l for confusion between olfactory and irritant modalitie s has produced extreme variability in direct scaling of upper airway irritant sensations by i ndividuals with intact olfactory and trigeminal systems. This confusion is exacerbated when chemicals present in the air at levels that stimulate only odor sensation can prompt exposed individuals to re port Â‘irritationÂ’, even if the perception is largely mediated through the psychological di scomfort incurred by smelling the odor of an unfamiliar or unpleasant chemical, or even from misattributions of unrelated symptoms that happen to coincide with chemical exposure.2 More simply, subjects may report to researchers "irritation" by odors that are merely unpleasant or unfamiliar, so this study design yields confusing results. Psychologically, hypotheses have focused on ps ychiatric disorders, personality traits13, 14 and mass hysteria.15 Responses to inhaled chemicals are subjective, variable and subject to suggestibility bias. Suggestion and/or suggestibility bias eff ects were demonstrated in a study in which people given positive informati on about odors to which they were exposed had less perceived irritation compared to people given negative information.7 In another study, reports of sensory irritation in the workplace were in fluenced by the reactions of coworkers and other bystanders to an odor.16 Instructing persons to attend to evidence of Â“nasal obstructionÂ” as they breathed induced mo re symptoms than instructing them to attend to the Â“free passage of air.Â” A significant amount of the vari ation in irritant and symptom perception in normal, healthy i ndividuals can be attributed to differences in personality orientations. Positive affective orientations appear to lower individuals' expectancies of
4 becoming ill, while negative orientations app ear to heighten those same expectancies.2 Although psychiatric causes may be an important source for symptoms, studies of patients with multiple chemical sensitivity showed no psychiatric diagnosis in up to 50%.13 Demographic studies have shown that women, persons of relatively higher socioeconomic status, and in office rather than heavy industr y positions are more likely to report symptoms from inhaled irritants.17 It is not understood why workers in heavy industry, who intuitively would have greater exposure to airborne chemicals, have less reported symptoms. Physiology affects irritant a nd odor perception. Irritants ty pically obey a sigmoid doseresponse relationship with a threshold under which no irritation reaction occurs. Many toxicological irritants show reversible effects Â– the infl ammation disappears after exposure to the irritant ceases, however, acute exposures to high levels of some irritants can result in irreversible effects.9, 18 Human physiology can also affect this perception, as experiments have shown that odor intensity depends on the effort associated with inspiration.19 Interestingly, research shows that odors ar e perceived to be more pleasant and odor identification is more accurate when smelled with the right nostril than with the left.20 Study limitations aside, there ar e at least three mechanisms by which ambient odors may produce health symptoms. First, symptoms can be induced by exposure to odorants at levels that also cause irritation or ot her toxicological effects. That is, irritation, ra ther than the odor is the cause of the heath symptoms, and odor, the sensation, simply serves as an exposure marker. Second, health symptoms fr om odorants at nonirritant concentrations can be due to innate (genetically coded) or learned aversions. Third, symptoms may be due to a copollutant (such as endotoxin) that is part of an odorant mixture.4 A problem has developed, however, as there is a group of pa tients who have been given a diagnosis most commonly called Multiple Chemical Sensitivit y. The incidence of Multiple Chemical Sensitivity is said to be increasing.3, 17 It has been defined as, "an acquired disorder characterized by recurrent symptoms, referabl e to multiple organ systems, occurring in response to demonstrable exposure to many ch emically unrelated compounds at doses far below those established in the general population to cause harmfu l effects. No single widely accepted test of physiological function can be shown to be correlated with the symptoms.Â”21 Media sources have sensationalized some asp ects of this sensitivity by reporting on "sick buildings" and "20th Century Dise ase." Once the media and legal system have reinforced to the general public that a product is dangerous, scientific evidence th at comes after this process is not likely to be reassu ring, and the misperception endures.22 Waddell's comments in 1992 are still true. "The salient problem w ith MCS is that there is no consistent and specific effect from exposure to any speci fic chemical. This does not allow for any objective test for any disease entity that might be caused by the chemicals as indicated by the theory of MCS. The effects of exposur e to chemicals as defined today by MCS are subjective, and no report is available to convi ncingly demonstrate that these effects would not have occurred merely by chance."23 To summarize, odor and irritant percepti on is a complicated field of study, with disagreement about case definition, prevalen ce, and methodological problems that exist on
5 physiological and psychological levels. It is obvious that this field needs more study in order to answer the question, "Â… is it the agents or the respo nder?" To that end, we have created a questionnaire to identify normal i ndividuals who seem to develop more severe symptoms from inhaled environmental odors, ch emicals, and irritants. These individuals may be suitable for further physiological studies in the Breath Laboratory at the University of South Florida. Hopefully, this and further studies in this laboratory will help determine the cause of symptoms in this troubled population.
6 Methods The Institutional Review Board of the Univer sity of South Florida reviewed the study proposal and exempted the study from further review. This study describes the development and implementation of a self-administered questionnaire to demonstrate the frequency and severity of symptoms a normal adult population recalls when queried about their e xperience with exposure to four commonly encountered airborne odors and irritants -cigarette smoke, pe rfumes, automobile exhaust, and paint fumes and relates these responses to their responses to quality of life queries. The questionnaire was presented to volunteers as part of a packet contained in a selfaddressed University of South Florida in teroffice envelope from February 1, 2007, to February 28, 2007. The envelope contai ned a cover letter (A ppendix B), a 23 item demographic and medical questionnaire (Appe ndix C), and a two page 42 item symptom and quality of life quest ionnaire (Appendix D). The symptom questionnaire was created with input from consultants of the College of Industrial and Organizational Ps ychology at the University of South Florida, who have reviewed the questionnaire for r eadability and structure. Quality of life items are derived from intuitive queries in addition to selected items from the Centers for Disease Control Prevention Health-Related Quality of Life 14 Item Measure's "Healthy Days Core Module."24 These items were included since they have been validated in other studies and increase the information contained in the data from ordinal to the interval level. The medical questionnaire was used to define the study population afte r the results were returned. It contained items concerning smoking history and medi cal and psychological conditions that could cause interference with nor mal olfaction and irritant sensation, all of which were used as exclusion criteria. Th e demographic portion of the questionnaire obtained sex and age information. No persona lly identifiable information was obtained. Information on age was used to study adults from 18 to 80 years old. The entire questionnaire took about 10 minut es to present and complete. The questionnaire itself asked respondents to rate their likelihood of agreement with statements relating to symptoms they experi enced when they encountered cigarette smoke; strong smells; cologne, perfumes or scented ca ndles; automobile exhaus t; or paint fumes in the past year. The responses were scored on a 5-point Likert scale (1 = almost never; 5 = almost always). The questionnaire containe d subscales for ear, nose, and throat, lower respiratory, general, gastrointestinal, a nd neuropsychological symptoms. To simplify symptom scoring, the form of the statements is such that agreement with the experience of
7 symptoms results in a higher Likert score. In that way, more agreement with the statements resulted in higher raw symptom scores. In this questionnaire, two formats were compared to each other to quantify quality of life. In one format, four items were on a Likert scale, and in the other format based on the CDC items, 3 ite ms were in a fill in the blank format, which will be referred to henceforth as Â“CDC styledÂ” items. Volunteers were obtained by submitting the que stionnaire to male and female staff and students of the College of Public Health at the University of South Florida during business hours, personally delivered by the Principal Inve stigator. A kiosk in the College of Public Health was also used to attract volunteers from student and st aff traffic in the lobby of the College of Public Health, also attended by the Principal Inve stigator. A flyer on the kiosk was used to attract attention and provide initial informati on to potential volunteers. Professors allowed the questionnaire to be distributed to a class in two cases. Using standard deviation from the mean to define "high" and "low" scores, Magnavita15 was able to demonstrate a significant difference (p<0.001) in total symp tom scores between cacosmic and noncascomic patients in a population of 47 subjects. We proposed, then, to obtain 50 included subjects by submitting 100 ques tionnaires, expecting to exclude about half of the returned questionnaires. Volunteers were asked to respond to the questio nnaire and mail their re sults via inter-office mail to the College of Public Health inbox, addressed to the principal investigator. The subjects completed the questionnai re privately in order to minimi ze social acceptability bias. A new pencil with the USF logo (value $0.50) was enclosed in the envelope with the questionnaire to compensate respondents for their time to complete the questionnaire. In order to protect the privacy of participants exclusion from analysis was performed after the questionnaires were returned. Responses were excluded if any of the items were answered positively in the medical questionnai re, except for one item, #3, which asked if the respondent considered themselves a healthy pe rson. Items that were left blank on the Medical and Demographic portion of the questio nnaire were considered a positive response and resulted in exclusion from the normal group. All raw data has been dated and filed in a locked cabinet in the Respiratory Laboratory in the College of Public Health for two years, and then destroyed. Respiratory Labo ratory personnel will hold the key. Quality of life was measured by 4 questions wi th Likert scales and with 3 questions using CDC styled items. The responses on the Likert format were summed to achieve a quality of life score, and the CDC styled qu estions were added to obtain a nother quality of life score. The possible range of values for summed Likert responses is 4 to 20. The possible range of values for the CDC styled scores was 0 to 90. Each of these scores were compared to total symptom scores to determine the relationshi p between quality of life scores and total symptom scores in response to inhaled odors, chemicals, and irritant s, using simple linear regression.
8 Statistical calculations were performed usi ng Microsoft Excel 2003. A symptom score with a possible range of 29-145 was obtained by sum ming the symptom ratings for each included subject. Scores for the five symptom gr oups were obtained by selecting the items corresponding to each system and summing the responses by system.
9 Results Ninety-six questionnaires were submitted to volunteers, and 82 we re returned. Fifty respondents identified themselves as female, a nd 28 identified themselves as male. Four respondents did not specify age or sex. Me dian age for all respondents was 31 years, ranging from 17 to 73. (Table 1) For refere nce, all items were numbered from the first demographic item, through the medical excl usion items, through the symptom items, and through the final quality of life items. These numbers appear in the far left column in the appended questionnaires. Table 1: Demographic Summary for All Respondents n Lowest AgeMedian AgeHighest Age All 8217 31 73 Males 2818 33.5 73 Females 5017 29 61 Unspecified 4 The intended group of respondents for this study, the inclusion group, were obtained by using the responses given on th e Medical and Demographic In formation page (Appendix C) to obtain a population without major physiologi cal or psychological disease between the ages of 18 and 80. Four responses had data of such poor quality they were excluded on this basis alone. Exclusion criteria removed 55 more responses, leaving 23 responses included for further statistical analysis, which was cal led the normal sample. The inclusion group consisted of 13 females, 9 males and one unsp ecified age and sex. The median age for all normal respondents for which data was specified was 26, ranging from 18 to 47. (Table 2) Table 2: Demographic Summ ary for Normal Respondents n Lowest AgeMedian AgeHighest Age All 2318 26 47 Males 9 18 28 38 Females 1319 27 47 Unspecified 1 Total symptom scores for the inclusion group were derived from a Li kert scale and thus treated as ordinal variables. These scores ranged from 35 to 106, representing the sum of all
10 Likert responses to specific symptom queries The median score is 63.5. The frequency distribution of symptom scor es is shown in Figure 1. Figure 1: Histogram of Total Symp tom Scores in Normal Respondents 0 1 2 3 4 5 6 730 35 40 45 50 55 60 65 70 75 80 85 90 95 10 105 110 115Symptom ScoreN The quality of life scores for the inclusion group ranged from 4 to 10 on the Likert scaled items, out of a possible range of 4 to 20, repres enting the sum of responses to each of the 4 Likert items. A histogram of the quality of life scores on the Likert sc ale is shown in Figure 2. Increasing values on the Likert scale represented diminishing quality of life. Figure 2: Histogram of Likert Quality of Life Scores in the Inclusion Group 0 2 4 6 8 10 12 345678910 Likert QOL ScoreN
11 The histogram in Figure 3 shows the frequency of responses to the fill in the CDC styled quality of life scores in the inclusion group. Three responses from the above group of Likert respondents were not included in Figure 3 due to missing data. Increasing QOL score is in the direction of more days of disability. Figure 3: Histogram of CDC Styled Quality of Life Scores 0 1 2 3 4 5 6 01234567891011 CDC QOL Score N The adjusted Spearman rank correlation = .463 for the relationship between Likert-type items and the CDC styled items in the inclusion group. The scatter plot of the raw quality of life scores is repr esented in Figure 4. Figure 4: Scatter Plot of Like rt versus CDC Styled Responses 0 2 4 6 8 10 12 024681012 CDC QOL ItemsLikert QOL Items
12 To show the relationship between quality of life score and symptom score, each of the two types of quality of life (Likert and CDC styl ed) scores were compared to total symptom scores in the normal patients. Again, there are three less comparisons in the CDC styled comparison than the Likert formatted items due to missing data. The relationship between the Likert styled items and symptom scores in normal patients yields an adjusted Spearman rank correlation = .916. The scatte r plot and the best fit regre ssion line for these points are in Figure 5. Figure 5: Scatter Plot of Symptom Score versus Likert Quality of Life Scores 0 2 4 6 8 10 12 020406080100 Symptom ScoreLikert QOL Score When the CDC styled quality of life scores are compared with the symptom scores, the regression function gives an adjusted Spearma n rank correlation = 0.725. These data are represented in Figure 6 with a trendline show ing the best fit for the regression function between the variables.
13 Figure 6: Scatter Plot of Symptom Score versus CDC Styled Quality of Life Score -2 0 2 4 6 8 10 12 020406080100 Symptom ScoreCDC QOL Score The responses in each symptom group were averag ed to give a system score, and the sum of these averages gave an adjusted symptom score that adjusted for the num ber of items in each symptom group, which ranged from 4 to 9 items. This controlled for the different weights given to the total symptom score due to th e different number of items in each symptom group. The unadjusted and adjusted total symptom scores were very similar, with adjusted Spearman correlation coefficient between the two scores = 0.99. The possible adjusted symptom score ranged from 5 to 25. The 5 symptom groups were compared by relating their average score to th e adjusted symptom score. This was done for several subsets of the responses, including the inclusion group, th e excluded group, an Â“atopic groupÂ”, and smokers. These populations are not necessarily exclusive. The excluded group consisted of all responses between the ages of 18 and 80 that are not in the inclus ion group. The Â“atopic groupÂ” consists of all excluded responses that would have been in the inclusion group but for a positive response to the items, Â“Do you get hay fever, seasonal allergies, or allergic rhinitis?Â” or Â“Do you have eczema or hives?Â” Sm okers were defined as those excluded, but which would have been in the inclusion gr oup except for a positive response on the items, Â“Are you a smoker?Â” or Â“Have you smoked in th e last 10 years?Â”. Adjusted Spearman correlation coefficients for these relationshi p between the system symptom score and the total adjusted symptom score are presented in Table 3. LoResp represents the lower respiratory system, and NeuroP repr esents neuropsychological symptoms.
14 Table 3: Correlation Coe fficients of Symptom Group with Adjusted Symptom Score NormalExcludedAtopic Smokers Average General 0.86 0.87 0.75 0.73 0.80 ENT 0.92 0.83 0.90 0.91 0.89 GI 0.93 0.76 0.24 0.89 0.71 LoResp 0.66 0.91 0.48 0.95 0.75 NeuroP 0.80 0.44 0.81 0.62 0.67 Average 0.83 0.76 0.64 0.82 0.76 The symptom scores of these patient groups we re compared to the Likert and CDC styled quality of life scores. The adjusted Spearman correlation coefficients for the Likert styled items appear in Table 4, and for the CDC styled items in Table 5. Table 4: Correlation Coefficients of Sy mptom Group with Likert Styled Quality of Life Items NormalExcludedAtopic Smokers Average General 0.84 0.79 0.43 0.81 0.72 ENT 0.89 0.86 0.07 0.85 0.67 GI 0.88 0.70 0.64 0.82 0.76 LoResp 0.49 0.90 0.52 0.82 0.68 NeuroP 0.80 0.77 0.48 0.77 0.70 Average 0.78 0.80 0.43 0.81 0.71 Table 5: Correlation Coefficients of Symptom Group with CDC Styled Quality of Life items NormalExcludedAtopic Smokers Average General 0.65 0.11 -0.31 0.26 0.18 ENT 0.72 -0.05 -0.33 0.00 0.09 GI 0.60 0.04 -0.13 0.00 0.13 LoResp 0.35 0.04 -0.46 0.07 0.00 NeuroP 0.65 -0.19 -0.30 0.26 0.11 Average 0.59 -0.01 -0.31 0.12 0.10 The Mann-Whtney U test was used to determine the probability that the distributions of the different symptom groups in the different pati ent groups were different than that of the inclusion group. The p values are presented in Table 6:
15 Table 6: Probability of Difference from Inclusion Group ExcludedAtopic Smokers General 0.00 0.23 0.08 ENT 0.01 0.01 0.19 GI 0.31 0.25 0.73 LoResp 0.10 0.49 0.07 NeuroP 0.27 0.22 0.07 Items # 25 Â– 28 asked about respondents' reaction to cigarette smoke, automobile exhaust, strong smells, cologne, perfumes or scented ca ndles, and fresh paint vapors or fumes (the substances in the table below). These res ponses were summed and compared with the symptom scores of the different patient groups. The adjusted Spearman correlation coefficients are shown in Table 7: Table 7: Correlation Coefficients of Substances with Patient Groups NormalExcludedAtopic Smokers Cigarette 0.21 0.49 0.48 0.27 Exhaust 0.12 0.40 0.23 0.17 Scents 0.11 0.50 0.60 0.52 Paint 0.18 0.50 0.47 0.27 General health was queried twi ce, using two different formats. The first format gave a yes/no choice, and the second gave a 5 level Like rt range of responses. In all cases, items answered "yes" to the query wh ether the respondent considered themselves a healthy person were answered from 1-3 in the Likert s cale, representing excellent, very good, and good health. All "no" answers in the yes/no format were answered from 4 to 5 in the Likert representation of the query, representing fair or poor health. The adjusted Spearman correlation coefficient of this relation = .363. There was a group of 15 respondents who were labeled "atopic", as they were excluded from the normal group only by their positive response to the items on the medical portion of the questionnaire relating to asthma, antihistamin e use, hay fever, and eczema. The median symptom score for this Â“atopicÂ” group was 65 ( n = 15), and it was 54 ( n = 23) for the normal group. Wilcoxon's two group rank-sum method25 was used to test th e hypothesis that the means of the groups were the same. The differe nce in the groups was significant for the two tailed test at = 0.1, with obtained t = 3.53 and critical t = 2.673. The Mann-Whitney U test was used to dete rmine the likelihood that patient groupsÂ’ distribution of symptom sc ores and quality of life scores ( on the two different quality of life styles) differed from the inclusion group scores These are presented in Tables 8 and 9.
16 Table 8: Probability of Different Di stribution from Inclusion Group Symptom Scores Excluded 0.23 Atopic 0.16 Smokers 0.03 Table 9: Probability of Different Distribution from Inclusion Group Quality of Life Scores LQOL CQOL Excluded 0.40 0.37 Atopic 0.30 0.39 Smokers 0.12 0.28 Symptom scores for males were significantly less than those for females only in the inclusion group, with a one tailed p = 0.048. This same test shows that the mean symptom scores are not significantly di fferent for males and females in the normal group, where p = 0.5 for the difference. The adjusted SpearmanÂ’s rank correlation coeffi cient relating age with total symptom scores = .239 in the normal group and = -.054 in the excluded group. Item # 24 asked respondents to identify their agreement with the stat ement, "I am more sensitive to inhaled chemicals, irritants, odors, or strong fragra nces than the average person." The adjusted SpearmanÂ’s rank correlation coefficients relating the response to this single item to the symptom groups in the di fferent patient groups in Table 10. Table 10: Correlation coefficients between Item 24 and Symptom Group in Various Patient Groups Normal ExcludedAtopic Smokers Average General 0.30 0.41 0.47 -0.03 0.29 ENT 0.54 0.39 0.43 0.36 0.43 GI 0.38 0.27 -0.04 0.42 0.26 LoResp 0.09 0.45 0.42 0.47 0.36 NeuroP 0.36 0.46 0.69 0.17 0.42 Average 0.33 0.40 0.40 0.28 0.35
17 Discussion Demographic comparison showed a similar age distribution between males and females, but female respondents numbered twice those of ma les, reflecting the student population of the College of Public Health at th e University of South Florida. The statistical analysis of the data in this cr oss-sectional study show that there is a bimodal distribution of symptom scor es that adults between th e ages of 18 and 80 without confounding disease states (the inclusion group) reports on thei r experience with cigarette smoke, automobile exhaust, strong smells, col ogne, perfumes or scented candles or fresh paint vapors or fumes. This could represent a physiologic di fference in individuals in the two groups, the normal individuals without sy mptoms and the normal individuals with symptoms. The nadir of the curve between mo des provides a good point to divide "normal asymptomatic" and "normal symptomatic" re sponses. This point would be a total symptom score somewhere between 65 and 70, as this score seems to discriminate between the groups on gross inspection of the histogram of symptom scores. The quality of life scores obtained from th e Likert scale items in this included group correlate fairly well with symptom scores, whereas the quality of life scores from the fill in the blank items correlate less well with sympto m scores. Also, the quality of life scores from the two different item formats correlate po orly with each other. None of the system subscores agreed well with the CDC styled quality of life in any patient group other than the inclusion group. The hypothesis that increasing symptoms aff ects the health-related quality of life was supported, as shown in Figures 5 and 6, wher e the regression slope is not zero, but is positive in this case. This shows that increasi ng symptoms result in decreased health related quality of life perceptions. On average, of all the symptom categories, ea r, nose, and throat symptoms correlated best with total symptom scores, and neuropsychologi cal symptoms correlated least well. High correlation of the total symptom score with ENT symptoms may result from the fact that this system is likely evolved to detect inha led odors, chemicals, and irritants. Interestingly, the quality of life scores for th e atopic individuals was very poorly correlated with quality of life. Perhaps the symptoms of atopy have incorporated into the background sensations of these individuals to the point that these symptoms are not perceived as severe enough to report.
18 The only patient group system scores that differe d at the 0.05 level of significance from the inclusion group was the General and ENT scores in excluded respondents and ENT scores in the atopic group. Item # 24, regarding the overall question rega rding oneÂ’s perceived sensitivity to odors, chemicals, and irritants correlated fairly well with final symptom scores in all groups. Magnavita, in his study of odor sensitivity in healthy workers, asked food store workers to rate the likelihood that each of 10 substances we re able to cause them symptoms of physical illness. He found that self-identified odor into lerance was significantly associated with the frequency of physical symptoms. Pearson pr oduct-moment calculations in the present study agree with this conclusion, finding a coeffi cient = 0.55 between th e item asking if an individual considered himself sensitive to odor s, chemicals, and irr itants, and their total symptom score. He also found that females developed symptoms more frequently than males without regard to preexisti ng disease. A similar analysis of our data shows the same effect when exclusions in this study ar e not applied, similar to his method. Comparisons between similar information asked at different points or asked a different way are useful to assess internal va lidity. There was agreement in direction of the answers to the items. All those who felt they were generally he althy also agreed that they enjoyed good to excellent health. All those who felt they were not generally healthy later agreed that they suffered fair or poor health. Biases in this questionnaire implementation include a severe selection bias. This study sample represents a mostly young, ambulatory, healthy, educated populat ion. This reflects the environment in which it was distributed. Ma les and females are not represented equally. The ethnic characteristics of this sample ar e not quantitatively known, but qualitatively, it can be said that the responde nts were of several different ethnicities, including African, Caucasian, Hispanic, Black American, and Asia n, with Caucasian representing the great majority of respondents. This will affect external generalization. As experience with the questionnaire grows, and the population to which it is submitted broadens, external generalizability will become more possible.
19 The Final Questionnaire This questionnaire was a pilot study to create a final questionnaire to inexpensively identify normal respondents who may suffer from symp toms when exposed to inhaled odors, chemicals, and irritants in a stronger way than other normal individuals. This is to select those for physiological testing that are more likely to demons trate a physiological difference when tested by inhalation challenge which satisfies in part the ethical duty to minimize risks to those not likely to bene fit from investigation. After analysis of this pilot questionnaire, it can be simplifie d and it needs modification. Several respondents reflected the teaching in the College of Pub lic Health that sex does not equal gender, and the more appropriate que stion in this item would be gender. Item # 3, asking, Â“Do you consider yourself a hea lthy person?Â”, can be dropped, as it is such a general query it should not re sult in inclusi on or exclusion in the normal group; that is, a volunteerÂ’s responses should not be segregated into the exclusi on group if this were the only item answered positively on the medical questionnaire of ex clusion criteria. Only one respondent reported an abnormal sens e of taste and a normal sense of smell. All other cases would have been excludable on the basis of abnormal sense of smell only. Also, the sense of taste is not critical to sensations of odors, chemicals, and irritants, so it (Item # 11) can be dropped. Consideration should be given to dropping the item on heart di sease (Item # 14). The item asking about heart medication should be more specific, as some respondents may consider antihypertensive medication to be "h eart medication", yet hypertension per se does not indicate heart disease. This problem is magnified by the relatively high prevalence of antihypertensive treatment compared to the preval ence of significant heart disease. If Item # 14 is kept, it should be preceded by an item as king about hypertension, such as "Do you take medicine for high blood pressure?". Also, it is not criti cal that persons with heart disease are excluded from the normal group, as there is no known link between heart disease and perception of inhaled odors, chemicals, and ir ritants. Excluding re spondents on the basis of this disease may be too restrictive. The item regarding psychiatric disorders (Item # 19) should be more specific to definitely include those suffering from de pression, which has been shown to influence responses to inhaled odors. Antidepressant therapy could be used as a proxy for the diagnosis of significant depression. It would be more accura te than self-diagnosis of depression without
20 developing another questionnaire to merely establish this point. The proposed items for this medical item are A: "Do you take medicine fo r depression?" and B: "Do you suffer from any other psychiatric disorder?" A positive resp onse to either should result in exclusion from the normal group. Near-identical responses to similar symptoms can be simplified by throwing out one or the other. These are the burning (Item # 31) and ti ngling (Item # 32) of the face items. They elicited identical respons es, and therefore one is predictive of the other 100% of the time, so either can be discarded. As few agreed with either statement, both could be dropped. Item # 61 is moved upward in the item bank to follow the other symptoms and allow the quality of life items to finish the secti on in order to ease analysis and scoring. Two scales were tested to indica te quality of life. The Likert styled items, as they correlated more closely with symptom scores than the CD C styled items, should be used if quality of life is to be measured in the future. Ther efore, in the final version, Items # 62-65 are dropped. (Appendix E) The final exclusion panel change s are shown in gray. It will consist of 20 items, including age and gender. The final sy mptom questionnaire, excluding quality of life items, will consist of 33 items. It is expected that, using the considerations above, this simpler questionnaire will retain the ability of the pilot questio nnaire to detect the frequency and severity of symptoms re ported by normal individuals on their experience with odors, chemicals, and irritants, and relate that with their quality of life. As the items that were excluded from the symp tom queries were mostly removed due to low prevalence of positive responses, the divi ding symptom score between normal and responsive normal subjects can continue to be somewhere between 65 and 70 until the distributions are better defi ned for the symptom scores.
21 List of References 1. Merchant J. Occupational Respiratory Di seases. U.S. Department of Health and Human Services, 1986. 2. Dalton P. Upper airway irr itation, odor perception and heal th risk due to airborne chemicals. Toxicol Lett 2003; 140-141:239-48. 3. Ziem G, McTamney J. Profile of patien ts with chemical injury and sensitivity. Environ Health Perspect 1997; 105 Suppl 2:417-36. 4. Moen BE. Chemical sensitivity and the wo rk place environment: research needs. Psychoneuroendocrinology 2005; 30(10):1039-42. 5. Cullen MR. Epidemiologic methods for th e study of occupational asthma. Current problems and solutions. Chest 1996; 109(3 Suppl):51S-54S. 6. Nowak D. Chemosensory irritation and the lung. Int Arch Occup Environ Health 2002; 75(5):326-31. 7. Chrostowski P, Foster S. Odor Percepti on and Health Effects. Water Environment Federation 2003 Annual Technical Exhibi tion and Conference. Los Angeles, California: CPF Associates, Inc., 2003. 8. Schiffman SS, Williams CM. Science of odor as a potential health issue. J Environ Qual 2005; 34(1):129-38. 9. Bascom R. The upper respiratory tract: mu cous membrane irritation. Environ Health Perspect 1991; 95:39-44.
22 10. Meggs WJ, Dunn KA, Bloch RM, et al. Prevalence and nature of allergy and chemical sensitivity in a general populat ion. Arch Environ Health 1996; 51(4):27582. 11. Kreutzer R, Neutra RR, Lashuay N. Pr evalence of people reporting sensitivities to chemicals in a population-based surv ey. Am J Epidemiol 1999; 150(1):1-12. 12. Shim C, Williams MH, Jr. Effect of odors in asthma. Am J Med 1986; 80(1):18-22. 13. Caccappolo-van Vliet E, Kelly-M cNeil K, Natelson B, et al. Anxiety sensitivity and depression in multiple chemical sensitivities and asthma. J Occup Environ Med 2002; 44(10):890-901. 14. Dalton P. Odor, irritation and perception of health risk. Int Arch Occup Environ Health 2002; 75(5):283-90. 15. Magnavita N. Cacosmia in healthy worker s. Br J Med Psychol 2001; 74(Pt 1):121-7. 16. Dalton P. Cognitive influences on health symptoms from acute chemical exposure. Health Psychol 1999; 18(6):579-90. 17. Miller CS. The compelling anomaly of ch emical intolerance. Ann N Y Acad Sci 2001; 933:1-23. 18. Alberts WM, Brooks SM. Reactive airway s dysfunction syndrome. Curr Opin Pulm Med 1996; 2(2):104-10. 19. Landis BN, Hummel T, Hugentobler M, et al. Ratings of overa ll olfactory function. Chem Senses 2003; 28(8):691-4. 20. Lledo PM, Gheusi G, Vincent JD. In formation processing in the mammalian olfactory system. Physio l Rev 2005; 85(1):281-317. 21. Cullen MR. The worker with multiple chem ical sensitivities: an overview. Occup Med 1987; 2(4):655-61.
23 22. Nanagas KA, Kirk MA. Perceived pois ons. Med Clin North Am 2005; 89(6):135978. 23. Waddell WJ. The science of toxicology a nd its relevance to MCS. Regul Toxicol Pharmacol 1993; 18(1):13-22. 24. Centers for Disease Control and Prev ention. Centers for Disease Control and Prevention Health-Related Quality -of-Life 14-Item Measure. 2005. 25. Blair RC, Taylor RA. Biostatistics for the Health Sciences: Prentice Hall, 1999.
25 Appendix A: Flyer. Formatting altered to fit format constraints. A Questionnnaire and 15 minutes could score you a shiny new USF pencil! Details here Pencil Image (813) 943-7960 (813) 943-7960 (813) 943-7960 (813) 943-7960 (813) 943-7960 (813) 943-7960 (813) 943-7960 (813) 943-7960 (813) 943-7960 (813) 943-7960 (813) 943-7960 (813) 943-7960 (813) 943-7960
26 Appendix B: Cover Letter To the volunteer: I am inviting you participate in a research proj ect to study symptoms that one develops after exposure to airborne irritants. Along with this letter is a s hort questionnaire that asks a variety of questions about this. I am aski ng you to look over the questionnaire and, if you choose to do so, complete it and send it back to me using the in teroffice envelope in which it came. It should take you about 15 minutes to complete. The results of this project will be used to re fine this questionnaire so it can be used to identify subjects for further study. Through your participation, I hope to identify those questions that are most and leas t useful. The development and results of this study will be used as the subject of my Masters Thes is in the College of Public Health. I do not know of any risks to you if you decide to participate in this survey, and I guarantee that your responses will not be identified with you. I promise not to share any information that identifies you with anyone. You should not volunteer to put your name or any other information on the questionnaire other than th at which is requested. If you do not feel comfortable completing the survey, discard it. The survey should take you about 15 minutes to complete. I hope you will take the time to complete this questionnaire and return it. Your participation is voluntary, and there is no penalty if you do not participate. Regardless of whether you choose to participate, the results will be on file at the University of South Florida Shimberg Health Sciences Library after April 7, 2007. If you have any questions or concerns about co mpleting the questionnaire or about being in this study, you may contact me at (813) 9437960. Alternatively, if you have questions about your rights, general questions, complaints, or issues as a person taking part in this study, you may also call the Division of Re search Integrity and Compliance of the University of South Florida at (813) 974-9343. Sincerely, and thank you, signature Stephen E. Williamson, M.D. Chief Investigator
27 Appendix C: Medical and Demographic Information. Formatting altered to fit format constraints. Demographics and History 1 Sex 2 Age Please place an "X" in the appropriate box below Yes No 3 Do you consider yourself a healthy person? 4 Do you take antihistamines? 5 Do you get hay fever, seasonal allergies, or allergic rhinitis? 6 Do you cough every day? 7 Do you suffer from respiratory problems? 8 Do you have asthma? 9 Are you a smoker? 10 Do you have a normal sense of smell? 11 Do you have a normal sense of taste? 12 Have you smoked in the last 10 years? 13 Have you received systemic steroids or antibiotics within the past 4 weeks? 14 Do you have congestive heart failure, cardiomyopathy, valvular heart disease, angina, cardiac arrhythmia, or had a myocardial infarction within the last 6 months? 15 Are you taking heart medication? 16 Do you have hepatitis or cirrhosis? 17 Do you suffer from renal failure? 18 Do you suffer from any neurologic disorder? 19 Do you suffer from any psychiatric disorder? 20 Are you pregnant or think you might be? 21 Do you have eczema or hives? 22 Do you have arthritis? 23 Has a doctor ever told you that you have fibromyalgia, chronic fatigue syndrome, or multiple chemical sensitivity?
28 Appendix D: Symptom and Qua lity of Life Questionnaire. Formatting altered to fit format constraints. Chemical, Odorant And Irritant Sensitivity Questionnaire This questionnaire asks about how you feel now and over the past year. Please check the box that most closely describes how you feel. Strongly Disagree Disagre e Uncertai n Agree Stron gly Agree 24 I am more sensitive to inhaled chemicals, irritants, odors, or strong fragrances than the average person If I am around the following I get this reaction: Nothing unusual I am bothere d A mild reaction Become somewhat ill Beco me very Ill 25 Cigarette smoke 26 Automobile exhaust 27 Strong smells, cologne, perfumes or scented candles 28 Fresh paint vapors or fumes If I am exposed to cigarette smoke, autom obile exhaust, strong smells, perfumes or colognes, or fresh paint vapors: Strongly Disagree Disagre e Uncertai n Agree Stron gly Agree 29 I suffer discomfort 30 I become sick 31 I develop burning in the skin of my face 32 I develop tingling in the skin of my face 33 I develop a funny sensation of the skin of my face 34 I develop eye irritation 35 I develop eye pain 36 I develop eye itching 37 I develop sore or burning nasal
29passages 38 I develop burning in my nasal passages 39 I develop a sore throat 40 I feel nauseated 41 I develop indigestion 42 I develop diarrhea 43 I get gas 44 I may cough 45 I may cough phlegm up 46 I feel like I can't get my breath. 47 I start wheezing 48 I feel tightness or pressure in my chest Continued on next page If I am exposed to cigarette smoke, autom obile exhaust, strong smells, perfumes or colognes, or fresh paint vapors: Strongly Disagree Disagre e Uncertai n Agree Stron gly Agree 49 I develop aching joints 50 I develop trouble sleeping 51 I develop numbness or tingling in my hands or feet 52 My body feels hot or cold 53 I become emotional 54 I get a headache 55 I become anxious 56 I have trouble concentrating 57 I miss work 58 I miss social or business appointments 59 I feel stress at home or work 60 I find it hard to interact with other persons 61 My symptoms ease if I can get away Excellent Very good Good Fair Poor 62 Would you say that in general your health is:
3063 Now thinking about your physical health, which includes physical illness and injury, for how many days during the past 30 days was your physical health not good? 64 Now thinking about your mental health, which includes stress, depression, and problems with emotions, for how many days during the past 30 days was your mental health not good? 65 During the past 30 days, for about how many days did poor physical or mental health keep you from doing your usual activities, such as self-care, work, or recreation?
31 Appendix E: Proposed Revised Medi cal and Demographic Information Demographics and History 1 Sex Gender 2 Age Please place an "X" in the appropriate box below Yes No 3 Do you consider yourself a healthy person? 4 Do you take antihistamines? 5 Do you get hay fever, seasonal allergies, or allergic rhinitis? 6 Do you cough every day? 7 Do you suffer from respiratory problems? 8 Do you have asthma? 9 Are you a smoker? 10 Do you have a normal sense of smell? 11 Do you have a normal sense of taste? 12 Have you smoked in the last 10 years? 13 Have you received systemic steroids or antibiotics within the past 4 weeks? 14 Do you have congestive heart failure, cardiomyopath y, valvular heart disease, angina, cardiac arrhythmia, or had a myocardial infarction within the last 6 months? 15 Are you taking heart medication? 16 Do you have hepatitis or cirrhosis? 17 Do you suffer from renal failure? 18 Do you suffer from any neurologic disorder? 19 Do you suffer from any psychiatric disorder? Do you take medication for depression? 20 Are you pregnant or think you might be? 21 Do you have eczema or hives? 22 Do you have arthritis? 23 Has a doctor ever told you that you have fibromyalgia, chronic fatigue syndrome, or multiple chemical sensitivity?
32 Appendix F: Proposed Revised Symptom and Quality of Life Questionnaire Chemical, Odorant And Irritant Sensitivity Questionnaire This questionnaire asks about how you feel now and over the past year. Please check the box that most closely describes how you feel. Strong ly Disagr ee Disa gree Unce rtain Agree Strongly Agree 24 I am more sensitive to inhaled chemicals, irritants, odors, or strong fragrances than the average person If I am around the following I get this reaction: Nothin g unusu al I am both ered A mild react ion Become somewhat ill Become very Ill 25 Cigarette smoke 26 Automobile exhaust 27 Strong smells, cologne, perfumes or scented candles 28 Fresh paint vapors or fumes If I am exposed to cigarette smoke, automobile exhaust, strong smells, perfumes or colognes, or fresh paint vapors: Strong ly Disagr ee Disa gree Unce rtain Agree Strongly Agree 29 I suffer discomfort 30 I become sick 31 I develop burning in the skin of my face 32 I develop tingling in the skin of my face 33 I develop a funny sensation of the skin of my face 34 I develop eye irritation 35 I develop eye pain 36 I develop eye itching 37 I develop sore or burning nasal passages 38 I develop burning in my nasal passages
3339 I develop a sore throat 40 I feel nauseated 41 I develop indigestion 42 I develop diarrhea 43 I get gas 44 I may cough 45 I may cough phlegm up 46 I feel like I can't get my breath. 47 I start wheezing 48 I feel tightness or pressure in my chest Continued on next page If I am exposed to cigarette smoke, automobile exhaust, strong smells, perfumes or colognes, or fresh paint vapors: Strong ly Disagr ee Disa gree Unce rtain Agree Strongly Agree 49 I develop aching joints 50 I develop trouble sleeping 51 I develop numbness or tingling in my hands or feet 52 My body feels hot or cold 53 My symptoms ease if I can get away Moved from item 61 to new position 54 I become emotional 55 I get a headache 56 I become anxious 57 I have trouble concentrating 58 I miss work 59 I miss social or business appointments 60 I feel stress at home or work 61 I find it hard to interact with other persons Delete the following if quality of life is assessed. Use above Likert styled items, # 58-61 Excell ent Very good Goo d Fair Poor 62 Would you say that in general your health is: 63 Now thinking about your physical health, which includes physical illness and injury, for how many days during the past 30 days was
34 your physical health not good? 64 Now thinking about your mental health, which includes stress, depression, and problems with emotions, for how many days during the past 30 days was your mental health not good? 65 During the past 30 days, for about how many days did poor physical or mental health keep you from doing your usual activities, such as self-care, work, or recreation?