| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
RESEARCH AND PRACTICE |
The authors are with the Industrial Relations Center, University of Minnesota, Minneapolis.
Correspondence: Requests for reprints should be sent to Irwin B. Horwitz, MSIR, MBA, 3-300 Carlson School of Management, 321 19th Ave S, Minneapolis, MN 55455 (e-mail: ihorwitz{at}csom.umn.edu).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONReferences |
|---|
Objectives. This study explored whether the prevalence of latex-related health conditions has increased among individuals employed in medical occupations relative to those employed in nonmedical occupations since the issuance of universal precautions in 1987.
Methods. Data derived from the 1983 to 1994 versions of the National Health Interview Survey were used to obtain odds ratios comparing prevalence rates of latex-related symptoms over time.
Results. No statistical evidence was found that the universal precautions resulted in increased prevalence rates of latex-related health conditions among medical workers relative to those employed in nonmedical occupations.
Conclusions. Increased use of latex gloves among health care personnel subsequent to the implementation of universal precautions appears to have had no effect on latex allergic reactions experienced by these workers.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONReferences |
|---|
Estimates of NRL sensitization among health care workers gained through the use of different methodologies, such as skin prick and patch tests, in vitro tests, and questionnaires, vary greatly between studies. For example, Kim et al. found a sensitization rate of 1.5%, whereas Yassin et al. reported a rate of 17%.7,8 Other studies involving health care workers have reported rates varying widely between these 2 extremes.9–12 In contrast, studies aimed at obtaining NRL sensitivity estimates for the general population have consistently reported sensitization rates between 5% and 7%.13–16 In many cases, point estimates of health care worker sensitization rates fall within the confidence intervals of general population rates.17–20 It is difficult, on the basis of these findings, to draw conclusions regarding sensitization prevalence rates among health care workers relative to those employed in other occupations.
A study conducted by the National Institute for Occupational Safety and Health that specifically attempted to determine whether NRL gloves were a source of sensitization by comparing health care workers with varying levels of exposure to gloves revealed no significant differences in sensitization rates between those who used gloves frequently and those who did not.21 Another hypothesis that has emerged from research on reactions to NRL is that cornstarch powder, frequently used as a donning agent, binds to the gloves’ NRL allergens and may act as an aeroallergen transmitter in health care environments. Similar to the hypotheses offered by other researchers focusing on glove use and sensitization, this hypothesis has received mixed support.
Several studies, for example, have produced evidence that workers who are exposed to NRL glove powder have a higher risk of becoming sensitized than do those who are not exposed.22,23 However, several large studies have recently presented a challenge to the earlier findings on the glove powder hypothesis.21,24 As with research on health care worker sensitization rates, there is currently no conclusive evidence regarding the hypothesis that NRL glove powder contributes to increased sensitization among health care employees relative to workers in other occupations who are less exposed to airborne NRL allergens.
Although studies have explored sensitization rates at varying exposure levels, there has been little comparative research on reported reactions to NRL. In the research project described here, we used self-report data on conditions commonly associated with NRL reactions to estimate relative risks among health care workers and workers in other occupations. Using symptoms as a means of assessing reactivity is a viable method for estimating occupational disease rates over time, and it has been successfully employed to analyze workers’ compensation claims potentially stemming from NRL reactions.25–27
The data included in the present study covered the periods both before and after adoption of universal precautions; thus, we were able to assess over time the effect of increased glove use on the relative risks of potential NRL-related conditions for health care workers as compared with those involved in other occupations. If reactions from occupational exposures to NRL are widespread, then our results should show not only significantly greater overall prevalence rates of symptoms commonly associated with NRL reactions among health care workers than among other workers but also a rise over time in ailments commonly associated with NRL among health care workers after the implementation of universal precautions.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONReferences |
|---|
The participation rate for the NHIS is generally between 95% and 98%,28 which is higher than rates in many of the past research endeavors on NRL reactions involving the use of survey methods.29–35 NHIS data have been used to investigate, for example, differences in occupational injuries between Black and White employee populations, differences in occupational injury rates between disabled and nondisabled workers, and prevalence rates of chronic health conditions including respiratory ailments, according to various population demographic characteristics.36–38
The sample included all individuals older than 18 years who were employed at the time of the survey. Those who were employed in the health care industry but were not working in medical occupations were excluded from the analysis. Thus, comparisons involved individuals who worked in medical occupations and had regular exposure to NRL gloves and those who worked in nonmedical occupations outside the health care industry. Of the 182 482 respondents, 6181 were employed in a medical occupation. Individuals were classified as being involved in a medical occupation if they were physicians, dentists, registered nurses, clinical laboratory technologists or technicians, licensed practical nurses, health technologists or technicians (not elsewhere classified), or nursing aides, orderlies, or attendants. The NHIS did not provide detailed occupational data after 1994.
Because the NHIS does not specifically assess causative factors, we identified potential NRL reactions by symptoms consistent with the research literature on NRL.39–42 Irritant contact dermatitis, which is not a true allergic reaction, is the most prevalent condition. This reaction takes the form of dry, itchy, or irritated areas on the skin, usually localized on the hands. Type IV allergic reactions are the most common allergic response, and they are characterized by swelling and itching on the skin surface, possible cracking of the fingertips, thickening of the dermal surface, eczema, and open lesions. Type I allergic reactions, the least common type of reaction to NRL, are IgE mediated. Symptoms of type I reactions include rhinitis, urticaria (hives), and asthma.
Asthmatic and respiratory conditions have been proposed in particular as potential responses to inhalation of airborne NRL allergens bound to the powder used as a donning agent. The NHIS categorization of conditions, which involves codes derived from the International Classification of Diseases, Ninth Revision (ICD-9), allows not only a generic look at overall conditions that may be associated with NRL but also a breakdown of conditions into dermatologic and respiratory categories. A list of the ICD-9 codes used to define potential NRL-related claims is available from the authors on request.
Each year the NHIS sample is divided into 6 subsamples. Members of each subsample are asked to provide relevant information on conditions corresponding to 1 of 6 chronic disease checklists. Conditions are reported in response to a checklist focusing on each body system. Allergic rhinitis and asthma appear on the same checklist, and dermal conditions appear on a separate checklist. We restricted our analyses to respondents who completed a symptom checklist for the category consistent with latex allergic reaction. Two sets of logistic regression analyses were estimated: one in which the dependent variable indicated whether the respondent had a potential NRL-related respiratory condition and one in which the dependent variable indicated whether the respondent had a potential NRL-related skin condition.
Because the dependent variable was dichotomous, we conducted statistical analyses by logistic regression.43 Results are presented as odds ratios comparing the conditional probability of an individual in the medical group reporting symptoms and the conditional probability of an individual in the nonmedical group reporting symptoms. If these 2 groups had the same likelihood of reporting symptoms, the odds ratio would be 1. Stata (release 7) software (Stata Corp., College Station, Tx) was used in calculating estimates. Because the survey involved a multistage sampling design, standard errors were adjusted for the design, and sample weights were used in maximum-likelihood estimations via the Stata svylogit command.
To test whether the relationship between employment in a medical vs nonmedical occupation and reports of potential NRL-related health conditions had changed since the issuance of universal precautions, we computed additional logistic regression estimates that allowed odds ratios for NRL-related health conditions among those involved in medical occupations to differ before and after these precautions had been adopted. These analyses allowed determination of whether increased NRL glove use resulted in an increase in potential NRL-related health conditions among those employed in medical occupations. Moreover, to test for possible changes over time in odds ratios, we estimated models that allowed odds ratios to differ for each calendar year.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONReferences |
|---|
presents sample means for male and female respondents broken down by medical and nonmedical occupations. As can be seen, among both men and women, those employed in medical occupations had completed more years of education than those employed in nonmedical occupations. There were also significant differences in marital status and veteran status between those employed in medical and nonmedical occupations. These potential confounds were included in the statistical models.
|
Table 2 presents estimated odds ratios for those in medical occupations as compared with those in the general working population. The first 2 rows show the estimates obtained when no other controls were added to the logistic regression analysis. The estimated odds ratios for both men and women were above 1, but in neither case were they significantly greater than 1. Age-adjusted odds ratio estimates are also presented; these estimates were similar to those in the unadjusted odds ratio estimates. Education-adjusted odds ratio estimates were considerably lower, reflecting the positive relationship between education level and the probability of experiencing dermal conditions. The final row of the first panel of Table 2 shows that inclusion of other potential confounders lowered the odds ratio estimates to less than 1, although these results were not significant.
|
shows the results for respiratory conditions. Again, the estimated odds ratios for both men and women were above 1; the estimated odds ratio for women was significantly greater than 1. Age-adjusted odds ratio estimates were similar to the unadjusted estimates. Education-adjusted odds ratio estimates were considerably lower than in the education-adjusted dermal conditions. Adjusting for education eliminated the positive significance of the odds ratios among women. Inclusion of other potential confounders lowered the odds ratio estimates further, particularly among men, although neither estimated odds ratio was significantly different from 1.
As noted previously, after the CDC’s issuance of universal precautions in 1987, NRL glove use in medical occupations increased tenfold. If NRL glove usage were a primary contributor to prevalence rates of potentially NRL-related health conditions among those employed in medical occupations, we would expect to observe increases in the odds ratios for such conditions among these individuals (relative to those employed in nonmedical occupations) after the adoption of universal precautions. However, these increases did not occur, as demonstrated by the estimates presented in Table 3 of changes in log odds ratios after the issuance of universal precautions.
|
show that the estimated percentage changes in odds ratios for dermal conditions among men were negative, whereas the changes in odds ratios among women were positive; in neither case, however, were the estimated changes significantly different from zero. Confidence intervals for dermal conditions were large, reflecting the low baseline reporting rates for these conditions. Adjustment for age, education, and other potential confounders did little to change these estimates.
The bottom panel of Table 3 shows the results for respiratory conditions. Here the estimated odds ratios for both men and women were negative but not statistically distinguishable from zero. Adjustment for age, education, and other potential confounders did little to change these estimates. The estimates indicate that, regardless of whether adjustments were made for potential confounding, the odds ratios for respiratory conditions decreased among men and among women after issuance of the universal precautions, although in no cases were the percentage change estimates statistically different from zero.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONReferences |
|---|
Although not reported in the tables, there was some indication that dermal and respiratory conditions increased over time for the sample as a whole. However, this increase in prevalence was roughly equal among those employed in medical occupations and those employed in other occupations. Thus, even though NRL glove use significantly increased during the late 1980s and 1990s in medical settings, we found no evidence to support the claim that the increase in potential NRLrelated health conditions among those employed in medical occupations was significantly different from the increase among those employed in other occupations. This finding is important, because it does not appear that increased use of NRL gloves has had a widespread adverse impact on the health of medical professionals.
Particularly noteworthy was the absence of significant changes in prevalence rates of respiratory conditions among health care workers relative to the general population after the issuance of universal precautions. Given the large increase in glove use over the period examined, this finding does not support the hypothesis that airborne glove powder contributed significantly to NRL reactions in health care settings. Of the confounders, education had the strongest relationship with both dermal and respiratory conditions. The reasons for such a relationship are not clear, but our findings suggest an important direction for future research.
One limitation of our findings stems from the inherent problems of using self-report data in medical research. Use of such data may have resulted in our estimates of potential NRL-related conditions being somewhat less specific than those obtained in previous studies involving serological or in vivo testing for allergic reactions. However, unlike the case in previous research, our focus was on reactions to NRL rather than simply on whether an individual was sensitized. Equivalent comparisons cannot be made between our results and those of earlier studies.
A second limitation was the possibility that our use of symptoms as an indication of potential NRL reactions in the absence of specific tests for NRL allergies may have led to misclassification of our dependent variables. It is unlikely that the majority of reactions in the dermal and respiratory categories measured among workers in nonmedical occupations were related to NRL. Likewise, we believe that many of the dermal and respiratory reactions observed among those working in medical occupations were probably due to factors other than NRL exposure.
Given the tenfold increase in NRL glove use among workers in medical occupations over this period and the consequent increase in exposure rates, if there were a substantial relationship between NRL exposure and reactivity, the change in symptom prevalence over time should have been statistically detectable given the large sample size. According to our estimates (made with Power and Precision software44), our methodology would have been able to detect a true odds ratio of 1.4 for respiratory conditions with a statistical power rate of 100% for women and 98% for men.
Because the baseline frequency of dermal conditions was somewhat lower than for the respiratory conditions, our statistical power was likewise lower for these estimates. In the case of dermal conditions, we would have been able to detect a true odds ratio of 1.3 with 97% power among women and a true odds ratio of 1.5 with 75% power among men. Therefore, as a result of this sensitivity, we would have been able to detect increases in these conditions owing to NRL exposure if the conditions had represented more than a minor proportion of the dermal or asthmatic conditions experienced.
A third limitation is that we analyzed only whether an individual reported a potential NRL-related health condition; we did not analyze the level of severity or the medical costs associated with such conditions. Future research might explore whether there are any differences in condition costs and severity between individuals employed in medical occupations and individuals employed in nonmedical occupations.
A fourth limitation is that those employed in medical occupations who suffered from NRL-related health conditions may have dropped out of the labor market or entered an occupation not requiring contact with NRL (i.e., "the healthy worker effect"). If this indeed occurred, a sample restricted to those employed at the time of the survey may have biased the estimated odds ratio downward. Unfortunately, the NHIS does not contain information on respondents’ previous occupations, and thus we were not able to investigate this hypothesis. Thus, an examination of the validity of this explanation is left for future research that includes information on individuals’ previous as well as current employment situations.
Overall, we found no statistical evidence that increased glove use in the period after the issuance of universal precautions by the CDC resulted in a significant increase in potential NRL-related conditions among those employed in medical occupations relative to those employed in nonmedical occupations. Moreover, the odds ratios for potential NRL-related health conditions among those involved in medical occupations were no higher across the entire study period. These results emphasize the need for future investigations examining the relationship between actual reactivity rates and NRL exposure, as opposed to studies exclusively examining sensitization rates.
|
Acknowledgments |
|---|
We are grateful for the input, comments, and critique of Richard D. Arvey and Richard J. Butler, from the University of Minnesota’s Industrial Relations Center.
Human Participant Protection
The National Health Interview Survey data used in this study were collected by the National Center for Health Statistics. Because the study used preexisting public data, no ethical clearance was necessary.
|
Footnotes |
|---|
Accepted for publication July 4, 2002.
|
References |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONReferences |
|---|
2. Kellet PB. Latex allergy: a review. J Emerg Nurs.1997;23:27–36.[Medline]
3. Jackson D. Latex allergy and anaphylaxis—what to do. J Intravenous Nurs.1995;18:33–59.
4. Evangelisto M. Latex allergy: the downside of standard precautions. Today’s Surgical Nurse.1997;19:28–33.
5. Melton AL. Managing latex allergy in patients and healthcare workers. Cleve Clin J Med.1997;64:76–82.[ISI][Medline]
6. Dakin MJ, Yentis SM. Latex allergy: a strategy for management. Anaesthesia.1998;53:774–781.[ISI][Medline]
7. Kim KT, Wellmeyer EK, Miller KV. Minimum prevalence of latex hypersensitivity in health care workers. Allergy Asthma Proc.1999;20:387–391.[ISI][Medline]
8. Yassin MS, Lierl MB, Fischer TJ, O’Brien K, Cross J, Steinmetz C. Latex allergy in hospital employees. Ann Allergy.1994;72:245–249.[ISI][Medline]
9. Turjanmaa KA. Incidence of immediate allergy to latex gloves in hospital personnel. Contact Dermatitis.1987;17:270–275.[ISI][Medline]
10. Arellano R, Bradley J, Sussman G. Prevalence of latex sensitization among hospital physicians exposed to latex gloves. Anesthesiology.1992;77:905–908.[ISI][Medline]
11. Wrangsjo K, Osterman K, van Hage-Hamsten M. Glove-related skin symptoms among operating theatre and dental care unit personnel. Contact Dermatitis.1994;30:139–143.[ISI][Medline]
12. Konrad C, Fieber T, Gerber H, Schuepfer G, Muellner G. The prevalence of latex sensitivity among anesthesiology staff. Anesth Analg.1997;84:629–633.[Abstract]
13. Ownby DR, Ownby HE, McCullough J, Shafer AW. The prevalence of anti-latex IgE antibodies in 1000 volunteer blood donors. J Allergy Clin Immunol.1996;97:1188–1192.[ISI][Medline]
14. Lebenbom-Mansour MH, Oesterle JR, Ownby DR, Jennett MK, Post SK, Zaglaniczy K. The incidence of latex sensitivity in ambulatory surgical patients: correlation of historical factors with positive serum immunoglobin E levels. Anesth Analg.1997;85:44–49.[Abstract]
15. Porri F, Lemiere C, Birnbaum J, et al. Prevalence of latex sensitization in subjects attending health screening: implications for perioperative screening. Clin Exp Allergy.1997;27:413–417.[ISI][Medline]
16. Saxon A, Ownby D, Huard T, Parsad R, Roth DH. Prevalence of IgE to natural rubber latex in unselected blood donors and performance characteristics of AlaSTAT testing. Ann Allergy Asthma Immunol.2000;84:200–206.
17. Vanderplas O, Delwiche J-P, Evrard G, et al. Prevalence of occupational asthma due to latex among hospital personnel. Am J Respir Crit Care Med.1995;151:54–60.[Abstract]
18. Grzybowski M, Ownby DR, Peyser PA, Johnson CC, Schork MA. The prevalence of anti-latex IgE antibodies among registered nurses. J Clin Immunol.1996;98:535–544.
19. Kibby T, Akl M. Prevalence of latex sensitization in a hospital employee population. Ann Allergy Asthma Immunol.1997;78:41–44.[ISI][Medline]
20. Mace SR, Sussman GL, Liss G, et al. Latex allergy in operating room nurses. Ann Allergy Asthma Immunol.1998;80:252–256.[ISI][Medline]
21. Page EH, Esswein EJ. NIOSH Health Hazard Evaluation Report: St. Joseph Exempla Hospital, Denver, Colorado. Atlanta, Ga: Centers for Disease Control and Prevention; 1999.
22. Brehler R, Kolling R, Wastell C. Glove powder—a risk factor for the development of latex allergy? Eur J Surg Suppl.1997;579:23–25.
23. Baur X, Chen, Z, Henning A. Can a threshold limit value for natural rubber latex airborne allergens be defined? J Allergy Clin Immunol.1998;101:24–27.[ISI][Medline]
24. Sussman GL, Liss GM, Deal K, et al. Incidence of latex sensitization among latex glove users. J Allergy Clin Immunol.1998;101:171–178.[ISI][Medline]
25. Horwitz IB, Arvey RD. Workers’ compensation claims from latex glove use: a longitudinal analysis of Minnesota 1988–1997. J Occup Environ Med.2000;42:932–938.[ISI][Medline]
26. Horwitz IB, Kammeyer-Mueller JD, Butler RJ. Workers’ compensation claims as a measure of healthcare worker reaction to latex gloves in healthcare settings: Rhode Island 1992–1997. J Workers Compensation.2000;10:49–59.
27. Horwitz IB, Kammeyer-Mueller JD, McCall BP. Assessing latex allergy among health care employees using workers’ compensation data: results from North Dakota and a comparison with findings in Minnesota and Rhode Island. Minn Med.2001;84:47–50.
28. Collins JG. Prevalence of selected chronic conditions: United States, 1990–1992. Vital Health Stat 10. 1997;No. 194.
29. National Center for Health Statistics. Current estimates from the National Health Interview Survey. Vital Health Stat 10. 1988;No. 173.
30. Tarlo SM, Sussman GL, Holness DL. Latex sensitivity in dental students and staff: a cross-sectional study. J Allergy Clin Immunol.1997;99:396–401.[ISI][Medline]
31. Kujala VM, Reijula KE. Glove-induced dermal and respiratory symptoms among health care workers in one Finnish hospital. Am J Ind Med.1995;28:89–98.[ISI][Medline]
32. Mizutari K, Kuriya N, Tomomichi O. Immediate allergy to rubber gloves: a questionnaire study of hospital personnel. J Dermatol.1995;22:13–19.
33. Tarlo SM, Wong L, Roos J, Booth N. Occupational asthma caused by latex in a surgical glove manufacturing plant. J Allergy Clin Immunol.1990;85:626–631.[ISI][Medline]
34. Sinha A, Harrison PV. Latex glove allergy among hospital employees: a study in the north-west of England. Occup Med.1998;48:405–410.
35. Liss G, Sussman GL, Deal K, et al. Latex allergy: epidemiological study of 1351 hospital workers. Occup Environ Med.1997;54:335–342.[Abstract]
36. Kujala VM, Karvonen J, Laara E, Kanerva L, Estlander T, Reijula KE. Postal questionnaire with latex allergy among health care workers in Finland. Am J Ind Med.1997;32:197–204.[ISI][Medline]
37. Wagener DK, Winn DW. Injuries in the workplace: black-white differences. Am J Public Health.1991;81:1408–1414.
38. Zwerling C, Whitten PS, Davis CS, Sprince NL. Occupational injuries among workers with disabilities: the National Health Interview Survey, 1985–1994. JAMA.1997;278:2163–2166.[Abstract]
39. Coombs RRA, Gell PGH. The classification of allergic reaction underlying disease. In: Gell PBH, Coombs RRA, eds. Clinical Aspects of Immunology. Oxford, England: Blackwell Scientific Publications; 1963:317.
40. Shoup A. Risk factors related to the use of latex gloves. QRC Advisor.1995;11:1–6.
41. Gliniecki CM. Management of latex reactions in the occupational setting. Am Assoc Occup Health Nurses J.1998;46:82–93.
42. Hamann CP. Natural rubber latex protein sensitivity in review. Am J Contact Dermatitis.1993;4:4–21.
43. Rothman KJ, Greenland S. Modern Epidemiology. Philadelphia, Pa: Lippincott-Raven; 1998.
44. Borenstein M, Rothstein H, Cohen J, Schoenfeld R, Berlin J. Power and Precision [computer software]. Englewood, NJ: Biostat Inc; 2001.
This article has been cited by other articles:
|
|
 |
|
  Latex-Induced Allergic Reactions: Are Healthcare Workers at Greater Risk? Journal Watch Emergency Medicine, May 14, 2003; 2003(514): 8 - 8. [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
