HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:
Guest Access | Sign In via User Name/Password

This Article Abstract Full Text (PDF) Free Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Article Archive Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (68) Citing Articles via Google Scholar Google Scholar Articles by Chapman, K. R. Articles by Pye, D. J. Search for Related Content PubMed PubMed Citation Articles by Chapman, K. R. Articles by Pye, D. J.

Gender Bias in the Diagnosis of COPD^*

^* From the Division of Respiratory Medicine (Dr. Chapman), Department of Medicine, University of Toronto, Toronto, Ontario, Canada; UCLA School of Medicine (Dr. Tashkin), Los Angeles, CA; and Westmount Research Consultants (Dr. Pye), Toronto, Ontario, Canada.

Correspondence to: Kenneth R. Chapman, MD, FCCP, Asthma Center of the University Health Network, Suite 4–011 ECW, 399 Bathurst St, Toronto, Ontario M5T 2S8, Canada; e-mail: kchapman{at}inforamp.net

	Abstract

TOP Abstract Introduction Materials and Methods Results Discussion References

 
Background: COPD is thought to be more prevalent among men than women, a finding usually attributed to higher smoking rates and more frequent occupational exposures of significance for men. However, smoking prevalence has increased among women and there is evidence that women may be more susceptible to the adverse pulmonary function effects of smoking than men. There may also be underdiagnosis and misdiagnosis of COPD in both sexes because objective measures of lung function are underused.

Objectives: We undertook the present study to determine if there is gender bias in the diagnosis of COPD, such that women are less likely than men to receive a diagnosis of COPD. We also attempted to determine if underuse of lung function measurements was a factor in any bias detected.

Methods: We surveyed a random sample of 192 primary-care physicians (96 American and 96 Canadian; 154 men and 38 women) using a hypothetical case presentation and a structured interview. The case of cough and dyspnea in a smoker was presented in six versions differing only in the age and sex of the patient. After presentation of the history and physical findings, physicians were asked to state the most probable diagnosis and to choose the diagnostic studies needed. Physicians were then presented with spirometric findings of moderate or severe obstruction without significant bronchodilator response, and the questions repeated. Finally, the negative outcome of an oral steroid trial was described.

Results: Initially, COPD was given as the most probable diagnosis significantly more often for men than women (58% vs 42%; p < 0.05). The likelihood of a COPD diagnosis increased significantly and initial differences between sexes decreased as objective information was provided. After spirometry, COPD diagnosis rates for men and women were 74% vs 66% (p = not significant); after the steroid trial 85% vs 79% (p = not significant). Only 22% of physicians would have requested spirometry after the initial presentation.

Conclusions: In North America, primary-care physicians underdiagnose COPD, particularly in women. Spirometry reduces the risk of underdiagnosis and gender bias but is underused.

Key Words: asthma • misdiagnosis • physician decision making • spirometry • underdiagnosis

	Introduction

TOP Abstract Introduction Materials and Methods Results Discussion References

 
COPD is the fourth most common cause of death in North America and remains the only common cause of mortality that is increasing in prevalence. Despite an overall decline in the prevalence of cigarette smoking since the pivotal report of the Surgeon General in the mid-1960s, we continue to witness the late pulmonary sequelae of tobacco consumption. Diagnosed COPD appears to be more prevalent in men than women, whether expressed in terms of mortality or hospital separation data. This is usually attributed to the historically higher rates of cigarette smoking among men and their greater likelihood of exposure to significant occupational respiratory irritants. However, these risk factors have shifted dramatically in the past several decades. Cigarette smoking is now as common or more common among women as compared to men, and more women than before now work outside the home in occupations once held almost exclusively by men. There is also growing evidence that women may be more susceptible to the adverse pulmonary consequences of tobacco smoking. In the Lung Health Study¹ of mild COPD, the prevalence of a positive methacholine challenge finding was 25% among men but 48% among women. In two Danish population studies² of 13,897 subjects followed for between 7 and 16 years, women who smoked had a higher rate of lung function decline per amount smoked than men and were at greater risk of being hospitalized for the treatment of COPD.

Although women are clearly at increasing risk of developing COPD, the diagnosis continues to be made much more commonly in men. By contrast, in studies³ of emergency department and ambulatory clinic care of asthma, women present with doctor-diagnosed asthma more often than men. Such findings could be explained by a gender bias among physicians in their diagnosis of different types of respiratory diseases. This has been well documented for other tobacco-related illnesses such as coronary artery disease.

It is plausible that misdiagnosis, including possible gender bias, could be reduced by the appropriate use of objective laboratory studies. Unfortunately, previous studies⁴ have shown that physicians make little use of spirometry in their investigation of chronic respiratory symptoms. We therefore undertook the following study to determine if North American physicians reached similar provisional diagnoses for hypothetical male and female patients presenting with identical chronic respiratory symptoms and identical smoking histories. We also sought to examine test-ordering behavior by physicians presented with these clinical scenarios, as well as the impact on their diagnoses when objective pulmonary function data were provided to them.

	Materials and Methods

TOP Abstract Introduction Materials and Methods Results Discussion References

 
Study Design
We devised one hypothetical case summary to present to primary-care physicians for their review and management. The summary described a middle-aged former smoker (of approximately 40 pack-years consumption) who presented with morning cough persisting for 4 years and worsening intermittently following viral respiratory tract infections. For 3 years, the patient had been troubled by breathlessness on moderate exertion. On physical examination, the only abnormal finding was the presence of expiratory wheezes throughout the chest. In randomized fashion, half the physicians were told that the patient was a woman, and half were told that the patient was a man. Also in random fashion, the age of the patient was given as 47 years or 59 years in balanced fashion between the sexes. Using a structured questionnaire, trained interviewers who were blinded to the major objectives of the study asked physicians to state the most likely diagnosis and the diagnostic studies needed. Independent of the provisional diagnosis offered or the tests requested, all physicians were provided with a set of laboratory test results for the hypothetical patient. These comprised normal "routine" blood analyses (hemoglobin, WBC count, sedimentation rate, electrolytes, liver enzymes, BUN, and creatinine), a normal ECG finding, and a normal chest radiograph finding. In addition, an abnormal spirometric result was reported. In balanced randomized fashion, physicians were shown one of two sets of numerical spirometric data (FVC, FEV₁, and FEV₁/FVC) accompanied by interpretations characterizing the results as a "moderate" or "severe" obstructive defect. These corresponded with values of FEV₁ of 52% or 32% of predicted, respectively, and with values of FEV₁/FVC of 43% and 27%, respectively. Half the respondents also received postbronchodilator numerical results characterized as showing "minimal" change. Once again, physicians were asked to offer their most probable diagnosis. After this was given, physicians were provided with negative oral steroid trial results for the hypothetical patient. That is, the patient was described as using prednisone, 40 mg/d, for 2 weeks with no significant change in FEV₁. After this information had been given, physicians were asked for the last time to provide a diagnosis.

The diagnostic and test ordering terminology used by physicians was categorized by an a priori scheme. Terms not accounted for by the a priori scheme were categorized by two of the investigators blinded to the gender, age, and degree of obstruction that had elicited the response.

Physicians Surveyed
Before the survey instrument was used for the main physician sample, we tested its plausibility, clarity, and consistency on a convenience sample of 21 primary-care and respiratory specialist physicians from two Canadian cities and two US cities. With minor changes in terminology, the instrument was then administered to a sample of 96 US primary-care physicians and 96 Canadian primary-care physicians selected in random fashion from two commercial databases of practicing physicians. In each country, physician sampling was from seven geographic regions ranging from coast to coast.

Data Analysis
An analysis of variance model was used to assess the influence of patient gender and patient age on the frequency of COPD diagnoses offered by physicians following presentation of the history and physical examination results. A repeated-measures analysis of variance model was used to assess the influence of these and additional factors on the frequency of COPD diagnoses over the three stages of the interview as additional data were offered. These additional factors were severity of obstruction, availability of bronchodilator test results, and nationality of the responding physician. Proportions of tests ordered by US and Canadian physicians were compared by ² test. Results were considered significant at the p < 0.05 level.

	Results

TOP Abstract Introduction Materials and Methods Results Discussion References

 
Physicians Surveyed
We surveyed a random sample of 192 primary-care physicians (96 American and 96 Canadian; 154 men and 38 women). In the United States, interviews were conducted in the following geographic areas (major urban center and surrounding nonurban region): Chicago, San Francisco, Seattle, Houston, Denver, Long Island, and Atlanta. In Canada, interviews were conducted in the following geographic areas (major urban center and surrounding nonurban region): Toronto, Montreal, Vancouver, Calgary, Winnipeg, London, and Quebec City. For each physician who participated in the survey, three who were contacted did not. The most common reason given for nonparticipation was inability to schedule an appointment with the interviewer during the few days available for survey participation in each city. Rates of participation were similar between Canada and the United States.

Diagnoses
Only 57% of physicians offered COPD as the most likely diagnosis following presentation of the history and physical examination. As shown in Table 1 , COPD was significantly more likely to be offered as the provisional diagnosis for the hypothetical male patient than the hypothetical female patient (64.6% vs 49.0%; p < 0.05). Patient age was not significantly related to the diagnosis offered. The likelihood of a provisional COPD diagnosis increased significantly (to 70%) after presentation of abnormal spirometric findings (p < 0.001) and the disparity between male and female patients decreased. There was a further increase in the number of COPD diagnoses offered by physicians after presentation of the negative oral steroid trial results (p < 0.001). The disparity in diagnosis between hypothetical male and female patients decreased further. The most common alternative diagnosis was asthma, a provisional diagnosis offered 35% of the time after the initial presentation, 27% of the time after abnormal spirometric results were offered, and 14% of the time after a negative oral steroid trial was described. Although the availability of spirometric data caused a significant increase in the frequency of COPD diagnoses for both sexes, we could detect no effect on diagnoses offered by the severity of obstruction or the availability of negative bronchodilator response data.

Diagnostic Studies Requested
After the initial presentation of the hypothetical patient, the most commonly requested study was a chest radiograph; this was requested by approximately 80% of both American and Canadian physicians (Table 2 ). Other diagnostic studies were requested far less frequently as listed in Table 2 . In particular, only 21.8% of physicians requested spirometry, a test requested slightly more frequently by American physicians than Canadian physicians (p < 0.05). American physicians also ordered blood analyses, ECGs, and arterial blood gas measurements more frequently than their Canadian counterparts. A small number of requests for methacholine challenge tests came only from Canadian practitioners. There were no significant differences between men and women physicians in their test-ordering behavior.

	Discussion

TOP Abstract Introduction Materials and Methods Results Discussion References

Ours is not the first study to document a physician bias toward diagnosing tobacco-related diseases more readily in men as compared to women. Women who suffer chest pain are less likely than men to be referred for invasive testing such as angiography, implying a bias against the diagnosis of coronary artery disease in women or a bias toward the diagnosis in men.⁵ Various explanations have been offered for the differences in diagnostic rates between men and women in certain disease settings. For example, it has been suggested that men and women may perceive or report some symptoms differently.⁶ It is also been noted that men and women use the health-care system differently.⁷ Women use primary-care services more frequently than men and receive more health-care services even when reproductive-related care is excluded.⁸ Curiously, women are less likely to receive specialist referrals than men.⁸ However, our study does not support these hypotheses, given that gender bias was present in the absence of specific patient behavior or perceptions. The details presented in our hypothetical scenario were identical for male and female patients. Thus, our findings reflect a preexisting physician bias that is not related to differences in patient personality, communication skills, or expressed preferences.

Some might argue that such a bias in a tobacco-related disease is appropriate. That is, they would argue that the risk of COPD is truly higher in men then in women given the historically higher tobacco consumption rates in men. However, we believe that this argument is flawed in several ways. First, the argument is a tautology. That is, epidemiologic data comparing the prevalence of various diseases by sex and age are dependent on accurate physician diagnosis. There is no single and clearly defined laboratory test for COPD or asthma, and currently published diagnostic criteria are confusing and vague with overlapping features between the syndromes.^{9 10} Thus, to argue that men have COPD more commonly than women is indeed circular given the findings of the present study. Second, a slightly higher prevalence of a given disease is irrelevant in the diagnostic approach to individual patients. Even if COPD were somewhat more common in men than women, a physician would be unwise to dismiss the possibility of COPD in any woman suffering from exertional breathlessness and wheezing against a background of tobacco consumption.

Our study confirms our earlier report that spirometry is underused by primary-care physicians in North America.⁴ We have previously surveyed primary-care physicians in Canada, presenting them with a hypothetical scenario similar to the one used in the present study. We discovered that only 5% of Canadian physicians would request spirometry when presented with a middle-aged smoker who suffered from recurrent productive cough over 2 consecutive years and presented with expiratory wheezes on physical examination. In our earlier study, physicians increased their requests for spirometry if they were presented with the diagnostic term chronic bronchitis as part of the medical history. However, even with this prompting, the rate of requesting spirometry was < 40%. We note that spirometry and objective measures of lung function are also poorly utilized in the assessment of asthma. We have previously described the clinical characteristics of patients referred for methacholine challenge testing and found to have negative studies.¹¹ The majority of patients had received a diagnosis of asthma by their primary-care physicians and had endured treatment with an average of two or more antiasthma medications for > 2 years. Thus, the provisional diagnosis of airways disease is anything but temporary. It may sometimes lead to prolonged periods of inappropriate therapy.

Our study also reveals that there is considerable diagnostic confusion between COPD and asthma, the most common alternative diagnosis offered by physicians. We find it significant that the initial ratio of men (58%) to women (42%) accorded the diagnosis of COPD is almost the mirror image of the gender ratio for asthma in a variety of studies of emergency or ambulatory care.^{3 12} Some might argue that the phenomenon of diagnostic exchange between the sexes is trivial and might note that in our study the diagnosis of an obstructive lung disease (regardless of subtype) was similar between the sexes. However, this diagnostic imprecision ignores the marked and diverging differences in prognosis and therapy between asthma and COPD. Of particular concern to postmenopausal women incorrectly labeled as having asthma would be the increased likelihood of corticosteroid treatment. Several studies^{13 14} now report an increased likelihood of decreased bone density as a dose-related consequence of inhaled corticosteroid exposure.

It is disturbing to note that some physicians in our survey were remarkably reluctant to use the diagnostic term COPD or equivalent. Even after the description of persistent and unchanging obstruction following 2 weeks of oral steroid therapy, one in seven physicians would continue to use the diagnostic label asthma. This could help to explain our reported findings from other surveys of respiratory disease management. We have found that physicians often prescribe similarly for asthma and COPD patients despite distinguishing in a theoretical sense between the illnesses.^{4 15} It is also possible that some physicians have chosen not to use the diagnostic terms COPD or emphysema, as the former has little meaning for patients and the latter may be alarming to patients. They may prefer the term asthma themselves, regarding it as a disease more amenable to therapy and more rewarding to treat than COPD. We have no explanation for the responses of 4% of the physicians who continued to offer nonrespiratory diagnoses as the most likely explanation for dyspnea after presentation spirometric data showed moderate or severe airflow limitation unresponsive to treatment with bronchodilators or steroids.

Some limitations to our study must be noted. First, we did not test the actual practice behavior of physicians we surveyed, behavior that might differ somewhat from the responses recorded following the presentation of written materials. Nevertheless, our findings are consistent with published epidemiologic and clinical trial data and we believe they are a reasonable reflection of actual practice. Second, we did not attempt to determine the reasons for the provisional diagnoses offered; therefore, we have no explanation for the underdiagnosis, gender bias, and lack of pulmonary function testing seen. Further research is needed to determine the factors responsible for the behaviors observed if appropriate corrective steps such as physician education are to be taken. Third, we cannot identify which types of physician practice or training backgrounds are associated with a neglect of objective spirometric testing or with gender bias in the diagnosis of obstructive lung disease. We sought by our screening process to identify self-described "family doctors" in active practice and read a descriptive statement to each potential participant. It is possible that a small number of primary-care internists participated in the study in the United States but too few to allow meaningful comparison to the remaining primary-care physicians.

	Footnotes

 
The authors received a research grant-in-aid from Boehringer Ingelheim Canada Ltd and Boehringer Ingelheim Pharmaceuticals Inc.

Received for publication April 12, 2000. Accepted for publication January 3, 2001.

	References

TOP Abstract Introduction Materials and Methods Results Discussion References

 

1. Kanner, RE, Connett, JE, Altose, MD, et al (1994) Gender difference in airway hyperresponsiveness in smokers with mild COPD: the Lung Health Study. Am J Respir Crit Care Med 150,956-961[Abstract]
2. Prescott, E, Bjerg, AM, Andersen, PK, et al (1997) Gender difference in smoking effects on lung function and risk of hospitalization for COPD: results from a Danish longitudinal study. Eur Respir J 10,822-827[Abstract]
3. Rea, HH, Garrett, JE, Mulder, J, et al (1991) Emergency room care of asthmatics: a comparison between Auckland and Toronto. Ann Allergy 66,48-52[ISI][Medline]
4. Kesten, S, Chapman, KR (1993) Physician perceptions and management of COPD. Chest 104,254-258[Abstract/Free Full Text]
5. Ayanian, JZ, Epstein, AM (1991) Differences in the use of procedures between women and men hospitalized for coronary heart disease. N Engl J Med 325,221-225[Abstract]
6. Kühni, CE, Sennhauser, FH (1995) The Yentl syndrome in childhood asthma: risk factors for undertreatment in Swiss children. Pediatr Pulmonol 19,156-160[ISI][Medline]
7. Hall, JA, Roter, DL, Katz, NR (1988) Meta-analysis of correlates of provider behavior in medical encounters. Med Care 26,657-675[ISI][Medline]
8. Franks, P, Clancy, CM, Naumburg, EH (1995) Sex, access and excess. Ann Intern Med 123,548-550[Free Full Text]
9. Celli, BR, Snider, GL, Heffner, J, et al (1995) Standards for the diagnosis and care of patients with chronic obstructive pulmonary disease. Am J Respir Crit Care Med 152,S77-S121
10. Sheffer AL, Bousquet J, Busse WW, et al. International consensus report on diagnosis and management of asthma. Bethesda, MD: US Department of Health and Human Services, 1992; publication No. 92–3091
11. Joyce, DP, Chapman, KR, Kesten, S (1996) Prior diagnosis and treatment of patients with normal results of methacholine challenge and unexplained respiratory symptoms. Chest 109,697-701[Abstract/Free Full Text]
12. Jackevicius, C, Joyce, DP, Kesten, S, et al (1997) Prehospitalization inhaled corticosteroid use in patients with COPD or asthma. Chest 111,296-302[Abstract/Free Full Text]
13. Hanania, NA, Chapman, KR, Sturtridge, WC, et al (1995) Dose-related decrease in bone density among asthmatic patients treated with inhaled corticosteroids. J Allergy Clin Immunol 96,571-579[CrossRef][ISI][Medline]
14. Ip, M, Lam, K, Yam, L, et al (1994) Decreased bone mineral density in premenopausal asthma patients receiving long-term inhaled steroids. Chest 105,1722-1727[Abstract/Free Full Text]
15. Jackevicius, CA, Chapman, KR (1997) Prevalence of inhaled corticosteroid use among patients with chronic obstructive pulmonary disease: a survey. Ann Pharmacother 31,160-164[Abstract]

This article has been cited by other articles:

				M. K. Han, M. G. Kim, R. Mardon, P. Renner, S. Sullivan, G. B. Diette, and F. J. Martinez Spirometry Utilization for COPD: How Do We Measure Up? Chest, August 1, 2007; 132(2): 403 - 409. [Abstract] [Full Text] [PDF]

				H. H. Jensen, N. S. Godtfredsen, P. Lange, and J. Vestbo Potential misclassification of causes of death from COPD Eur. Respir. J., October 1, 2006; 28(4): 781 - 785. [Abstract] [Full Text] [PDF]

				C. Jenkins Mars and Venus in the GP's office. Eur. Respir. J., August 1, 2006; 28(2): 259 - 261. [Full Text] [PDF]

				W. M. Chatila, E. A. Hoffman, J. Gaughan, G. B. Robinswood, G. J. Criner, and for the National Emphysema Treatment Trial Researc Advanced emphysema in african-american and white patients: do differences exist? Chest, July 1, 2006; 130(1): 108 - 118. [Abstract] [Full Text] [PDF]

				J. B. Soriano, G. T. Visick, H. Muellerova, N. Payvandi, and A. L. Hansell Patterns of Comorbidities in Newly Diagnosed COPD and Asthma in Primary Care Chest, October 1, 2005; 128(4): 2099 - 2107. [Abstract] [Full Text] [PDF]

				M. A. Campos, A. Wanner, G. Zhang, and R. A. Sandhaus Trends in the Diagnosis of Symptomatic Patients With {alpha}1-Antitrypsin Deficiency Between 1968 and 2003 Chest, September 1, 2005; 128(3): 1179 - 1186. [Abstract] [Full Text] [PDF]

				J Vestbo COPD in the ECRHS Thorax, February 1, 2004; 59(2): 89 - 90. [Full Text] [PDF]

				B. R. Celli A 62-Year-Old Woman With Chronic Obstructive Pulmonary Disease JAMA, November 26, 2003; 290(20): 2721 - 2729. [Full Text] [PDF]

				J. B. Soriano, K. J. Davis, B. Coleman, G. Visick, D. Mannino, and N. B. Pride The Proportional Venn Diagram of Obstructive Lung Disease: Two Approximations From the United States and the United Kingdom Chest, August 1, 2003; 124(2): 474 - 481. [Abstract] [Full Text] [PDF]

				S.T. Weiss, D.L. DeMeo, and D.S. Postma COPD: problems in diagnosis and measurement Eur. Respir. J., June 1, 2003; 21(41_suppl): 4S - 12s. [Abstract] [Full Text] [PDF]

				R. Sinvhal, L. Tosiello, G. Cable, R. I. Harik-Khan, J. L. Fleg, and R. A. Wise Confounding Issues in COPD Risk Study? Chest, January 1, 2003; 123(1): 307 - 308. [Full Text] [PDF]

				J. M. Singh, V. A. Palda, M. B. Stanbrook, and K. R. Chapman Corticosteroid Therapy for Patients With Acute Exacerbations of Chronic Obstructive Pulmonary Disease: A Systematic Review Arch Intern Med, December 9, 2002; 162(22): 2527 - 2536. [Abstract] [Full Text] [PDF]

				R. G. Barr, J. Herbstman, F. E. Speizer, and C. A. Camargo Jr. Validation of Self-reported Chronic Obstructive Pulmonary Disease in a Cohort Study of Nurses Am. J. Epidemiol., May 15, 2002; 155(10): 965 - 971. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Free Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Article Archive Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (68) Citing Articles via Google Scholar Google Scholar Articles by Chapman, K. R. Articles by Pye, D. J. Search for Related Content PubMed PubMed Citation Articles by Chapman, K. R. Articles by Pye, D. J.