Practice Patterns of Pulmonologists and Family Physicians for Occupational Asthma

Original Research ASTHMA

Practice Patterns of Pulmonologists and Family Physicians for Occupational Asthma* D. Linn Holness, MD, FCCP; Shehrina Tabassum, MSc; Susan M. Tarlo, MBBS, FCCP; Gary M. Liss, MD; Frances Silverman, PhD; and Michael Manno, MSc

Background: The longer the duration of symptoms of occupational asthma (OA) before diagnosis, the poorer the outcome. Physicians can play a key role in the early recognition of occupational lung diseases (OLDs), including OA. Our objective was to document and compare the practice patterns, barriers, and needs for early diagnosis of OA among pulmonologists and family physicians. Methods: Based on information from the literature and interviews with pulmonologists and family physicians, a survey was developed to obtain information on practice patterns. The survey was sent to all pulmonologists and a random sample of 600 family physicians in Ontario. Results: Eight percent of pulmonologists and 7% of family physicians report seeing > 20 patients a year with OLD. The majority report taking a workplace exposure history. The most commonly stated barrier to obtaining a workplace exposure history was time constraints. Main reasons for referral to specialists for diagnosis include personal lack of expertise, testing facilities, and knowledge about workers’ compensation, while lack of timely access to specialists is a barrier for referral. While most physicians identified a need for further education, those who did not identify a need for further occupational respiratory education cited low volume of patients, access to specialists, and time constraints as reasons for not wanting further education. Conclusions: Opportunities are identified to improve health services delivery and educational initiatives for OA, with approaches tailored to each particular physician group. (CHEST 2007; 132:1526 –1531) Key words: barriers; education needs; exposure history; medical education; occupational asthma; occupational history; occupational lung disease; referral patterns Abbreviations: OA ⫽ occupational asthma; OLD ⫽ occupational lung disease

asthma (OA) is a common occupaO ccupational tional disease that often results in poor clinical and functional outcomes.1–11 There is often a long gap between the onset of symptoms and diagnosis.12–13 It has been shown that the longer the time between first symptoms and diagnosis, the poorer the outcome.13–17 The physicians who first see workers with OA include family physicians and community pulmonologists. It is known that family physicians may not take a complete work history, and inadequate knowledge and time constraints have been identified as barriers.18 –24 1526

Other than the work by Harber et al,25 we know little of the practice of pulmonologists relative to OA. The objective of this study was to document and compare the knowledge and practice patterns related to OA, the barriers to early recognition, and educational needs for family physicians and pulmonologists in Ontario. This information could then be used to design and implement strategies to improve earlier recognition, diagnosis, and management of workers with OA. We were also interested in identifying perceived knowledge gaps and educational needs for other occupational lung diseases (OLDs). Original Research

Materials and Methods The study was reviewed and approved by the Research Ethics Board of St. Michael’s Hospital, as part of a larger study26 that also examined dermatologist and family physician practice related to occupational contact dermatitis. To inform the content of the survey instrument, phase 1 consisted of interviews with pulmonologists and family physicians. The interviews were conducted by a consultant in medical education with much experience in interviewing physicians. Four pulmonologists and two family physicians were interviewed. The key themes that arose in the interviews included low volume of patients, lack of training and knowledge, time constraints, and the compensation process. Phase 2 consisted of a mailed survey to all 167 Ontario pulmonologists and a sample of 600 family physicians randomly selected from approximately 4,000 family physicians in Ontario. We used the Total Design Method, modified as follows.27 Everyone received at least two mailings. The first mailing consisted of a cover letter, survey, and postage-paid return envelope that was followed by a reminder card sent 2 weeks later. Those physicians who did not return the survey were sent a second package, and those who did not return this questionnaire were sent another reminder card about 2 weeks later. Response rates were 49% for pulmonologists and 26% for family physicians. As the main purpose of the study was descriptive, the data were analyzed using standard statistical methods including frequencies. Difference between physician groups were evaluated using Pearson ␹2 tests.

Results Demographic information is presented in Table 1. The majority of respondents practiced in large urban or metropolitan areas, were men, and were ⬍ 50 years old. Fewer than 10% of pulmonologists and family physicians reported seeing ⬎ 20 patients per year with occupational lung disease (OLD). There were no significant differences between the two groups with respect to age, gender, or year of graduation, but there was a significant difference in practice setting, with family physicians being more likely to work in rural areas. The reported frequency with which physicians take a workplace exposure history and the barriers to taking a *From Gage Occupational and Environmental Health Unit, St Michael’s Hospital, University of Toronto, Toronto, ON, Canada. This work as performed at Gage Occupational and Environmental Health Unit, St. Michael’s Hospital, University of Toronto. This project was funded by a research grant from the Ontario Workplace Safety and Insurance Board, project 02036. Drs. Holness, Tarlo, Liss, and Silverman receive research grant support from the Ontario Workplace Safety and Insurance Board. Ms. Tabassum and Mr. Manno have no conflicts of interest to disclose. Manuscript received September 7, 2006; revision accepted July 11, 2007. Reproduction of this article is prohibited without written permission from the American College of Chest Physicians (www.chestjournal. org/misc/reprints.shtml). Correspondence to: D. Linn Holness, MD, FCCP, Gage Occupational and Environmental Health Unit, St Michael’s Hospital, 30 Bond St, Toronto, ON, Canada M5B 1W8; e-mail: HOLNESSL@ smh.toronto.on.ca DOI: 10.1378/chest.06-2224 www.chestjournal.org

Table 1—Demographics of Survey Respondents* Demographics

Pulmonologists

Family Physicians

Total respondents, No. Response rate Practice location (population) Metropolitan (⬎ 500,000) Large urban (50,000–500,000) Urban (25,000–50,000) Rural (⬍ 25,000) Male gender Age ⬍ 50 yr Patients with OA seen per year ⬍5 5–20 ⬎ 20

65 49

107 26

55 38 6 0 70 64

34 29 7 30 58 68

42 51 8

48 45 7

*Data are presented as % unless otherwise indicated.

history are shown in Table 2. More pulmonologists than family physicians report taking a workplace exposure history all or most of the time (p ⬍ 0.0001). The most important barrier to taking a workplace exposure history for both groups was time constraints, but lack of

Table 2—Frequency and Barriers To Taking a History of Workplace Exposures*

Variables

Pulmonologists (n ⫽ 65)

Family Physicians (n ⫽ 107)

Take history of workplace exposures† Always/most of the time 92 57 Sometimes 6 40 Rarely/never 2 3 If you do not routinely ask about workplace exposures, why not? Time constraints 60 86 Lack of confidence in 20 26 taking history Patients usually not able to 40 47 provide Lack of knowledge 60 74 pertaining to workers’ compensation Lack of adequate 0 23 reimbursement Complicated/excessive 20 19 forms to fill in Not my job 20 2 Forget to ask 40 67 Three most important Time Time constraint; reasons for not constraint; lack of knowledge; taking a history forget to ask; forget to ask lack of confidence *Data are presented as %. †p ⬍ 0.0001 for comparison between pulmonologists and family physicians. CHEST / 132 / 5 / NOVEMBER, 2007

1527

knowledge, lack of confidence, and forgetting to ask were other main reasons cited. There were no differences in taking a workplace history related to age, gender, or year of graduation. Responses to questions regarding the investigation of OA are presented in Table 3. There was some difference between the pulmonologists and family physicians groups, with 27% of pulmonologists and 8% of family physicians reporting that they always diagnose diseases in patients themselves. Both groups noted lack of Table 3—Investigation of OA*

Variables

Pulmonologists, %

Family Physicians, %

Investigation of patients with possible OA† Always diagnose 27 8 condition myself Sometimes refer 50 72 Mostly refer to 23 20 specialist Why not? Do not have the 74 77 expertise Do not have 87 62 necessary testing facilities Complicated/ 33 12 excessive forms Lack of knowledge 67 27 pertaining to administration issues and workers’ compensation Lack of 24 12 cooperation from employer or workers’ compensation Patient attitude 13 23 Concern that 26 12 patient will suffer financially Wish to avoid 22 11 medicolegal issues Lack of adequate 13 8 reimbursement Time constraints 35 29 Three most Do not have Do not have important expertise; do not expertise; do not reasons for not have facilities; lack have facilities; lack investigating by of administration of administration yourself or WSIB or WSIB knowledge knowledge *WSIB ⫽ Workplace Safety and Insurance Board: the provincial workers’ compensation board. †p ⫽ 0.0025 for comparison between pulmonologists and family physicians. 1528

necessary testing facilities, lack of expertise, and lack of knowledge about the administrative and workers’ compensation process as the three key reasons for referring patients for diagnosis. Those reporting that they carry out the diagnostic investigations themselves were asked their reasons for not referring. The three key reasons were feeling competent to diagnose the condition themselves (75% for pulmonologists, 83% for family physician), lack of access to specialists (32% for pulmonologists, 33% for family physicians), and lack of timely access to specialists (25% for pulmonologist, 63% for family physicians). There were no differences in investigation practice related to age, gender, or year of graduation. The physicians were asked what advice regarding work they provide to patients with suspected OA during the diagnostic process. Five percent of pulmonologists and 4% of family physicians reported that they always advised their patients with suspect OA to leave work until the diagnosis is confirmed, and 27% of pulmonologists reported that they did this most of the time compared with 9% of family physicians. The physicians were asked to rate their knowledge of various OLDs and whether they wanted more education on specific topics (Table 4). For all topics, pulmonologists were more likely to rate their knowledge at a higher level than family physicians (p ⬍ 0.0001). While ⬎ 50% of family physicians indicated an interest in more education for all the listed topics, the majority of pulmonologists did not indicate a need for further education for asbestosrelated lung disease, silicosis, and lung cancer. There were no statistically significant differences between the two groups related to need for education on a particular topic except for silicosis where more family physicians indicated a desire for more education (p ⫽ 0.0051). For both pulmonologists and family physicians, the most commonly cited reasons for not wanting further education were that they were not seeing enough patients with these problems, the availability of specialists, and time constraints. The most important reported source of information about OA was different among the two physician groups (Table 5). The three key sources were continuing education conferences, journal articles, and consultation notes from specialists. Family physicians were more likely to view consultation reports from specialists and information booklets from government or professional organizations as important, while pulmonologists were more likely to identify continuing medical education conferences as an important source. Very few listed information from professional or government organizations or Web sites as important sources. Finally, physicians were asked what would make recognition and management of OA easier (Table Original Research

Table 4 —Knowledge of OLD*

Variables OA (including RADS)† Excellent/good Average Fair/poor Want more education on this topic Work-aggravated asthma† Excellent/good Average Fair/poor Want more education on this topic Asbestos-related disease† Excellent/good Average Fair/poor Want more education on this topic Silicosis† Excellent/good Average Fair/poor Want more education on this topic Industrial bronchitis† Excellent/good Average Fair/poor Want more education on this topic Lung cancer† Excellent/good Average Fair/poor Want more education on this topic Barriers to further education

Table 5—Information Sources

Pulmonologists, %

Family Physicians, %

49 45 6 82

24 48 29 75

58 40 2 68

27 49 24 76

75 23 2 43

22 41 37 60

57 35 8 42

8 30 63 70

31 41 28 71

8 35 57 78

72 20 6 48

35 50 16 60

Do not see enough; have access to specialists; time constraints

Do not see enough; have access to specialists; time constraints

*RADS ⫽ reactive airway dysfunction syndrome. †p ⬍ 0.0001 for comparison between pulmonologists and family physicians for rating their knowledge of the particular topic.

6). A readily available and timely referral source was the most common factor cited by both groups. Family physicians were more likely to identify easily available standard tests and readily available and timely referral sources compared with pulmonologists. Discussion Although response rates were low in this survey, they were in keeping with, or better than, that www.chestjournal.org

What Is the Most Important Source of Information About OLD? Continuing medical education, conferences Journal articles Consultation reports from specialists Newsletters from professional organizations Information booklets from government or professional organizations Web sites

Pulmonologists, %

Family Physicians, %

p Value*

43

50

0.3491

58 25

32 50

0.0007 0.0013

9

14

0.3952

2

15

0.0051

5

5

0.9435

*Comparing pulmonologist and family physician responses.

reported in a similar study of members of the American College of Chest Physicians by Harber et al25 (response rate, 25.5%). Our response rate from pulmonologists was 49%, and 26% for family physicians. The demographics of the respondents to the survey are similar to those of the total respective groups of Ontario physicians. The Canadian Medical Association reported that 25% of pulmonologists were ⱖ 55 years old (in our study, 36%) and 78% were men (in our study, 64%).28 Values for family physicians were 32% ⬎ 55 years of age (in our study, 32% ⬎ 50 years old), and 64% were men (in our study, 68%). There were no significant associations between age, gender, and year of graduation for responses to key questions, suggesting that the dif-

Table 6 —Reported Ways To Improve Recognition and Management of OA

Variables Improved remuneration Easily available standard tests Readily available and timely referral sources Templates for asking questions during history taking Education on how to initiate a claim Toll-free telephone numbers or Web sites for information Better education to enable early detection and referral

Pulmonologists, %

Family Physicians, %

p Value*

50 11

53 53

0.7190 ⬍ 0.0001

58

78

0.0068

50

63

0.1147

44

30

0.0725

48

43

0.5216

42

70

0.0003

*Comparing pulmonologist and family physician responses. CHEST / 132 / 5 / NOVEMBER, 2007

1529

ference in age and gender distribution between responders and nonresponders probably did not significantly bias the results. The majority of both pulmonologists and family physicians report that they see ⬍ 20 patients per year with OLD, findings similar to those of Harber et al.25 The majority of pulmonologists report taking a workplace exposure history all or most of the time compared with just over half of family physicians. This is in contrast to the findings of a number of studies18 –20 that found physicians tend not to take a workplace-exposure history. It is also in contrast with the recent findings of Poonai et al,12 who studied barriers to the diagnosis of OA in Ontario. They reported, from patient-completed questionnaires, that the main reason for delay in diagnosis was the lack of enquiry about work relatedness by family physicians (41%). Without actual examination of physician notes, it is impossible to know the accuracy of patient reporting. Physicians noted time constraints and forgetting to ask as barriers to taking a workplace exposure history. A short workplace exposure history template could be developed for use to enhance this component of history taking. More than 50% of both groups identified such as template as a way to improve recognition of OA. This was also identified in a UK Health and Safety Executive study29 on occupational health and primary care. One of its recommendations was for the “development of simple screening questionnaires to highlight those patients with work-related ill health.” Similarly, Harber and Merz21 also suggested the use of focused, brief histories and possibly computer-based methods in their study of barriers to recognizing occupational disease. The use of tools and systems that support information collection are indicated. Electronic medical records also provide the opportunity to encourage the inclusion of workplace exposure information in the medical record. More than 70% of both physician groups report that they sometimes or mostly refer these patients out for diagnosis. Key reasons for referral for both groups include personal lack of expertise, lack of necessary testing facilities, and lack of knowledge about administrative issues related to the Workplace Safety and Insurance Board, the workers’ compensation board in Ontario. Key reasons for not referring reported by physicians who mostly do the investigation themselves include feeling competent to diagnose the condition themselves and lack of access or timely access to specialists. These findings suggest that readily accessible expertise for the diagnosis of OA is needed. Canadian guidelines for occupational asthma have recommended that investigations be performed to confirm or refute the diagnosis of OA before advising 1530

the patient to stop work, unless it is considered dangerous to the patient to continue exposures.30 The investigations for OA are more helpful if they can be performed while the patient is still working in the suspected causative work area in order to document changes in peak flows and methacholine responsiveness at work and away from work. Job termination may result in the inability to confirm the diagnosis of OA without specific inhalation challenge testing, which is of limited availability, is costly, and is time consuming. It is somewhat disconcerting that this survey indicated a significant minority of pulmonologists, almost one third, always or usually advise the patient to leave work before a diagnosis is confirmed. This may be expected to delay a conclusive diagnosis and for some patients may result in unnecessary job loss if the investigation reveals the patient does not have OA. This concept should be emphasized future medical education. We obtained detailed information on physician perception of their knowledge of specific occupational lung diseases. Generally, specialists rated their knowledge higher than the family physicians. Both groups of physicians also indicated an interest in further education about a number of these topics. We have identified similar findings in relation to occupational contact dermatitis.26 Harber et al25 identified significant educational needs related to occupational respiratory disease among members of the American College of Chest Physicians. The findings of the recent Health and Safety Executive study confirm the need for basic medical education and continuing education related to occupational diseases such as OLD and OA.29 It is important that such education be included in medical school training and that appropriate continuing education is also available. A challenge for those involved in occupational medicine educational programs is competing for curriculum time with other topics. As indicated in the Health and Safety Executive report, it is the responsibility of key system stakeholders to support these educational initiatives.29 The Workplace Safety and Insurance Board in Ontario funds a Workplace Health Champions Program. This program provides support for a workplace health champion at each of the six medical schools in Ontario who champion the inclusion of workplace health content in both medical school and continuing education courses. We also obtained information on the preferred sources of information. The most important sources of information were continuing education events, journals, and reports from specialists. The importance of the consultation note as an educational tool is interesting and suggests there is an opportunity for specialists seeing a high volume of OA to consider the potential of their consultation note an important Original Research

information source. Of equal note, newsletters from professional organizations, booklets from government or professional associations and Web-based materials were not commonly cited sources. The most important item indicated from the survey that would make recognition and management of OA easier was readily available and timely access to specialists. In Ontario, the Workplace Safety and Insurance Board has developed a specialty clinic model of care to facilitate referral of workers to specialized centers in academic hospitals for diagnosis and management. The occupational disease specialty clinic is at St. Michael’s Hospital. The multidisciplinary team includes pulmonologists, allergists, dermatologists, and occupational medicine specialists as well as nursing, occupational hygiene for detailed work exposure assessment and rehabilitation professionals to assist with return to work. Other items to facilitate recognition identified in the survey responses included templates for asking specific questions during history taking and further education on the workers compensation process and OA. Our findings can help to inform policy decisions related to the delivery of health services for workers with possible OA. References 1 Blanc PD, Eisner MD, Israel L, et al. The association between occupation and asthma in general medical practice. Chest 1999; 115:1250 –1264 2 Tarlo SM, Leung K, Broder I, et al. Prevalence and characterization of asthmatics symptomatically worse at work among a general asthma clinic population. Chest 2000; 118:1309 – 1314 3 Blanc PD, Toren K. How much adult asthma can be attributed to occupational factors? Am J Med 1999; 107:580 –587 4 Adams WG. Long-term effects on the health of men engaged in the manufacture of toluene diisocyanate. Br J Ind Med 1975; 32:72–78 5 Chan-Yeung M. Fate of occupational asthma and follow-up study of patients with occupational asthma due to Western Red Cedar (Thuja plicata). Am Rev Respir Dis 1977; 116: 1023–1029 6 Chan-Yeung M, Lam S, Koerner S. Clinical features and natural history of occupational asthma due to Western Red Cedar (Thuja plicata), Am J Med 1982; 72:411– 415 7 Burge PS. Occupational asthma in electronic workers caused by colophony fumes: follow-up of affected workers. Thorax 1982; 37:348 –353 8 Paggiaro PL, Loi Am, Rossi O et al. Follow-up study of patients with respiratory disease due to toluene diisocyanate (TDI). Clin Allergy 1984; 14:463– 469 9 Hudson P, Cartier A, Pineon L, et al. Follow up of occupational asthma due to various agents. J Allergy Clin Immun 1985; 76:682– 688 10 Paggiaro PL, Vagaggini B, Bacci E, et al. Prognosis of occupational asthma. Eur Respir J 1994; 7:761–767

www.chestjournal.org

11 Malo J-L, Chan-Yeung M. Occupational asthma. J Allergy Clin Immunol 2001; 108:317–328 12 Poonai N, van Diepen S, Bharatha A, et al. Barriers to diagnosis of occupational asthma in Ontario. Can J Public Health 2005; 96:230 –233 13 Tarlo SM, Banks D, Liss G, et al. Outcome determinants for isocyanate-induced occupational asthma among compensation claimants. Occup Environ Med 1997; 54:756 –761 14 Boulet L-P, Boutet M, Laviolette M, et al. Airway inflammation after removal from the causal agent in occupational asthma due to high and low molecular weight agents. Eur Respir J 1994; 7:1567–1575 15 Perfetti L, Cartier A, Ghezzo H, et al. Follow-up of occupational asthma after removal from or diminution of exposure to the responsible agent. Chest 1998; 114:398 – 403 16 Tarlo SM, Liss GM, Yeung KS. Intervention measures and changes in compensation claims for asthma due to diisocyanates. Occup Environ Med 2002; 59:58 – 62 17 Tarlo SM, Liss G, Corey P, et al. A workers compensation claim population for occupational asthma: comparison of subgroups. Chest 1995; 107:634 – 641 18 McCurdy SA, Morrin LA, Memmott MM. Occupational history collection by third-year medical students during internal medicine and surgery inpatient clerkships. J Occup Environ Med 1998; 40:680 – 684 19 Institute of Medicine. Environmental medicine: integrating a missing element into medical education. Washington, DC: National Academy Press, 1995 20 Burnstein JM, Levy BS. The teaching of occupational health in US medical schools: little improvement in 9 years. Am J Public Health 1994; 84:846 – 849 21 Harber P, Merz B. Time and knowledge barriers to recognizing occupational disease. J Occup Environ Med 2001; 43:285–288 22 Merrill RN, Pransky G, Hathaway J, et al. Illness and the workplace: a study of physicians and employers. J Fam Pract 1990; 31:55–59 23 Taylor R, Ruth D. A survey of environmental and occupational health needs of GPs. Aust Fam Phys 1995; 24:1433– 1439 24 Russell G, Brown JB, Stewart M. Managing injured workers. Can Fam Phys 2005; 51:78 –79 25 Harber P, Scanlon PD, do Pico G, et al. Role of chest physicians in detection and treatment of occupational and environmental respiratory disease. Chest 1995; 107:1156 – 1161 26 Holness DL, Tabassum S, Tarlo SM, et al. Dermatologist and family practitioner practice patterns for occupational contact dermatitis. Australas J Dermatol 2007; 48:22–27 27 Dillman DA. The design and administration of a mail survey. Annu Rev Sociol 1991; 17:225–249 28 cma.co. Statistical information on Canadian physicians. Available at: http://cma.ca/index.cfm/ci_id/16959/1a_id/1.htm#1. Accessed September 24, 2007 29 Health and Safety Laboratory. The profile of patients’ occupational health in primary care: health and safety executive, research report 254. Norwich, UK: Her Majesty’s Stationary Office, 2004 30 Tarlo SM, Boulet L-P, Cartier A, et al. Canadian Thoracic Society guidelines for occupational asthma. Can Respir J 1998; 5:289 –300

CHEST / 132 / 5 / NOVEMBER, 2007

1531