ARTICLE IN PRESS Respiratory Medicine (2008) 102, 738–743
Under-diagnosis of chronic obstructive pulmonary disease: A qualitative study in primary care Julia A. Waltersa,, Emily C. Hansenb, E. Haydn Waltersa, Richard Wood-Bakera a
Menzies Research Institute, University of Tasmania, 43 Collins Street, Hobart TAS 7001, Australia Primary Health Care Research, Evaluation and Development Group, Menzies Research Institute, University of Tasmania, 43 Collins Street, Hobart TAS 7001, Australia b
Received 31 May 2007; accepted 9 December 2007 Available online 28 January 2008
KEYWORDS Diagnosis; Qualitative research; Chronic obstructive pulmonary disease; Primary health care
Summary Background: Chronic obstructive pulmonary disease (COPD) is under-diagnosed and diagnosis often occurs late thereby reducing opportunities to prevent deterioration. Investigation of causes has focused on the use of spirometry but the role of attitudes of doctors and patients has not been directly investigated. Methods: A cross-sectional study of patients diagnosed with COPD and their general practitioners in two general practices in Tasmania who participated in semi-structured interviews and focus groups. Practice records were examined and patients underwent spirometry, quality of life and symptom assessment. Iterative content analysis identified themes that were compared with quantitative data. Findings: For the 32 participating patients, mean recorded duration of COPD was 8.0 (SD 6.8) years and mean FEV1% predicted 38.4% (SD 19.8). Qualitative data were obtained from 14 (44%) of patients with COPD (5M/9F) and 16 general practitioners (GPs) (10M/6F). We identified three themes around the diagnosis of COPD in primary care: patients’ labels, acquiring and communicating a diagnosis and consequences of delaying or withholding diagnosis. Doctors correctly recognised that patients were unfamiliar with COPD. They intentionally avoided early diagnosis as a result of nihilism and misperception of patients’ attitudes. Patients often received the diagnosis from other sources and found delayed diagnosis frustrating. Interpretation: Nihilistic attitudes to COPD may explain the disappointing results from promotion of spirometry in primary care. Education about COPD for doctors in primary care is needed and action to increase awareness and understanding of COPD in the community. & 2008 Elsevier Ltd. All rights reserved.
Corresponding author. Tel.: +61 3 62264798; fax: +61 3 62264894.
E-mail addresses:
[email protected] (J.A. Walters),
[email protected] (E.C. Hansen),
[email protected] (E.H. Walters),
[email protected] (R. Wood-Baker). 0954-6111/$ - see front matter & 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.rmed.2007.12.008
ARTICLE IN PRESS Under-diagnosis of chronic obstructive pulmonary disease
Introduction Previous studies have found that chronic obstructive pulmonary disease (COPD) is both under-diagnosed and diagnosed late in primary care.1,2 Earlier diagnosis would increase the opportunity for intervention, especially through smoking cessation,3 which confers greatest benefit the earlier it is achieved.4 Spirometry is required for the diagnosis of COPD,5,6 particularly in its early stages, as reliance on symptoms will miss up to 50% of cases.7 We have previously reported on the low rate of use of spirometry in the diagnosis of COPD in primary care,8 but little is known about other barriers to its early diagnosis due to attitudes of patients or doctors. Qualitative research, with its inductive and interpersonal nature,9 is well suited to exploring barriers and providing new insights that can improve practice. In this study we investigated attitudes influencing the diagnosis of COPD amongst doctors and patients with COPD in a primary care setting.
Methods Selection of participants Nineteen general practitioners (GPs) in two practices in Southern Tasmania were invited to participate and asked to select patients they recognised as having COPD from searches of practice databases using COPD diagnosis or repeat prescriptions for respiratory medications. Patients with cognitive impairment or another serious medical condition were excluded. Subjects were invited to participate by letter. Informed consent was obtained from respondents and GPs. The study was approved by the Southern Tasmanian Health and Medical Human Research Ethics Committee (2002 H006650).
Clinical assessment Participating patients underwent spirometry with bronchodilator reversibility testing10 using European Community for Steel and Coal reference values,11 data collection on demographics and smoking history. They completed the anxiety and depression questionnaire for general medical settings,12 the St. George’s Respiratory Quality of Life Questionnaire (SGRQ)13 and MRC functional dyspnoea scale.14 COPD was diagnosed by a post-bronchodilator FEV1/FVC ratio o0.76 with severity determined by FEV1% predicted: mild X80%, moderate 50–79%, severe 30–49%, very severe o30%.15 Practice records were reviewed to extract data from the time of diagnosis and subsequently using a structured template.
Interviews with patients Interviews were conducted at participants’ homes before clinical assessments to avoid bias through contact with researchers. An interview guide based on open-ended questions for eliciting an explanatory model was used in semi-structured interviews (see Appendix A) with additional reminder prompts for the interviewer to explore issues not spontaneously mentioned by the interviewee.16 The study
739 aimed for a non-probability sample of 20 participants with a balance of gender, age and geographical location.17
Focus groups with general practitioners All doctors in the involved practices were invited to participate in an interview or focus group lasting from 60 to 90 min conducted during working hours at doctors’ premises. An experienced facilitator used interview prompts and questions (Appendix A) to guide discussions exploring attitudes to the diagnosis and management of COPD.
Data analysis Interviews and focus group discussions were recorded and transcribed verbatim. Transcripts were analysed independently by two authors (JW and EH) within an iterative interpretive framework drawing on aspects of grounded theory.18 Each transcript was read as soon as possible after the interview and given initial codes and memos. Notice was paid to repeated stories and links made between events and illness and how participants spoke about the diagnosis. This iterative process was repeated as more interviews were conducted and transcripts were then compared with each other. Codes were grouped into larger categories and after ongoing iteration these categories were grouped into major themes. The process of analysis was performed using NVivo (version 2, Qualitative Solutions & Research International, Melbourne, Vic.). Emergent themes were verified by JW and EH and discussed among all authors, while a more detailed in-depth analysis of some thematic areas was performed independently by JW and EH.19,20 Quantitative results were analysed using SPSS. Variables are presented as mean and standard deviation (SD) if normally distributed or median and interquartile range (IQR) for non-normal distribution. Continuous variables were compared between groups using Student’s t-test or Mann–Whitney U-test. Chi square tests were used to compare categorical variables.
Results Recruitment Of 106 patients identified by searches, 32 patients with COPD completed clinical assessments of whom 14 patients underwent interview, nine from an urban and five from a rural/suburban practice (Figure 1). Three interviewees did not complete clinical assessments (one required treatment for another condition and two said their respiratory symptoms were too severe) thus no comparisons can be made between these patients and those whose data were analysed. Four interviewees were subsequently found to have been incorrectly classified with COPD (two had asthma and two had chronic bronchitis without airflow limitation). Their interview data are not included in this analysis. Amongst the 32 participants with COPD, 13 (41%) were male and 27 (84%) were ex-smokers (Table 1). The dates of diagnosis were unavailable in five patient records and eight patients did not provide dates. COPD was classified as: mild
ARTICLE IN PRESS 740
J.A. Walters et al.
2 (6%), moderate 13 (41%), severe 9 (28%) and very severe 8 (25%). The mean duration of COPD from practice records was 7.9 (6.4) years and from patient report 10.0 (6.9) years. Participants who were interviewed were younger (mean difference 8 years, SE 3) with more severe airflow limitation (mean difference FEV1% predicted 15, SE 6) than those not interviewed. However, there were no significant differences in quality of life, functional dyspnoea or presence of symptoms of anxiety and depression (Table 1). Sixteen (84%) GPs provided qualitative data (10M/6F), their length of time in practice ranged from 5 to 30 years.
Findings This study revealed findings about the diagnosis of COPD in primary care in three thematic areas which we titled: 1. Patients’ labels, 2. Acquiring and communicating a diagnosis, 3. Consequences of delaying or withholding diagnosis.
Patients’ labels The diagnosis of COPD was rarely named directly by any patient. When asked to describe their illness, 13 interviewees used the term emphysema, one used COPD in addition to emphysema and one called it asthma. The diagnosis was rarely given as a direct, specific answer. Frequently it was qualified with a description of another perceived prior respiratory condition, usually ‘‘asthma’’: I think it is classed as emphysema but it is basically COPD. Asthma, all that sort of thing it stems from. It started off as asthma. As I got older it progressed onto bigger titles. (Female, 49) It was also common for the diagnosis given by health professionals to vary over relatively short periods of time: Well I was originally given the tag of asthma. We were travelling and my condition got worse. Eventually, I think they’d probably diagnosed emphysema and asthma. (Male, 69)
Figure 1
Patients used diagnostic terms interchangeably to describe their own symptoms: The emphysema has I think brought on the bronchitis. Before the bronchitis sets in I get a lot of coughing, so you
Table 1 Characteristics of participants with COPD according to participation in semi-structured interviews. Participant characteristics
Interviewed (n ¼ 14)
No interview (n ¼ 18)
5 (35.7) 8 (44.4) Male (%) 10 (71.4) 15 (83.3) Ex-smoker (%) 67.0 (16.0) 72.5 (15.0) Age (years),y 12.0 (9.5) 9.5 (9.8) Duration COPD patient report (years) 5.5 (11.3) 9.0 (7.3) Duration COPD practice record (years) 55.21 (26.2) 66.7 (23.1) Smoking exposure (pack years)z Years smoking pre-diagnosisz 40.57 (11.0) 42.40 (8.4) 38.4 (19.8) 52.9 (14.5) FEV1% predictedy 0.50 (0.21) 0.61 (0.13) FEV1/FVC ratio 52.3 (24.5) 47.8 (16.1) SGRQ score: overallz 67.9 (31.3) 64.3 (21.7) SGRQ score: activity limitationz 66.7 (23.9) 64.6 (19.6) SGRQ score: symptomsz 38.8 (23.8) 31.2 (17.5) SGRQ score: impactsz 3.1 (1.7) 2.7 (1.3) Functional dyspnoea (MRC grade)z 8 (57.1) 7 (38.9) Anxiety screening: positive (%) 9 (64.3) 13 (72.2) Depression screening: positive (%) SGRQ, St. George’s Respiratory Questionnaire. Median (IQR). y po0.05. z Mean (SD).
Recruitment of patients and general practitioners in a qualitative study of COPD in primary care.
ARTICLE IN PRESS Under-diagnosis of chronic obstructive pulmonary disease could say that it is an asthma attack which then brings on the bronchitis. (Female, 76) GPs themselves noted that changes in diagnostic terminology used by specialist respiratory groups during their working life were confusing for them and difficult to explain to patients. From the population point of view changes just confuse them. If you say emphysema, they say ‘Oh, that’s what you’re really talking about!’ They won’t have heard of COAD, and they certainly wouldn’t have heard of COPD. (Female GP) GPs would generally use the term ‘‘emphysema’’ to a patient when they made a diagnosis of COPD as they felt patients were familiar with emphysema but did not understand the acronym.
Acquiring and communicating a diagnosis Consideration of a diagnosis of COPD in the presence of risk factors, especially smoking, was part of GPs’ diagnostic process: If someone has been smoking for say 20 years, they are going to have some element of COPD. (Male GP) However they were reluctant to label the condition. This stemmed mainly from the GPs’ perceived implications of a diagnosis of emphysema for a patient, who would understand it as a serious, ultimately terminal disease: You don’t actually say they have emphysema, they are not keen to have that. They don’t like emphysema; they have seen Grandad starved of oxygen. I find that they will fight that label. (Female GP) Formal diagnosis was often delayed as no apparent advantage was seen by GPs in applying the diagnosis: I don’t suppose it really matters whether one puts a label to it. (Male GP)
741 and this confirmed that of those seven had the diagnosis of COPD made in hospital. More unusually, the diagnosis resulted from the process of ‘‘gradual evolution’’ in patients being seen regularly: It’s just something that dawns on us after a while. (Male GP) Even after the diagnosis had become clear in the GP’s mind, this may have been withheld from the patient, with eventual disclosure occurring incidentally via a pharmacist or allied health practitioner. Well no one actually told me I had it. One day they gave me the Atrovent and I said ‘This is a new one’ and they said ‘That is for your emphysema.’ (Female, 50) Negative personal assessments of COPD were held by GPs and this could influence them against communicating the diagnosis: (COPD) is a horrible way to spend the last 10 or 15 years of your life. (Male GP) Diagnosis was often preceded by frequent presentation by patients to a GP: I think you see a lot of people over a period of time, who are perhaps in the middle period of life, who might have recurrent respiratory infections- smokers. (Male GP) The delay in communicating the diagnosis is supported by discrepancies in the time since diagnosis. The duration in practice records exceeded patient-reported duration for 5/22 (23%) patients, mean difference 7.4 years (range 1–12). Of these patients at the time of study, all had clinical depression on screening. Overall quality of life scores and individual domain scores for symptoms, activity limitations or impacts did not differ significantly between patients with discrepant or nondiscrepant disease duration.
Treatment was usually initiated before formal diagnosis of the emerging clinical picture by the GP:
Consequences of delaying or withholding a diagnosis
Gradually a picture evolves- patients chronically short of breath, tend to have cough and wheeze, who perhaps benefit from bronchodilators and Atrovent regularly rather than just during exacerbations. (Male GP)
Patients expressed frustration when not given a diagnosis. Delay meant doctors often failed to provide information about COPD, causing patients to seek information elsewhere, or remain poorly informed:
Delayed recording of a diagnosis was evident with the duration since diagnosis being greater according patients’ reports than from practice records in 14 (64%) of 22 patients with available data. The mean discrepancy was 5.4 years (range 3–15) and among these patients, seven (50%) screened positively for clinical depression at the time of this study. Eventually, formal diagnosis often resulted from admission to hospital for an acute exacerbation of COPD:
I read up in a couple of medical books. I have been like this for two or three years and no one actually told me I had emphysema. (Female, 50) I don’t really know that much about it. I haven’t really had it explained to me. (Male, 55)
They get crook and get admitted to hospital and come back with a diagnosis. (Male GP) Extraction of data from practice records contemporary with the time of diagnosis was possible for 10 participants
A major disadvantage of withholding a diagnosis is potentially losing the opportunity for preventing deterioration by achieving smoking cessation. GPs appeared pessimistic that a diagnosis of COPD could assist patients to stop smoking: The percentage of patients diagnosed with COPD that have stopped smoking are far less than those diagnosed with ischaemic heart disease. (Male GP)
ARTICLE IN PRESS 742 GPs reported routinely giving smoking cessation advice to any smoker, but felt smokers with COPD were difficult to motivate: They all imagine that they will have a heart attack or get lung cancer, and they don’t think from fifty onwards they will not be able to walk down the street. (Female GP) However patients referred positively to their doctors’ advice to quit smoking: The doctor told me ‘Smoke or die’. So I gave up. That is when I first had an outbreak of emphysema. (Female, 60) Patients universally recognised the difficulty of quitting smoking, and were apologetic if they continued to smoke yet were still receptive to advice: ‘If I had a doctor telling me I had to give it up it would give me a bit of a shock, but I just can’t tell myself to do it.’ (Female, 60)
Discussion This study found that making a diagnosis of COPD by doctors in primary care is often delayed, and that this delay may be intentional. There was also evidence that even after the diagnosis had been made, this was not communicated to patients. These findings were consistent with both the expressed attitudes and clinical practice of doctors and reports of patients’ experiences. The delayed diagnosis of COPD differed from some other chronic diseases such as hypertension, where under-diagnosis is likely to be due to conflicting diagnostic criteria.21 In COPD, the delay was rationalised by misperception of patients’ unwillingness to be given a diagnosis and doctors’ pessimistic attitudes to prognosis. Similar communication deficits in COPD between primary care physicians and patients were found in the US22 and in Spain where only 11.2% of smokers or ex-smokers had any spontaneous knowledge of COPD.23 Perhaps the reluctance on the part of GPs was to protect patients against psychological impact in a population in whom depression and anxiety are common.24 Although doctors perceived their advice on smoking cessation to be relatively ineffective, this was not generally supported by participants who assessed it as valuable. Other studies have shown that doctors may be reluctant to give advice on smoking cessation for fear of harming the doctor–patient relationship or because of a preference for addressing the patient’s agenda.25 The strengths of this study included the use of data from practice records to validate patient-reported experience and the direct comparison of the attitudes of patients with those of their doctors.17 Participating patients were symptomatic, had significantly impaired quality of life and moderate or severe disease. This reflects the general situation in primary care in Australia, where patients with mild disease are less likely to be diagnosed.26 There are limitations to the study. We may not have identified the entire COPD population in the practices, firstly due to the response rate, although this is line with previous studies27,28 and secondly because of incomplete
J.A. Walters et al. recording of the diagnosis.29 To overcome the latter, we supplemented our search using respiratory medication use, resulting in an age distribution in our sample similar to that in a UK practice cohort.29 The number of patient interviews available for analysis was constrained by substantial misclassification, although the rate (22%) was similar to that found in a study of UK general practice.29 Although this reduced the number of patient interviews available for qualitative analysis, themes that emerged were consistent between interviewees.17 We believe our findings are representative of Australian general practice, despite the majority of participants being female. An Australian prevalence study in 1998 found selfreported emphysema to be only marginally more prevalent in males than females.30 In assessing time of diagnosis, it is acknowledged that patients’ recall of diagnosis was subjective and may not have been accurate. However the high proportion of patients (86%) with discrepancies in duration is striking. There was no reason to expect that this sample of doctors would differ from others in their clinical practice, as they were not selected on the basis of having a special interest in chronic respiratory diseases. These findings are likely to be generalisable to general practice in Australia as a whole and relevant to similar systems of primary care internationally. There have been disappointing results from efforts to increase the use of spirometry in primary care and achieve earlier diagnosis of COPD.31,32 These results indicate that attitudes of GPs to the diagnosis of COPD are also an important cause of diagnostic delay. They highlight the need for education about the nature, natural history and treatment of COPD for GPs to reduce their nihilism and for efforts to increase awareness and understanding of COPD in the community.
Acknowledgements Professor P. Mudge assisted with study concept and recruitment. Dr. J Gartlan, Dr. R. Boland, P. Fitzpatrick, S. Davoren, E. Hammer assisted with data collection. We gratefully acknowledge the cooperation of doctors and staff from the participating general practices and the contribution of patients who shared their experiences. Dr. Julia Walters is the recipient of a Glaxo Smith Kline Postgraduate Support grant. Competing Interests None
Appendix A. Supplementary data Supplementary data associated with this article can be found in the online version at doi:10.1016/j.rmed.2007.12.008.
References 1. Van Schayck C, Loozen J, Wagena E, Akkermans R, Wesseling G. Detecting patients at a high risk of developing chronic
ARTICLE IN PRESS Under-diagnosis of chronic obstructive pulmonary disease
2.
3.
4.
5.
6.
7.
8.
9.
10. 11.
12.
13. 14.
15.
16.
obstructive pulmonary disease in general practice: cross sectional case finding study. Br Med J 2002;324:1370–5. Takahashi T, Ichinose M, Inoue H, Shirato K, Hattori T, Takishima T. Underdiagnosis and undertreatment of COPD in primary care settings. Respirology 2003;8:504–8. Anthonisen NR, Connett JE, Kiley JP, Altose MD, Bailey WC, Buist AS, et al. Effects of smoking intervention and the use of an inhaled anticholinergic bronchodilator on the rate of decline of FEV1. The lung health study. J Am Med Assoc 1994;272: 1497–505. Doll R, Peto R, Boreham J, Sutherland I. Mortality in relation to smoking: 50 years’ observations on male British doctors. Br Med J 2004;328:1519–28. National Collaborating Centre for Chronic Conditions. Chronic obstructive pulmonary disease. National clinical guideline on management of chronic obstructive pulmonary disease in adults in primary and secondary care. Thorax 2004;59:1–232. Pauwels RA, Buist AS, Calverley PM, Jenkins CR, Hurd SS. Global strategy for the diagnosis, management, and prevention of chronic obstructive pulmonary disease. NHLBI/WHO Global Initiative for Chronic Obstructive Lung Disease (GOLD) Workshop summary. Am J Respir Crit Care Med 2001;163:1256–76. Buffels J, Degryse J, Heyrman J, Decramer M. Office spirometry significantly improves early detection of COPD in general practice: the DIDASCO Study. Chest 2004;125:1394–9. Walters JA, Hansen E, Mudge P, Johns DP, Walters EH, WoodBaker R. Barriers to the use of spirometry in general practice. Aust Fam Physician 2005;34:201–3. Morgan DL. Practical strategies for combining qualitative and quantitative methods: applications to health research. Qual Health Res 1998;8:362–76. American Thoracic Society. Standardization of Spirometry, 1994 Update. Am J Respir Crit Care Med 1995; 152:1107–36. Quanjer PH, Tammeling GJ, Cotes JE, Pedersen OF, Peslin R, Yernault JC. Lung volumes and forced ventilatory flows. Report of the Working Party Standardization of Lung Function Tests, European Community for Steel and Coal. Eur Respir J 1993; 6:5–40. Goldberg D, Bridges K, Duncan-Jones P, Grayson D. Detecting anxiety and depression in general medical settings. Br Med J 1988;297:897–9. Jones P, Quirk F, Baveystock C. The St. George’s respiratory questionnaire. Respir Med 1991;85:25–31. Fletcher C. Standardised questionnaire on respiratory symptoms: a statement prepared and approved by the MRC committee on the aetiology of chronic bronchitis (MRC breathlessness score). Br Med J 1960;2:1665. GOLD Executive Committee. Workshop report, global strategy for diagnosis, management, and prevention of COPD (updated 2005) 2005, accessed June 2005. Kleinman A. The illness narratives: suffering, healing and the human condition. New York: Basic Books; 1988.
743 17. Patton MQ. Enhancing the quality and credibility of qualitative analysis. Health Serv Res 1999;34:1189–208. 18. Charmaz K. Qualitative interviewing and grounded theory analysis. In: Gubrium J, Holstein J, editors. Handbook of interveiw research: context and method. Thousand Oaks, CA: Sage; 2002. 19. Sandelowski M. Divide and conquer: avoiding duplication in the reporting of qualitative research. Res Nurs Health 2006; 29:371–3. 20. Hansen E, Walters JAE, Wood-Baker R. Explaining chronic obstructive pulmonary disease (COPD): perceptions of the role played by smoking. Sociol Health Illn 2007;29:1–20. 21. Pater C. The blood pressure ‘‘Uncertainty Range’’—a pragmatic approach to overcome current diagnostic uncertainties (II). Curr Control Trials Cardiovasc Med 2005;6:5–21. 22. Nelson M, Hamilton HE. Improving in-office discussion of chronic obstructive pulmonary disease: results and recommendations from an in-office linguistic study in chronic obstructive pulmonary disease. Am J Med 2007;120:S28–32. 23. Miravitlles M, de la Roza C, Morera J, Montemayor T, Gobartt E, Martin A, et al. Chronic respiratory symptoms, spirometry and knowledge of COPD among general population. Respir Med 2006;100:1973–80. 24. van Manen JG, Bindels PJ, Dekker FW, CJ IJ, van der Zee JS, Schade E. Risk of depression in patients with chronic obstructive pulmonary disease and its determinants. Thorax 2002;57: 412–6. 25. Coleman T, Murphy E, Cheater F. Factors influencing discussion of smoking between general practitioners and patients who smoke: a qualitative study. Br J Gen Pract 2000;50:207–10. 26. Frith P, Crockett A, Esterman A, McKenzie D. Airflow limitation and respiratory symptoms in a general practice population sample. Respirology 2004;9:A9. 27. Guthrie SJ, Hill KM, Muers MF. Living with severe COPD. A qualitative exploration of the experience of patients in Leeds. Respir Med 2001;95:196–204. 28. McHorney CA, Bricker DE. A qualitative study of patients’ and physicians’ views about practice-based functional health assessment. Med Care 2002;40:1113–25. 29. Faulconer ER, de Lusignan S. An eight-step method for assessing diagnostic data quality in practice: chronic obstructive pulmonary disease as an exemplar. Inform Prim Care 2004;12:243–54. 30. Abramson M, Matheson M, Wharton C, Sim M. Prevalence of respiratory symptoms related to chronic obstructive pulmonary disease and asthma among middle aged and older adults. Respirology 2002;7:325–31. 31. Kaminsky DA, Marcy TW, Bachand M, Irvin CG. Knowledge and use of office spirometry for the detection of chronic obstructive pulmonary disease by primary care physicians. Respir Care 2005;50:1639–48. 32. Enright PL. Office spirometry is 30 years old, but is not mature. Respir Care 2005;50:1619–20.
