Article Text


COPD: identification and testing

Statistics from


I. Jarrold, N. Eiser, K. Leach, T. Lethbridge. British Lung Foundation, London, UK

British Lung Foundation (BLF) surveys identified UK hotspots where the most acute admissions for chronic obstructive pulmonary disease (COPD) are expected. Two top hotspots, Nottingham and South Tyneside, were targeted with awareness-raising events. These events were supported by extensive publicity campaigns directed at GPs, pharmacies, community health centres and libraries with local media involvement. In South Tyneside there were also telemarketing campaigns directed towards at-risk groups. Day awareness-raising events were arranged at eight venues—supermarkets, shopping malls or Bingo halls—where spirometry was offered by trained respiratory nurses using Microlab portable spirometers. BTS criteria for acceptable results were used. People without known lung disease were invited to attend.

Complete datasets were obtained from 1273 attendees (table 1). 96% were Caucasian, with mean age 57 years (range 14–95). 60% were smokers or ex-smokers and 40% non-smokers. 20% attendees had abnormal spirometry. 11% had airflow obstruction and 9% had restrictive spirometry. Of 663 aged ⩾35 years with smoking histories, 18% had airflow obstruction while 9% had restrictive spirometry. In attendees with airflow obstruction, it was mild in 79% (FEV1 50–80% predicted), moderate in 19% (FEV1 30–49% predicted) and severe in 3% (FEV1<30% predicted). All those with abnormal spirometry were referred to their GPs, while all smokers were offered access to local quit-smoking programmes. 19% of the 621 people tested in Nottingham had abnormal spirometry compared with 22% of the 652 people in South Tyneside.

Abstract P185 Table 1

Those who suspected that they might have lung disease were presumably more likely to attend these events. A high percentage of attendees had abnormal results, particularly those aged ⩾35 years with smoking histories. Although additional telemarketing failed to increase the number of abnormalities found at events, possibly some at-risk people visited their GPs instead. As yet, we have not identified these people.

Our study demonstrates the benefits of identifying and targeting hotspots to identify and treat people with previously undiagnosed COPD. This is crucial to improving care and equally of access for this large population of disadvantaged patients. However, renewed efforts should be made to encourage members of the black and minority ethnic communities to attend such events.


I. Jarrold, N. Eiser, J. Morrell. British Lung Foundation, London, UK

Approximately 900 000 people in the UK have diagnosed chronic obstructive pulmonary disease (COPD). However, it is estimated that the true number is 3.7 million. The aim of this study was to help identify those undiagnosed COPD patients. The British Lung Foundation (BLF) commissioned a survey to calculate the prevalence of COPD in the UK by primary care organisation (PCO) and postcode, using Quality and Outcomes Framework data in general practice and Hospital Episode Statistics for hospital admission rates. Postcodes and PCOs were ranked for potential risk of COPD admissions and “hotspots” were identified. In addition, Experian Mosaic lifestyle segmentation was used to identify the key lifestyle characteristics of those at greatest risk of admission with COPD.

6.5 million people lived in the COPD “hotspot” PCOs selected by the BLF, with 1.9 million living in postcodes at high risk of future COPD admissions. The top 10 PCOs with highest proportions of people at risk of COPD are shown in table 1.

Abstract P186 Table 1

Lifestyles associated with greatest risk of admission with COPD were:

  1. Older people (60–80 years) living in crowded apartments in high density social housing (four times as likely to get admitted with COPD than the UK average).

  2. Older people, many in poor health from working in heavy industry, in low-rise social housing (three times more likely).

  3. Families with school-age children, living in large social housing estates on the outskirts of provincial towns (twice as likely).

  4. Older couples, mostly in small towns, who have bought their council houses (twice as likely).

This survey identified specific communities most at risk of future hospital admissions with COPD. These include ex-industrial and inner city areas, areas of particularly high social deprivation and unemployment and those with disproportionately high populations of older people. Those at risk had often worked in factories, steelworks, dockyards, mines or manufacturing, in semi-skilled jobs or were currently unemployed. Public awareness campaigns should be targeted at these “hotspots”. Better smoking cessation and diagnostic services are urgently needed in these areas to improve care for the missing millions of COPD patients.


1J. Upton, 1E. McCutcheon, 1C. Loveridge, 2J. Wiggins, 1S. Walker, 1M. Fletcher. 1Education for Health, Warwick, UK, 2Wexham Park and Heatherwood NHS Trust, Berkshire, UK

Introduction The National Institute for Health and Clinical Excellence guidelines for chronic obstructive pulmonary disease (COPD) recommends that medication should be initiated or changed in “patients who remain symptomatic”. In the absence of an agreed definition of “symptomatic”, less experienced healthcare professionals may be unclear when this should happen. The aim of the study was to gain expert consensus on which symptoms and signs are most important when considering treatment changes in patients with COPD.

Methods Leading COPD experts (members of COPD-related national committees or suggested by these members) from primary and secondary care were invited to participate in an online three-stage Delphi exercise. In round 1 participants listed the signs and symptoms that may be considered when initiating or changing treatment. Participants then scored the importance of each on a 5-point Likert scale (round 2). In round 3 participants were provided with the descriptive data analysis from round 2, and given the opportunity to revise their previous scores. Consensus was defined as ⩾80% of the panel scoring an item as 4 or 5 in round 3; these items were deemed by the panel to be the most important to consider when initiating or changing treatment in a patient with COPD.

Results 54 experts were identified, 37 (69%) of whom agreed to participate (providing a respectable sample size for this qualitative study). Only three participants dropped out during the three rounds (92% response rate). Consensus was gained on seven items (table 1).

Abstract P187 Table 1

Conclusion Assessing these seven symptoms and signs during COPD consultations may be a useful guide for all clinicians involved in managing patients with COPD. Clinical prompts (check lists) and education are needed in order to increase the opportunity to elicit this information during consultations.


1K. Ansari, 2A. Kay, 2I. K. Taylor, 1J. Munby, 2N. P. Keaney. 1Sunderland University, Sunderland, UK, 2Sunderland Royal Hospital, Sunderland, UK

Elderly patients have frequently lost height due to osteoporosis and related vertebral collapse. Osteoporosis is more prevalent in smokers and is a recognised co-morbidity associated with chronic obstructive pulmonary disease (COPD) and could affect the linear relationship between height and predicted forced expiratory volume in 1 s (FEV1). This in turn may lead to a misdiagnosing and misclassifying of COPD. We have studied arm span as a linear variable to calculate predicted values for spirometric measurements.

In 1999–2002 we studied a cohort of primary care patients thought to have COPD and re-examined 109 (55 men) of them during 2007–9. The demographic and spirometric measurements were recorded on each occasion. We calculated FEV1% predicted for measured and estimated height (arm span/1.03 and arm span/1.01 in men and women respectively: 1n 1999–2002 the subjects were aged 60.5±9.2 years, body mass index (BMI) was 26.4±4.7, measured FEV1 was 2.6±0.6 l (68.2±23.3% predicted), mean measured height was 1.66±0.08 m and their mean estimated height (from arm span in 2007–8) was 1.65±0.08 m. Thus FEV1% predicted would not have altered in 1999–2002 with the use of arm span.

However, we found that measured height changed significantly (p<0.001) by 2 cm between 1999–2002 (1.66±0.08 m) and 2007–8 (1.64±0.08 m), thus FEV1% predicted was significantly (p<0.001) lower when using estimated height (−67.7±24.6% vs 65.9±24.2%). As a consequence, in borderline non-COPD subjects (according to the standard of the Global Initiative for Obstructive Lung Disease (GOLD) FEV1% predicted ⩾80) we found that 6.4% (7/109) in the 1999–2002 cohort and 6.4% (7/109) in the 2007–8 cohort had values for FEV1% predicted of <80% when these were calculated using estimated height. For classifying severity we did not find that using arm span yielded a significant shifting of patients between mild and severe COPD either in 1999–2002 or 2007–8.

The results of this study suggest that: (1) current measured height may underestimate predicted FEV1; (2) use of arm span increased the proportion of patients with FEV1 <80%.


1O. Kurmi, 2S. Semple, 2P. P. Simkhada, 2W. C. S. Smith, 1J. G. Ayres. 1University of Birmingham, Birmingham, UK, 2University of Aberdeen, Aberdeen, UK

Introduction and Objectives Over half the world is exposed daily to the smoke from combustion of solid fuels. Chronic obstructive pulmonary disease (COPD) is one of the main contributors to the global burden of disease and can be caused by biomass smoke exposure. However, studies of biomass exposure and COPD show a wide range of effect sizes. The aim of this systematic review was to quantify the impact of biomass smoke on the development of COPD and define reasons for differences in the reported effect sizes.

Methods A systematic review was conducted of studies with sufficient statistical power to estimate the risk of COPD from exposure to solid fuel smoke which followed standardised criteria for the diagnosis of COPD, adjusted for smoking, were in English and contained original data. The results were pooled by fuel type and country to produce summary estimates using a random effects model. Publication bias was also estimated.

Results 4164 titles were identified which were reduced to 24 studies (11 relating to COPD, 11 to chronic bronchitis and 2 to both). Pooled estimates for the development of COPD with solid fuel use showed an odds ratio of 2.66 (95% CI 1.81 to 3.92) (fig 1) and an OR of 2.32 (95% CI 1.92 to 2.80) for chronic bronchitis. Pooled estimates by fuel type showed that exposure to wood smoke presents a greater risk of development of COPD (wood: OR 4.3; mixed biomass: OR 2.8). The findings for chronic bronchitis were similar. There was no evidence of publication bias but there was clear variation between studies which might be explained by study design, dealing with confounders, use of selected comparator groups or exposure assessment, although differential toxicity of different fuels is the most likely explanation.

Abstract P189 Figure 1

Forest plot of studies reporting COPD due to exposure to solid fuels.

Conclusion Despite heterogeneity across the selected studies, exposure to solid fuel smoke is consistently associated with COPD and chronic bronchitis. Efforts should be made to reduce exposure to solid fuel either by using cleaner fuel or relatively cleaner technology while performing domestic work.


1A. C. Darby, 2J. C. Waterhouse, 1C. Young, 3M. Aziz, 3J. Wight, 2C. G. Billings, 2C. Billings, 2V. Stevens, 1C. M. Burton, 1C. M. Barber, 1D. Morgan, 4P. D. Blanc, 1D. Fishwick. 1Health and Safety Laboratory, Buxton, UK, 2University of Sheffield, Royal Hallamshire Hospital, Sheffield, UK, 3Department of Public Health Medicine, Sheffield NHS Primary Care Trust, Sheffield, UK, 4Division of Occupational and Environmental Medicine, University of California, San Francisco, USA

Introduction and Objectives A recent BLF report has suggested there are 2.8 million people with undiagnosed chronic obstructive pulmonary disease (COPD) in the UK, the so-called “missing millions”. One of the biggest challenges for the NHS at present is how best to identify these missing cases to allow risk factor modification and disease treatment. Previous work has identified risk factors to assist population identification of such individuals,1 2 but these approaches have not been validated in UK populations.

Methods A random sample of 4000 residents in a UK city was sent a postal questionnaire enquiring about respiratory symptoms, respiratory disease, smoking and occupational exposures. A proportion had lung function measurement and the EQ-5D quality of life estimate performed. A post hoc descriptive analysis has been conducted along with more detailed logistic regression and Receiver Operating Characteristic (ROC) analysis in order to determine the optimal screening approach for airways obstruction.

Results 2001 participants returned questionnaires, of whom 572 had forced expiratory volume in 1 s (FEV1) and forced vital capacity (FVC) measured. 165 had evidence of airways obstruction (AO), defined as an FEV1/FVC ratio of <0.70, and 407 had no evidence of AO. Symptom and demographic differences between those with and without AO are shown in table 1. ROC analysis demonstrated that the previous screening questionnaire suggested by Martinez1 (AUC value 0.72) performed well for predicting AO in our local population, and that a separate model created from our data to identify AO included other significant predictors including exposure to occupational factors including steel processing work (AUC value 0.77).

Abstract P190 Table 1

Conclusions Despite differences in the demographics of the populations studied and the questions asked, a previously suggested population screening questionnaire for the presence of AO performed well in a local UK-based population. In addition, a locally-derived questionnaire from UK-based data has been produced to assist identifying those with as yet undiagnosed AO.



1A. C. Darby, 1J. C. Waterhouse, 1C. M. Barber, 1C. M. Burton, 2J. Wight, 2P. D. Blanc, 1V. Stevens, 1C. Billings, 1C. G. Billings, 1D. Fishwick. 1Centre for Workplace Health, University of Sheffield, Royal Hallamshire Hospital, Sheffield, UK, 2Department of Public Health Medicine, Sheffield NHS Primary Care Trust, Sheffield, UK

Introduction and Objectives Chronic obstructive pulmonary disease (COPD) causes a substantial burden to both individual patients and to society. While smoking is recognised as the predominant cause, other risk factors are identified. Consistent estimates suggest occupational exposure to dust contributes 15% to the overall disease burden. Many of these data are derived, however, from non-UK based studies. We aim to address this gap in the literature with an epidemiological study of COPD in a UK city.

Methods We conducted a questionnaire-based study of a random sample of 4000 people aged over 55. In addition, an enrichment sample of individuals with likely COPD was identified from a secondary care physiology department. The study recorded data on respiratory symptoms, self-reported respiratory diagnoses, smoking and a detailed occupational history including generic exposure to vapours, gases, dusts and fumes (VGDF). A proportion of respondents had spirometry performed. Regression analysis (including age and sex) estimated the association of ever having smoked and ever having VGDF exposure with reported doctor diagnosed COPD, using different definitions of COPD.

Results Having previously presented the cut-off point analysis1 excluding the enrichment population, these data represent the final analysis including the enrichment sample. 2061 completed questionnaires were received (2001 postal, 60 enrichment), 1935 of which had complete data for smoking and VGDF exposures and were used for this analysis. The mean age was 69.1 years and 49.5% were male. 115 reported a previous doctor diagnosis of COPD, 114 chronic bronchitis, 80 emphysema and 306 asthma. 57.0% had ever smoked and 49.2% reported ever being exposed to VGDF at work. Odds ratios and population attributable risk (PAR) for COPD from ever being exposed to VGDF or from ever smoking are shown in table 1.

Abstract P191 Table 1

Conclusions This study has documented high UK-based PAR% values for the occupational contributions to COPD. These values fall when the diagnosis is based on spirometric values, confirming the variable well-documented relationship between reported symptoms and lung function abnormalities. Even allowing for these issues, these data suggest that, in this UK population, ever exposure to VGDF has contributed significantly to the burden of COPD.



1D. M. Neville, 2S. M. S. Smith, 2S. L. Elkin. 1Imperial College London School of Medicine, London, UK, 2Imperial College Healthcare NHS Trust, London, UK

Introduction Chronic obstructive pulmonary disease (COPD) and cardiovascular disease share traditional risk factors including smoking. In some patients these diseases co-exist. We hypothesised that (1) spirometry could be performed in cardiology outpatient clinics and (2) that it might identify previously undiagnosed COPD.

Methods Over a 6-week period, patients aged 35 years and over attending cardiology outpatient clinics were invited to perform simple spirometry and to complete a respiratory questionnaire. This included smoking history, respiratory symptoms and previously diagnosed medical conditions.

Results A total of 163 patients (107 male, mean age 63.83±13.25 years) performed spirometry. 68 patients declined to participate in the study mostly due to time pressure. Of the participants, 79 (49%) had been diagnosed with ischaemic heart disease (IHD) and 4 had a previous diagnosis of COPD. Spirometry tests demonstrated airflow obstruction in 43 (26.4%) patients of whom 30 (69.8%) were male. In current smokers, 14 (58.3%) showed airflow obstruction. The patients with airflow obstruction had smoked a significantly greater number of mean pack years compared with those recording normal spirometry (p = 0.047). A significantly greater number of patients with airflow obstruction experienced severe breathlessness (of level 4 on the MRC dyspnoea scale) compared to those with normal spirometry (p = 0.015). A restrictive defect was found in 34 (20.9%) patients. Of these, 25 (73.5%) were overweight (BMI >25).

Conclusions Performing spirometry as a screening tool in a population with a high risk for COPD is feasible. Within cardiology outpatient clinics, spirometry detected a large proportion of patients with airflow obstruction and possible COPD. This method of screening has the potential to benefit a patient in terms of both their cardiovascular and respiratory health.


1N. Nzekwue, 1S. Bremner, 1S. Taylor, 2N. Pursey, 3M. Roberts, 2R. Stone. 1Department of Health Sciences, Queen Mary University of London, London, UK, 2The Royal College of Physicians, London, UK, 3Queen Mary University of London and The Royal College of Physicians, London, UK

Introduction and Objectives Organisation of COPD care and self-management have been thoroughly investigated using tools such as the National COPD Audit. Literature highlights burdens caused by COPD that help inform how to organise care. The government and NHS advocate self-management, citing well evidenced and potentialised benefits. What shapes one’s ability to aptly self-manage one’s own COPD? One of the main objectives of this study was to identify which factors are related to skilled self-management. Coupled with the 2008 National COPD Audit’s assessment of current service use and demands, the overall aim was to propose recommendations for those setting up a COPD service. This service would use necessary improvements to any level of care while integrating factors identified by this study that facilitate the success of already evidence based self-management techniques and initiatives.

Method This study used raw data from the 2008 National COPD Audit. Responses to specific questions were combined forming characteristic variables (condition severity, support and hospital attendance) and attribute variables that constitute certain aspects of self-management (understanding, monitoring, service use and behaviour). Logistic regression was performed to estimate the association between individual characteristic variables including age and sex and adequate self-management performance in each attribute variable. Correlations were made between characteristic variables and themes related to COPD care generated from the audit.

Results There were 2864 patient respondents. Age and gender were not significantly associated with self-management performance. Those living alone had 4–46% [0.72 (0.54–0.96)] less odds of having adequate understanding than those living with someone. First-time hospital patients had 32–66% (0.48 (95% CI 0.34 to 0.68)) less odds of having adequate understanding and 48–74% (0.37 (0.26 to 0.52)) less odds of having adequate monitoring performance than those admitted frequently. Compared with non-exacerbators, those who experienced exacerbations had 8–58% (1.30 (1.08 to 1.58)) more odds of having adequate understanding, and almost double the odds of having adequate monitoring (1.77 (1.49 to 2.11)) and service use performance (1.85 (1.32 to 2.60)) (fig 1).

Abstract P193 Figure 1

Monitoring performance.

Conclusion An improved COPD service will incorporate self-management. Increasing contact time with community and primary care members (most popular audit themes) will improve patients’ ability to self-manage their COPD.


J. L. Kelly, P. Mehta, D. Shrikrishna, D. Cramer, M. I. Polkey, N. S. Hopkinson. Muscle Laboratory, Royal Brompton Hospital, London, UK

Introduction Chronic obstructive pulmonary disease (COPD) is a progressive lung disease which has an impact on airways, lung parenchyma and the pulmonary vasculature. Damage to these different compartments is likely to impact differently on different lung function parameters. COPD is conventionally classified into stages according to the degree of airflow obstruction expressed as percentage predicted forced expiratory volume in 1 s (FEV1).

Aims We wished to evaluate the relationship between GOLD stage of COPD and absolute rate of decline of lung function over time.

Methods Patients were selected from our COPD research audit database if they had had two sets of complete lung function tests performed at least 3 years apart. We excluded patients who had had a lung volume reduction procedure or who had α1-antitrypsin deficiency. Annual rates of decline in FEV1, lung carbon monoxide transfer factor (Tlcoc), carbon monoxide transfer coefficient (Kcoc), total lung capacity (TLC) and residual value (RV) were calculated both for absolute change and for change in percentage predicted values.

Results Annual decline in FEV1 and Tlcoc varied significantly with disease stage (ANOVA, p = 0.0025 and p = 0.0494). Decline in FEV1 was most rapid in those with the least severe disease (ANOVA, p = 0.002) with a change of −52 ml/year in stage I compared with +1 ml/year in stage IV. Gas transfer, however, appeared to be static in stage I. The largest decline in Tlcoc was in stage II disease, decreasing sequentially as GOLD stage increased. However, in contrast to FEV1, there was an appreciable decline even in GOLD stage IV patients (fig 1). Change in RV and TLC did not differ according to GOLD stage.

Abstract P194 Figure 1

GOLD stage versus annual change in absolute lung function.

Conclusion In patients with the most severe disease, the decline in FEV1 appears to plateau. Gas transfer decline is most prominent in GOLD stage II disease but continues to decline appreciably in stage IV.

View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.