Sage Journals: Discover world-class research

Abstract

Background: Individuals with Preserved Ratio Impaired Spirometry (PRISm), defined as FEV₁/FVC ≥0.7 and FEV1 <80% predicted, are at higher risk of developing COPD. However, data for Australian adults are limited. We aimed to describe prevalence of PRISm and its relationship with clinical characteristics in Australia. Method: Data from the Burden of Lung Disease (BOLD) Australia study of randomly selected adults aged ≥40 years from six sites was classified into airflow limitation, PRISm, or normal spirometry groups. Demographic, clinical characteristics, and lung function were compared between groups. Results: Of the study sample (n = 3518), 387 (11%) had PRISm, 549 (15.6%) had airflow limitation, and 2582 (73.4%) had normal spirometry. PRISm was more common in Indigenous Australian adults. Adults with PRISm had more frequent respiratory symptoms, more comorbidities, greater health burden and poorer quality of life than those with normal spirometry. Pre- and post-bronchodilator FEV₁ and FVC were lower in adults with PRISm than those with airflow limitation. Adults with PRISm were less likely to use respiratory medicine than those with airflow limitation (OR = 0.56, 95% CI 0.38–0.81). Conclusions: PRISm was present in 11% of adults in this study and they had similar respiratory symptoms and health burden as adults with airflow limitation.

Keywords

Chronic obstructive pulmonary disease preserved ratio impaired spirometry symptoms spirometry burden of obstructive lung disease

Introduction

Chronic obstructive pulmonary disease (COPD) is one of the most common chronic respiratory diseases, that is associated with a high medical and economic burden for patients and healthcare systems.¹ According to the 2024 Global initiative for Obstructive Lung Disease (GOLD), making the diagnosis of COPD in a patient with respiratory symptoms and/or a toxic exposure such as tobacco smoke requires post-bronchodilator (BD) spirometry demonstrating a ratio of forced expiratory volume in the first second (FEV₁) to forced vital capacity (FVC) less than 0.7.² COPD patients are often under-diagnosed or diagnosed only in advanced stages of the disease.³ In Australia, the Burden of Obstructive Lung Disease study (BOLD) found that 6.9% of adults aged ≥40 years had persistent airflow limitation, without a diagnosis of COPD.⁴ Early diagnosis, avoidance of risk factors and subsequent management of COPD could potentially slow disease progression and reduce the impact on patient health and well-being.⁵ For this reason, it is important to identify individuals who may eventually develop airflow limitation consistent with a diagnosis of COPD.

In 2001, the GOLD report proposed COPD stage 0 defined by symptoms among smokers with normal spirometry.⁶ However, this category was later abandoned as the GOLD stage 0 only identified a small proportion of individuals who ultimately progressed to COPD and was not demonstrated to be an effective strategy to identify individuals who will eventually develop COPD.^5,7 The GOLD 2023 report included the term ‘PRISm’ (Preserved Ratio Impaired Spirometry),⁸ which had been proposed in 2014 to describe individuals with post-BD FEV₁/FVC ≥0.7 and FEV₁ <80% predicted. This spirometric pattern was often previously called ‘restrictive’ or ‘non-specific’.^5,8,9 Inclusion of PRISm in GOLD 2023 was based on the observation that some individuals with PRISm are at risk of rapid progression to COPD over time.^9–11

The prevalence of PRISm in epidemiological studies has ranged from 7.1% to 20.3%.¹² Although there are some limitations of PRISm, including that it is calculated from two binary variables, which further increases likely error,¹³ some studies found that PRISm was associated with respiratory symptoms, adverse cardiovascular and respiratory outcomes, and increased all-cause mortality.^9,14,15 To our knowledge, there have not been any studies comparing clinical characteristics between individuals with PRISm, normal spirometry, and airflow limitation in the general Australian population.

In the BOLD Australia study, self-reported symptoms and clinical test information, including standardised quality controlled post-BD spirometry results, were collected from a large number of Australian adults. Therefore, we are able to use these results to define individuals with PRISm among the BOLD Australia population, and to understand PRISm more comprehensively in a large population of Australian adults. Our study aimed to describe prevalence of PRISm, and to investigate the characteristics associated with PRISm in Australian adults aged 40 years and over.

Methods

Study population

BOLD Australia was a cross-sectional study conducted between 2006 and 2012 of individuals aged ≥40 years from six study sites across Australia, including Sydney, rural New South Wales, Melbourne, Tasmania (Hobart and Launceston) and Busselton and Broome in Western Australia. Participants were selected from the Australian Electoral Roll for all sites except, in Western Australia, participants at the Kimberley site were recruited from a household census in Broome and from local aboriginal communities within the region, while at the Busselton site participants were randomly recruited from the Busselton Health Study. Participants who could not be contacted, were institutionalised, or were younger than 40 years old were excluded. The detailed information for the study design and sample selection was described previously.^16,17 As part of the international BOLD study, some results from the Sydney site including individuals with PRISm (described at the time as restricted spirometry) have been published with data from centres in 14 countries.¹⁸ This current report includes data from both Sydney and the five other sites in Australia. The participants included in this analysis provided both core questionnaire data and acceptable post-BD spirometry data.

Minimal data were collected from those who did not complete the full protocol and included 6 key questions about smoking, respiratory disorders and additional comorbidities (Supplemental information S1).

Study questionnaire

The BOLD core questionnaires were utilised across all sites. The questionnaires covered details of demographics, body mass index (BMI), smoking status, occupational exposures, respiratory symptoms, treatments, comorbidities, time lost from work or daily activities, and healthcare utilisation.^16,19 Demographic characteristics included age, sex, ethnicity, education level, and socioeconomic status. Socioeconomic status was reported using quintiles of Socio-Economic Indexes for Areas (SEIFA), with 1 being the “most disadvantaged” and 5 being the “least disadvantaged”.²⁰

Respiratory symptoms included cough, phlegm, and breathlessness. Breathlessness was assessed by the modified Medical Research Council (mMRC) dyspnoea scale.²¹ As in previous studies, ‘clinically important breathlessness’ was defined as an mMRC dyspnoea grade ≥2.²² Treatments included respiratory medication use and lung volume reduction surgery (LVRS). Specific comorbidities included current asthma, heart disease, hypertension, diabetes, lung cancer and stroke. Healthcare utilisation reported were visits to a general practitioner (GP) and hospitalisations in the last 12 months due to breathing problems.

Domain scores of the Short-Form (SF)-12 questionnaire were used to report health-related quality of life (HrQoL).²³ Because the SF-12 has not been validated in Indigenous Australians, the BOLD Australia survey collected answers only to the first question of the SF-12 questionnaire (‘In general, would you say your health is …’) among Indigenous participants. Therefore, we defined self-reported general health status for all participants by using the result of the first question (SF-1) in the SF-12. Participants who answered excellent, very good, and good were defined as having good or above general health. The physical and mental component summary scores (PCS, MCS) of the SF-12 were calculated using US norms because there were no Australian norms available.²³

Spirometry

Lung function was assessed using the EasyOne spirometer (ndd Medizintechnik, Zürich, Switzerland), according to 2005 American Thoracic Society (ATS)/European Respiratory Society spirometry (ERS) standard,²⁴ as previously described.^16,19 All spirograms were quality control reviewed and evaluated for acceptability by a senior respiratory scientist.¹⁶ The highest recorded FEV₁ and FVC in acceptable manoeuvres were collected, these did not need to occur in the same manoeuvre.²⁴ Airflow limitation was defined as FEV₁/FVC <0.7 by post-BD spirometry, based on then-current GOLD guideline.²⁵ For the present analyses, PRISm was defined as post-BD FEV₁/FVC ≥0.7 and FEV₁ <80% predicted. “Normal spirometry”, i.e. no airflow limitation and no PRISm, was defined as post-BD FEV₁/FVC ≥0.7 and FEV₁ ≥80% predicted (Table 1). The percent predicted FEV₁ and FVC were defined by using NHANES and Global Lung Function Initiative (GLI) reference equations,^26,27 and lower limit of normal (LLN) for FEV₁ were defined using GLI reference equations²⁷ for Caucasians, consistent with previous BOLD Australia publications and new ATS guidelines.²⁸ Although a recent ATS/ERS Technical Standard proposed new BD responsiveness criteria,²⁸ a positive BD response was defined in this analysis as having an increase in FEV₁ (or FVC) of ≥200 mL and ≥12% from baseline,²⁹ for consistency with previous BOLD Australia publications.

Table 1.

Spirometric diagnostic criteria for PRISm, airflow limitation and normal spirometry.

Normal spirometry	Post-BD FEV₁/FVC ≥0.7 and FEV₁ ≥80% predicted
PRISm	Post-BD FEV₁/FVC ≥0.7 and FEV₁ <80% predicted
Airflow limitation	Post-BD FEV₁/FVC <0.7

Statistical analysis

All statistical analyses were completed in SAS statistical software version 9.4. (SAS Institute Inc., Cary, NC). The study population was grouped according to spirometry results (airflow limitation, PRISm, or normal spirometry). Participant characteristics were tabulated by groups. For categorical variables, data are presented as numbers with proportions, and for continuous variables, data are presented as mean ± standard deviation (SD) or median with interquartile range [IQR]. Differences between individuals, with PRISm, normal spirometry, or airflow limitation were evaluated using Chi-squared tests for categorical variables, and analysis of variance for continuous variables that followed normal distributions. Wilcoxon Mann-Whiney U-tests were used to compare continuous variables that did not follow normal distributions.

Multivariate logistic regression models were used to estimate adjusted odds ratios (OR) with 95% confidence intervals (CI). We used Directed Acyclic Graphs (DAGs) to identify potential confounders,³⁰ with examples included in the supplement (Figures S1-3). Age, sex, being Indigenous Australians, BMI, smoking and socioeconomic status were identified as potential confounders because they could open a back-door pathway association between ‘exposure’ (spirometry label) and ‘outcome’ (clinical feature) and therefore should be adjusted for in the analyses.³⁰

Results

Prevalence and correlates of PRISm

Overall, 3518 participants performed acceptable pre- and post-BD spirometry, thus were included in this analysis. Compared to those who only provided minimal information, we found that the analysed sample was younger and more likely to have a self-reported diagnosis of respiratory disease.¹⁶ Of the 3518 participants with evaluable post-BD spirometric data, 387 (11.0%) participants were classified as having PRISm, 549 (15.6%) participants had airflow limitation, and 2582 (73.4%) participants had normal spirometry.

Characteristics of participants according to lung function category

Table 2 describes the demographic characteristics. Participants with PRISm were younger and more likely to be female or a non-smoker than those with airflow limitation, and they were much more likely than those with either normal spirometry or airflow limitation to be Indigenous Australians, living in socio-economically disadvantaged areas, or to be obese.

Table 2.

Demographic characteristics of adults ≥40 years with PRISm compared to those with airflow limitation and normal spirometry.

Characteristic	Normal spirometry (n = 2582)	PRISm (n = 387)	Airflow limitation (n = 549)	Total (n = 3518)	p-values
Characteristic	Normal spirometry (n = 2582)	PRISm (n = 387)	Airflow limitation (n = 549)	Total (n = 3518)	PRISm versus normal spirometry	PRISm versus airflow limitation
Mean age, years (SD)	57.3 (11.1)	55.9 (11.0)	66.3 (11.7)	58.5 (11.7)	p = 0.02	p < 0.0001
Age group					p = 0.04	p < 0.0001
40–49 years	795 (30.8)	135 (34.9)	56 (10.2)	986 (28.0)
50–59 years	810 (31.4)	128 (33.1)	107 (19.5)	1045 (29.7)
60–69 years	578 (22.4)	70 (18.1)	155 (28.2)	803 (22.8)
70–79 years	210 (12.4)	42 (10.9)	169 (30.8)	531 (15.1)
≥80 years	79 (3.1)	12 (3.1)	62 (11.3)	153 (4.3)
Sex, female	1377 (53.3)	204 (52.7)	238 (43.4)	1819 (51.7)	p = 0.82	p = 0.005
Ethnicity					p < 0.0001	p < 0.0001
N with data	2581	387	549	3517
Caucasian	2410 (93.4)	186 (48.1)	510 (92.9)	3106 (88.3)
Indigenous	64 (2.5)	172 (44.4)	24 (4.4)	260 (7.4)
Other	107 (4.2)	29 (7.5)	15 (2.7)	151 (4.3)
Smoking status					p < 0.0001	p = 0.10
Never smoked	1315 (50.9)	167 (43.2)	174 (31.7)	1656 (47.1)
Former smoker	1039 (40.2)	126 (32.6)	274 (49.9)	1439 (40.9)
Current smoker	228 (8.8)	94 (24.3)	101 (18.4)	423 (12.0)
Packs-year for ever smokers, mean (SD)	12.0 (19.5)	15.3 (28.1)	27.0 (39.0)	14.4 (25.0)	p = 0.016	p < 0.0001
BMI, kg/m²
N with data	2916	287	204	3447
Mean BMI, (SD) kg/m²	27.9 (4.8)	29.4 (6.4)	27.4 (5.2)	28 (5.1)	p < 0.0001	p < 0.0001
BMI status					p = 0.05	p = 0.49
<18.5 kg/m²	14 (0.6)	7 (1.8)	7 (1.3)	28 (0.8)
18.5≤ BMI <25.0 kg/m²	709 (27.9)	91 (24.2)	188 (35.4)	988 (28.7)
25.0≤ BMI <30.0 kg/m²	1079 (42.5)	107 (28.5)	191 (36.0)	1377 (39.9)
≥30.0 kg/m²	738 (29.1)	171 (45.6)	145 (27.3)	1054 (30.6)
Highest level of schooling					p = 0.31	p = 0.008
N with data	2178	181	456	2815
Primary education or less	33 (1.5)	6 (3.3)	23 (5.0)	62 (2.2)
High school	750 (34.4)	67 (37.0)	215 (47.2)	1032 (36.7)
Technical or further education	782 (35.9)	62 (34.3)	134 (29.4)	978 (34.7)
University	613 (28.2)	46 (25.4)	84 (18.4)	743 (26.4)
Work in a dusty job (≥1 year)	824 (31.9)	167 (43.2)	230 (41.9)	1221 (34.7)	p < 0.0001	p = 0.70
Socioeconomic status (SEIFA)					p < 0.0001	p < 0.0001
N with data	2555	260	538	3353
Quintile 1 (most disadvantaged)	592 (23.2)	108 (41.5)	139 (25.8)	839 (25.0)
Quintile 2	334 (13.1)	34 (13.1)	79 (14.7)	447 (13.3)
Quintile 3	839 (32.8)	74 (28.5)	190 (35.3)	1103 (32.9)
Quintile 4	265 (10.4)	16 (6.2)	44 (8.2)	325 (9.7)
Quintile 5 (least disadvantaged)	525 (20.6)	28 (10.8)	86 (16.0)	639 (19.1)

Notes: For percentages, the denominator is given when different from the total number of patients (N with data [excluding „unknown”]); data are presented as n (%), unless stated otherwise. Lung function category definitions: PRISm (FEV₁/FVC ≥0.7 and FEV₁ <80% predicted), normal spirometry (FEV₁/FVC ≥0.7 and FEV₁ >80% predicted) and airflow limitation (FEV₁/FVC <0.7 by post-BD spirometry). p-values for multilevel variables were the Chi-square for trend test.

Abbreviations: PRISm: preserved ratio impaired spirometry; SD: standard deviation; BMI: body mass index; SEIFA: socio-economic indexes for areas.

Current asthma and heart disease were most prevalent in participants with airflow limitation, while hypertension and diabetes were most prevalent in participants with PRISm (Table 3). The proportions of participants reporting cough or phlegm on most days for at least 3 months, and of those with clinically important breathlessness (mMRC ≥2), were similar in those with airflow limitation and PRISm (Table 3). A total of 32.6% of the airflow limitation group reported using respiratory medications, compared with 20.4% of those with PRISm, and 12.8% of those with normal spirometry (Table 4).

Table 3.

Comorbidities and symptoms of adults ≥40 years with PRISm compared to those with airflow limitation and normal spirometry.

Characteristic	Normal spirometry (n = 2582	PRISm (n = 387)	Airflow limitation (n = 549)	Total (n = 3518)	Adjusted OR (95% CI), p value
Characteristic	Normal spirometry (n = 2582	PRISm (n = 387)	Airflow limitation (n = 549)	Total (n = 3518)	PRISm versus normal spirometry	PRISm versus airflow Limitation
Asthma
N with data	2580	386	549	3515
Ever diagnosis with asthma	459 (17.8)	94 (24.3)	177 (32.2)	730 (26.2)	1.80 (1.30–2.46), p = 0.0003	0.54 (0.38–0.78), p = 0.001
Current asthma	222 (8.6)	59 (15.3)	132 (24.0)	413 (11.7)	2.02 (1.35–2.96), p = 0.0004	0.43 (0.28–0.66), p = 0.0001
Comorbidities
Heart disease	174 (6.7)	58 (15.0)	100 (18.2)	332 (9.4)	2.56 (1.70–3.81), p < 0.0001	1.79 (1.13–2.79), p = 0.01
Hypertension	789 (30.6)	192 (49.6)	232 (42.3)	1213 (34.5)	1.93 (1.44–2.60), p < 0.0001	1.90 (1.35–2.69), p = 0.0003
Diabetes	175 (6.8)	127 (32.8)	65 (11.8)	367 (10.4)	3.07 (2.12–4.39), p < 0.0001	2.96 (1.88–4.66), p < 0.0001
Lung cancer	4 (0.2)	6 (1.6)	9 (1.6)	19 (0.5)	13.2 (3.16–58.1), p = 0.0003	3.02 (0.79–10.7), p = 0.09
Stroke	41 (1.6)	14 (3.6)	27 (4.9)	82 (2.3)	1.85 (0.85–3.73), p = 0.10	1.16 (0.51–2.50), p = 0.71
Cough
N with data	2581	384	549	3514
Cough on most days for ≥3 months	211 (8.2)	54 (14.1)	90 (16.4)	355 (10.1)	1.54 (0.99–2.34), p = 0.05	0.75 (0.46–1.19), p = 0.23
Phlegm
N with data	2580	382	549	3511
Phlegm on most days for ≥3 months	145 (5.6)	55 (11.5)	72 (13.1)	261 (7.4)	2.15 (1.35–3.32), p = 0.0008	1.09 (0.66–1.78), p = 0.73
Clinically important breathlessness
N with data	2494	331	493	3318
mMRC ≥2	110 (4.4)	62 (18.7)	80 (16.2)	252 (7.6)	3.40 (2.18–5.20), p < 0.0001	1.01 (0.62–1.63), p = 0.96

Notes: For percentages, the denominator is given when different from the total number of patients (N with data [excluding „unknown”]); data are presented as n (%), unless stated otherwise; adjusted for age, sex, being Indigenous Australians, BMI status, smoking status and socioeconomic status. Lung function category definitions: PRISm (FEV₁/FVC ≥70% and FEV₁ <80% predicted), normal spirometry (FEV₁/FVC ≥70% and FEV1 >80% predicted) and airflow limitation (FEV₁/FVC <70% by post-BD spirometry).

Abbreviations: PRISm: preserved ratio impaired spirometry; mMRC: modified medical research council; OR: odds ratio; CI: confidence intervals.

Table 4.

Treatments reported by adults ≥40 years with PRISm compared to those with airflow limitation and normal spirometry.

Characteristic	Normal spirometry (n = 2582)	PRISm (n = 387)	Airflow limitation (n = 549)	Total (n = 3518)	Adjusted OR (95% CI), p value
Characteristic	Normal spirometry (n = 2582)	PRISm (n = 387)	Airflow limitation (n = 549)	Total (n = 3518)	PRISm versus normal spirometry	PRISm versus airflow limitation
Any respiratory medication used	330 (12.8)	79 (20.4)	179 (32.6)	588 (16.7)	2.05 (1.46–2.86), p < 0.0001	0.56 (0.38–0.81), p = 0.002
Muscarinic antagonists (Ipratropium or tiotropium)	24 (0.9)	7 (1.8)	40 (7.3)	71 (2.0)	2.83 (1.07–6.60), p = 0.02	0.65 (0.25–1.49), p = 0.34
SABA	237 (9.2)	65 (16.8)	132 (24.0)	434 (12.3)	2.12 (1.45–3.06), p < 0.0001	0.56 (0.37–0.84), p = 0.005
LABA	107 (4.1)	32 (8.3)	97 (17.7)	236 (6.7)	2.89 (1.79–4.56), p < 0.0001	0.56 (0.34–0.91), p = 0.02
ICS	179 (6.9)	44 (11.4)	123 (22.4)	349 (9.8)	2.29 (1.51–3.40), p < 0.0001	0.54 (0.34–0.83), p = 0.006
OCS	18 (0.7)	1 (0.3)	5 (0.9)	24 (0.7)	0.67 (0.04–3.86), p = 0.72	0.19 (0.01–1.37), p = 0.15
History of LVRS	9 (0.4)	3 (0.8)	11 (2.0)	23 (0.7)	4.85 (1.03–17.4), p = 0.02	1.07 (0.22–3.89), p = 0.92

Notes: For percentages, the denominator is given when different from the total number of patients (N with data [excluding „unknown”]); data are presented as n (%), unless stated otherwise; adjusted for age, sex, being Indigenous Australians, BMI status, smoking status and socioeconomic status. Lung function category definitions: PRISm (FEV₁/FVC ≥70% and FEV₁ <80% predicted), normal spirometry (FEV₁/FVC ≥70% and FEV₁ >80% predicted) and airflow limitation (FEV₁/FVC <70% by post-BD spirometry).

Abbreviations: PRISm: preserved ratio impaired spirometry; LVRS: Lung volume reduction surgery; SABA: short-acting β2-agonist; LABA: long-acting β2-agonist; ICS: inhaled corticosteroid; OCS: oral corticosteroid; OR: odds ratio; CI: confidence intervals.

Spirometry

Mean pre- and post-BD FEV₁ and FVC were lowest in the PRISm group (Table 5). Mean (SD) post-BD percent predicted FEV₁ (NHANES) was 101.7% (11.7) in participants with normal spirometry compared with 81.4% (20.2) in those with airflow limitation and 71.6% (10.1) in those with PRISm; mean (SD) post-BD percent predicted FVC (NHANES) was 98.5% (11.3) in participants with normal spirometry compared with 97.4% (18.9) in those with airflow limitation and 69.5% (9.3) in those with PRISm (Table 5). The mean post-BD FEV₁/FVC was similar in PRISm and normal spirometry groups and, by definition, was significantly lower in those with airflow limitation (Table 5). In addition, a higher proportion of participants in the airflow limitation group had significant BD responsiveness than in the PRISm group (Table 5).

Table 5.

Spirometry results in adults ≥40 years with PRISm compared to those with airflow limitation and normal spirometry.

Characteristic	Normal spirometry (n = 2582)	PRISm (n = 387)	Airflow limitation (n = 549)	Total (n = 3518)	p-values
Characteristic	Normal spirometry (n = 2582)	PRISm (n = 387)	Airflow limitation (n = 549)	Total (n = 3518)	PRISm versus normal spirometry	PRISm versus airflow limitation
Mean pre-BD spirometry, (SD)
FEV₁, % pred. (NHANES)^a	99.1 (12.3)	70.4 (10.5)	76.9 (21.2)	92.5 (17.8)	p < 0.0001	p = 0.01
FVC, % pred. (NHANES)^a	99.3 (11.9)	71.3 (10.5)	93.5 (20.4)	95.3 (16.0)	p < 0.0001	p < 0.0001
FEV₁, % pred. (GLI)^b	99.3 (11.9)	70.5 (10.4)	75.8 (20.5)	92.5 (17.7)	p < 0.0001	p = 0.08
FVC, % pred. (GLI)^b	101.5 (11.9)	72.8 (10.6)	95.2 (20.5)	97.4 (16.2)	p < 0.0001	p < 0.0001
Mean post-BD spirometry, (SD)
FEV₁, % pred. (NHANES)	101.7 (11.7)	71.6 (10.1)	81.4 (20.2)	95.2 (17.3)	p < 0.0001	p < 0.0001
FVC, % pred. (NHANES)	98.5 (11.3)	69.5 (9.3)	97.4 (18.9)	95.1 (15.5)	p < 0.0001	p < 0.0001
FEV₁, % pred. (GLI)^a	101.9 (11.3)	71.7 (10.0)	80.2 (19.5)	95.2 (17.1)	p < 0.0001	p < 0.0001
FVC, % pred. (GLI)^a	100.7 (11.4)	71.0 (9.5)	99.1 (19.1)	97.2 (15.7)	p < 0.0001	p < 0.0001
Mean post-BD FEV₁/FVC, (SD)	0.80 (0.05)	0.80 (0.05)	0.62 (0.08)	0.77 (0.08)	p = 0.22	p < 0.0001
Mean post-BD FEV1/FVC by LLN^a, (SD) (GLI)	0.67 (0.03)	0.67 (0.03)	0.65 (0.03)	0.67 (0.03)	p = 0.45	p = 0.03

Characteristic	Normal spirometry (n = 2582)	PRISm (n = 387)	Airflow limitation (n = 549)	Total (n = 3518)	OR (95% CI), p-values
Characteristic	Normal spirometry (n = 2582)	PRISm (n = 387)	Airflow limitation (n = 549)	Total (n = 3518)	PRISm versus normal spirometry	PRISm versus airflow limitation
BD responsiveness
Change of ≥12% and ≥200 mL in FEV1 and/or FVC from pre-BD value	72 (2.8)	13 (3.4)	92 (16.8)	177 (5.0)	1.35 (0.65-2.56), p = 0.38	0.17 (0.08-0.32), p < 0.0001

Notes: Data are presented as n (%), unless stated otherwise; adjusted for age, sex, being indigenous Australians, BMI status, smoking status and socioeconomic status. Lung function category definitions: PRISm (FEV₁/FVC ≥70% and FEV₁ <80% predicted), normal spirometry (FEV₁/FVC ≥70% and FEV₁ >80% predicted) and airflow limitation (FEV₁/FVC <70% by post-BD spirometry).

Abbreviations: BD: bronchodilator; SD: standard deviation; FEV₁: forced expiratory volume in 1 second; FVC: forced vital capacity; GLI: Global Lung Initiative; OR: odds ratio; CI: confidence intervals LLN: the lower limit of normal.

^aDoes not include 1 participant due to missing data.

^bDoes not include 2 participants due to missing data.

Health burden and HrQoL

The proportions of participants with PRISm reporting absences from work or daily activities (8.3%) and healthcare utilisation related to respiratory conditions (6.5% for GP visits and 3.1% for hospitalisations) were similar in the PRISm group and airflow limitation group, but significantly higher compared to participants with normal spirometry (Table 6). Regarding HrQoL, the proportions of participants reporting good or above general health were similar in the PRISm group and airflow limitation group, but significantly lower than in the normal spirometry group. The Physical Component Score (PCS) was not significantly different between the PRISm and airflow limitation groups but was significantly lower than in the normal spirometry group (Table 6).

Table 6.

Health burden and health-related quality of life (SF-12) of adults ≥40 years with PRISm compared to those with airflow limitation and normal spirometry.

Characteristic	Normal spirometry (n = 2582)	PRISm (n = 387)	Airflow limitation (n = 549)	Total (n = 3518)	Adjusted OR (95% CI), p-value
Characteristic	Normal spirometry (n = 2582)	PRISm (n = 387)	Airflow limitation (n = 549)	Total (n = 3518)	PRISm versus normal spirometry	PRISm versus airflow limitation
Lost time off work or social activities^a
≥1 in the past 12 months^b	100 (3.9)	32 (8.3)	44 (8.0)	176 (5.0)	2.72 (1.61–4.47), p = 0.0001	1.00 (0.55–1.77), p = 1.00
Healthcare utilisation^a
≥1 GP visit in the past 12 months^b	77 (3.0)	25 (6.5)	30 (5.5)	132 (3.8)	2.64 (1.43–4.62), p = 0.001	1.03 (0.52–1.99), p = 0.94
≥1 hospitalisations in the past 12 months^b	9 (0.4)	12 (3.1)	8 (1.5)	29 (0.8)	8.15 (2.71–24.2), p = 0.0001	2.07 (0.66–6.40), p = 0.20
General health
Good or above	2366 (91.6)	302 (78.0)	432 (78.7)	3100 (88.1)	0.50 (0.35–0.72), p = 0.0001	1.18 (0.78–1.78), p = 0.43
Mental and physical scores^b
N with data	2174	180	455	2809
MCS, Median (IQR)	54.8 (9.0)	54.2 (11.5)	55.2 (10.5)	54.8 (9.3)	p = 0.08	p = 0.46
PCS, Median (IQR)	53.9 (8.6)	49.7 (13.9)	49.8 (13.0)	53.0 (9.9)	p < 0.0001	p = 0.90

Abbreviations: PRISm: preserved ratio impaired spirometry; SF-12, short-form-12; PCS, physical component summary scores; MCS, mental component summary scores; OR: odds ratio; CI: confidence intervals; IQR, interquartile range.

^aDescribed as ‘When breathing problems got so bad that they interfered with usual daily activities or caused participants to miss work’; Mental and physical scores did not include Indigenous Australians.

^bDo not include all observations due to missing data.

Discussion

We have shown for the first time that the prevalence of PRISm among Australian adults aged ≥40 years enrolled in the BOLD Australia study was 11.00% (95% confidence interval (CI) = 9.97%–12.03%). Participants with PRISm had similar respiratory symptoms, similar or even more comorbidities, similar health burden, HrQoL, and lower pre- and post- BD spirometry results than those with airflow limitation but were less likely to report taking respiratory treatment. There were some striking differences in demographic characteristics compared with participants with either normal spirometry or airflow limitation. These findings have important implications for clinical practice and further research.

Individuals with PRISm have reduced lung function, but do not meet the spirometric definition of COPD.^2,9 They may often previously have been described as having ‘restrictive’ or ‘non-specific’ spirometry⁹; indeed, PRISm includes people with restricted total lung capacity due to pulmonary causes (e.g. sarcoidosis, pulmonary fibrosis) or extra-pulmonary causes (e.g. obesity, kyphoscoliosis, growth retardation). In our analysis, those with PRISm were younger, more likely to be female, and had smoked less compared to participants with airflow limitation, similar to previous studies.^9,15 We found that 44% of participants with PRISm were Indigenous Australians, which was significantly higher than in the airflow limitation or normal spirometry groups. Participants with PRISm were much more likely to be Indigenous Australians than those with either normal spirometry or airflow limitation, after adjusting for age, gender, BMI, smoking and socioeconomic status. This was due to the substantially higher prevalence of low FVC (<80% predicted) in Indigenous Australians compared with non-Indigenous Australians, as reported in a previous BOLD Australia study paper.¹⁷ We also found that 41.5% of participants with PRISm lived in the most disadvantaged socio-economic status areas, which was significantly higher than participants with airflow limitation or normal spirometry.

The proportion of participants who reported working in a dusty job was similarly high in the PRISm and airflow limitation groups, compared with the normal spirometry group. This result may indicate that occupational exposure is not only associated with COPD but also associated with PRISm.

Previous studies found that PRISm was associated with a greater burden of multimorbidity than was normal spirometry.^15,31,32 Our study also showed this trend for a range of comorbidities. Previous studies reported that the proportion of heart disease in participants with PRISm was higher than those with normal spirometry, but lower than those with airflow limitation, similar to our findings.^15,31 However, our findings showed that after adjusting for age, gender, being Indigenous Australian, BMI, smoking, and socioeconomic status, individuals with PRISm still have a higher odds of reporting heart disease than those with airflow limitation. We found that the presence of diabetes and hypertension was higher in participants with PRISm compared with participants with airflow limitation, which was also consistent with previous studies.^31,33 The higher proportions of diabetes and hypertension in the PRISm group may be associated with the increased total and abdominal adiposity known to be associated with reduced FEV₁ and FVC.³⁴

We found that the prevalence of respiratory symptoms including cough/phlegm on most days for at least 3 months was similar in participants with PRISm and airflow limitation. Previous studies also reported that PRISm was significantly associated with increased risk of breathlessness,³² which is similar to our findings. However, the proportion using respiratory medications in the PRISm group was significantly lower compared to the airflow limitation group, consistent with a previous study.⁹ These findings are consistent with the lack of specific interventions for individuals with PRISm in real-life clinical practice. This may be due to insufficient knowledge and understanding of the heterogeneous causes of PRISm.

Turning to lung function, mean percent predicted FEV₁ and FVC were lower in adults with PRISm, consistent with its definition, and similar to previous findings.¹⁵ The FEV₁/FVC ratio of the three groups were similar to the results reported by the OCEAN study which was conducted on sequentially recruited Japanese individuals ≥40 years old.³¹ We also found that the prevalence of BD responsiveness, which was highest in those with airflow limitation, was similarly low in the PRISm and normal spirometry groups. Despite this, self-reported diagnosis of current asthma was more common in those with PRISm (15%) than normal spirometry (9%). Since underdiagnosis and misdiagnosis of asthma are undoubtedly common,³⁵ further study may need to explore asthma in the PRISm population.

Previous studies found that PRISm, compared with normal spirometry, was strongly associated with respiratory-related events including hospitalisations or deaths due to respiratory diseases.¹⁵ We also showed that the proportions of participants with PRISm who reported time lost from work or daily activities, and healthcare utilisation, were significantly higher than among those with normal spirometry.

We observed that both PRISm and airflow limitation were associated with poor general health status. Previous studies reported that the negative impact of COPD was greater on the physical than the mental aspects of HrQoL.³⁶ We also observed that the negative associations of PRISm with the physical aspects of HrQoL were similar to those of airflow limitation. However, after adjustment for confounders, we found no significant difference in mental health across the three groups.

The main strengths of this study were the data from a large nationwide population sample, the use of standardised data collection methods, coupled with a high level of quality control, which improved the internal validity of the analysis.^16,19 Adopting this approach to spirometry testing provided standardised and high-quality data and reliable estimates of the prevalence of PRISm and airflow limitation. The study protocol and core questionnaire were harmonised with the BOLD International protocol, which allowed for comparisons across BOLD participating countries.^19,37 Another strength of this study was the large number of Indigenous participants studied, allowing us to identify the high prevalence of PRISm among Indigenous Australians. This study avoided using ethnic adjustment in defining the predicted values for spirometry , which is consistent with recent recommendations by ATS/ ERS to avoid normalising potential effects of early life events and socioeconomic deprivation on lung size and lung function.³⁸

A limitation of our study was the cross-sectional design that did not allow assessment of causality or long-term outcomes. The low overall response rate might have introduced the potential for selection bias, with those included in this analysis being slightly younger and more likely to self-report a diagnosis of chronic respiratory disease than those who provided only minimal data.¹⁶ With a sample response rate of 27% there is always opportunity for selection bias to influence the estimates. However, the prevalence of ever smoking and clinically significant breathlessness are not too dissimilar to other published estimates.²² The study participants were not a simple random sample of the Australian population, as the six study sites were not selected completely at random. Although post-hoc weights have been used in previous work to adjust prevalence estimates for a better representation of the Australian population¹⁶; sample prevalence estimates were used in this analysis. The single day spirometry measurement was also a limitation because spirometry results may vary on different days, leading to differences in diagnostic criteria.³⁹ Finally, the BOLD study did not include assessment of total lung capacity, so we were not able to identify PRISm participants in whom reduced FEV₁ and FVC were due to pulmonary or extra-pulmonary restriction of lung volumes.

Our findings have important implications for the development of management strategies for patients with chronic respiratory symptoms. According to our results, the PRISm population contained highly symptomatic individuals with significant health burden. Further research is needed to explore the underlying causes of PRISm, and whether the association between PRISm and adverse health outcomes is causal. Because PRISm does not meet the diagnostic criteria for airflow limitation, detection of these individuals with potential risk of COPD or who may have other serious underlying chest or lung conditions may be missed during routine health checks. Our findings emphasise the importance of recognising PRISm in clinical settings. Identification of PRISm should be made in every day real-life clinical practice in order to identify causes and initiate appropriate interventions in order to optimise patient prognosis. Because of the heterogeneous underlying causes of PRISm, there is no evidence on what the best treatment is for individuals with PRISm; this is an important gap that needs more research in the future.^2,12 However, there are limitations of PRISm, including that it still relies on an accurate FVC and is calculated from two binary variables, which leads to loss of information and increases the risk of error.¹³ Future research can include information including respiratory symptoms, results of tests such as DLco and CT scans, and lung function trajectories which may provide a more comprehensive indication of individuals at risk of COPD,^5,40 and the underlying pathophysiology.

Supplemental Material

Supplemental Material - Prevalence and characteristics of adults with preserved ratio impaired spirometry (PRISm): Data from the BOLD Australia study

Supplemental Material for Prevalence and characteristics of adults with preserved ratio impaired spirometry (PRISm): Data from the BOLD Australia study by Yijun Zhou, Maria R Ampon, Michael J Abramson, Alan L James, Graeme P Maguire, Richard Wood-Baker, David P Johns, Guy B Marks, Helen K Reddel and Brett G Toelle in Chronic Respiratory Disease.

Footnotes

Acknowledgments

The BOLD study in Australia was funded by the National Health & Medical Research Council, Project Grant 457385. The BOLD study in Sydney was funded by grants from Air Liquide P/L, AstraZeneca P/L, Boehringer Ingelheim P/L, GlaxoSmithKline Australia P/L and Pfizer Australia P/L. Operations Centre: Mrs Tessa E. Bird and Dr Wei Xuan (Woolcock Institute of Medical Research). Sydney: Professor Christine R. Jenkins, Mrs Tessa E. Bird, Dr Kate Hardaker and Dr Paola Espinel (Woolcock Institute of Medical Research). Busselton: the late Professor A. W. (Bill) Musk, Dr Michael L. Hunter, Ms Elspeth Inglis and Ms Peta Grayson (University of Western Australia). Kimberley: Professor David N. Atkinson, Mr Dave Reeve, Dr Nathania Cooksley, Dr Matthew Yap, Ms Mary Lane, Dr Wendy Cavilla and Ms Sally Young (University of Western Australia). Melbourne: Ms Angela Lewis, Ms Joan Raven, Ms Joan Green and Ms Marsha Ivey (Monash University). Tasmania: Professor E. Haydn Walters, Mrs Carol Phillips and Ms Loren Taylor (University of Tasmania). NSW Rural: Dr Phillipa J. Southwell, Dr Bruce J. Graham, Dr Brian Spurrell, Mrs Robyn Paton, Ms Melanie Heine, Ms Cassandra Eccleston and Dr Julie Cooke (Charles Sturt University).

Declaration of conflicting of interests

MJA holds investigator initiated grants from Pfizer, Boehringer-Ingelheim, Sanofi and GSK. He has conducted an unrelated consultancy for Sanofi. He has also received a speaker’s fee from GSK. RWB reports cohort grants from the National Health & Medical Research Council. GBM has received funding for advisory boards with AstraZeneca. HKR holds investigator initiated grants from AstraZeneca, GlaxoSmithKline, Chiesi, Sanofi and Perpetual Philanthropy. She has received consulting fees from AstraZeneca, Chiesi and Novartis and honoraria from Alkem, AstraZeneca, Boehringer Ingelheim, Chiesi, Getz, GlaxoSmithKline, Novartis, Sanofi and Teva. HKR holds a leadership role in the Global Initiative for Asthma (GINA) and is a member of the National Asthma Council Guidelines Committee. All other authors declare no competing interests.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The BOLD study in Australia was funded by the National Health & Medical Research Council, Project Grant 457385. The BOLD study in Sydney was funded by grants from Air Liquide P/L, AstraZeneca P/L, Boehringer Ingelheim P/L, GlaxoSmithKline Australia P/L and Pfizer Australia P/L.

Ethical statement

ORCID iDs

Yijun Zhou

David P Johns

Supplemental Material

Supplemental material for this article is available online.

References

Vos

Lim

Abbafati

. Global burden of 369 diseases and injuries in 204 countries and territories, 1990–2019: a systematic analysis for the global burden of disease study 2019. Lancet 2020; 396: 1204–1222. DOI: 10.1016/s0140-6736(20)30925-9.

Global Initiative for Chronic Obstructive Lung Disease (GOLD) . Global strategy for prevention, diagnosis and management of COPD: 2024 report 2024. https://goldcopd.org/2024-gold-report/ (accessed 01 December 2023).

Diab

Gershon

Sin

, et al. Underdiagnosis and overdiagnosis of chronic obstructive pulmonary disease. Am J Respir Crit Care Med 2018; 198: 1130–1139.

Petrie

Toelle

Wood-Baker

, et al. Undiagnosed and misdiagnosed chronic obstructive pulmonary disease: data from the BOLD Australia study. Int J Chronic Obstr Pulm Dis 2021; 16: 467–475.

Han

Agusti

Celli

, et al. From GOLD 0 to pre-COPD. Am J Respir Crit Care Med 2021; 203: 414–423.

Pauwels

Buist

Calverley

, et al. 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: 1256–1276.

Vestbo

Lange

. Can GOLD Stage 0 provide information of prognostic value in chronic obstructive pulmonary disease? Am J Respir Crit Care Med 2002; 166: 329–332.

Global Initiative for Chronic Obstructive Lung Disease (GOLD) . Global strategy for prevention, diagnosis and management of COPD: 2023 report 2023. https://goldcopd.org/2023-gold-report-2/ (accessed 10 December 2023).

Wan

Castaldi

Cho

, et al. Epidemiology, genetics, and subtyping of preserved ratio impaired spirometry (PRISm) in COPDGene. Respir Res 2014; 15: 89.

10.

Martinez

Agusti

Celli

, et al. Treatment trials in young patients with COPD and pre-COPD patients: time to move forward. Am J Respir Crit Care Med 2021; 205: 275–287.

11.

Kanetake

Takamatsu

Park

, et al. Prevalence and risk factors for COPD in subjects with preserved ratio impaired spirometry. BMJ Open Respir Res 2022; 9: e001298.

12.

Wan

. The clinical spectrum of PRISm. Am J Respir Crit Care Med 2022; 206: 524–525. DOI: 10.1164/rccm.202205-0965ED.

13.

Knox-Brown

Amaral

Burney

. Concerns about prism. Lancet Respir Med 2022; 10: e51–e52.

14.

Wan

Fortis

Regan

, et al. Longitudinal phenotypes and mortality in preserved ratio impaired spirometry in the COPDGene study. Am J Respir Crit Care Med 2018; 198: 1397–1405.

15.

Wan

Balte

Schwartz

, et al. Association between preserved ratio impaired spirometry and clinical outcomes in US adults. JAMA 2021; 326: 2287–2298.

16.

Toelle

Xuan

Bird

, et al. Respiratory symptoms and illness in older Australians: the burden of obstructive lung disease (BOLD) study. Med J Aust 2013; 198: 144–148.

17.

Cooksley

Atkinson

Marks

, et al. Prevalence of airflow obstruction and reduced forced vital capacity in an Aboriginal Australian population: the cross‐sectional BOLD study. Respirology 2015; 20: 766–774.

18.

Mannino

McBurnie

Tan

, et al. Restricted spirometry in the burden of lung disease study. Int J Tubercul Lung Dis 2012; 16: 1405–1411.

19.

Buist

Vollmer

Sullivan

, et al. The burden of obstructive lung disease initiative (BOLD): rationale and design. COPD 2005; 2: 277–283.

20.

Australian Bureau of Statistics (ABS) . Socio-economic indexes for areas (SEIFA). Canberra: Australian Bureau of Statistics, 2011.

21.

Bestall

Paul

Garrod

, et al. Usefulness of the Medical Research Council (MRC) dyspnoea scale as a measure of disability in patients with chronic obstructive pulmonary disease. Thorax 1999; 54: 581–586.

22.

Poulos

Ampon

Currow

, et al. Prevalence and burden of breathlessness in Australian adults: the national breathlessness survey—a cross‐sectional web‐based population survey. Respirology 2021; 26: 768–775.

23.

Kosinski

Ware

Turner-Bowker

, et al. User's manual for the SF-12v2 health survey: with a supplement documenting the SF-12® health survey. Johnston, Rhode Island: QualityMetric Incorporated, 2007.

24.

Miller

Hankinson

Brusasco

, et al. Standardisation of spirometry. Eur Respir J 2005; 26: 319–338.

25.

Global Initiative for Chronic Obstructive Lung Disease (GOLD) . Global strategy for the diagnosis, management and prevention of chronic obstructive pulmonary disease 2010. https://www.goldcopd.org/Guidelines/guideline-2010-gold-report.html (accessed Dec 07 2011).

26.

Hankinson

Odencrantz

Fedan

. Spirometric reference values from a sample of the general US population. Am J Respir Crit Care Med 1999; 159: 179–187.

27.

Quanjer

Stanojevic

Cole

, et al. Multi-ethnic reference values for spirometry for the 3–95-yr age range: the global lung function 2012 equations. Eur Respir J 2012; 40: 1324–1343. DOI: 10.1183/09031936.00080312.

28.

Stanojevic

Kaminsky

Miller

, et al. ERS/ATS technical standard on interpretive strategies for routine lung function tests. Eur Respir J 2022; 60: 2101499.

29.

Pellegrino

Viegi

Brusasco

, et al. Interpretative strategies for lung function tests. Eur Respir J 2005; 26: 948–968.

30.

Lederer

Bell

Branson

, et al. Control of confounding and reporting of results in causal inference studies. Guidance for authors from editors of respiratory, sleep, and critical care journals. Ann Am Thorac Soc 2019; 16: 22–28.

31.

Kaise

Sakihara

Tamaki

, et al. Prevalence and characteristics of individuals with preserved ratio impaired spirometry (PRISm) and/or impaired lung function in Japan: the OCEAN study. Int J Chronic Obstr Pulm Dis 2021; 16: 2665–2675.

32.

Higbee

Granell

Davey Smith

, et al. Prevalence, risk factors, and clinical implications of preserved ratio impaired spirometry: a UK Biobank cohort analysis. Lancet Respir Med 2022; 10: 149–157.

33.

Wijnant

SRA

De Roos

Kavousi

, et al. Trajectory and mortality of preserved ratio impaired spirometry: the Rotterdam study. Eur Respir J 2020; 55: 1901217.

34.

Jones

Nzekwu

M-MU

. The effects of body mass index on lung volumes. Chest 2006; 130: 827–833.

35.

Aaron

Boulet

Reddel

, et al. Underdiagnosis and overdiagnosis of asthma. Am J Respir Crit Care Med 2018; 198: 1012–1020.

36.

Janson

Marks

Buist

, et al. The impact of COPD on health status: findings from the BOLD study. Eur Respir J 2013; 42: 1472–1483.

37.

Buist

McBurnie

Vollmer

, et al. International variation in the prevalence of COPD (the BOLD Study): a population-based prevalence study. Lancet 2007; 370: 741–750.

38.

Bhakta

Bime

Kaminsky

, et al. Race and ethnicity in pulmonary function test interpretation: an official American thoracic society statement. Am J Respir Crit Care Med 2023; 207: 978–995.

39.

Schermer

Robberts

Crockett

, et al.

Should the diagnosis of COPD be based on a single spirometry test?

NPJ Prim Care Respir Med 2016; 26: 16059.

40.

Bui

Lodge

Burgess

, et al. Childhood predictors of lung function trajectories and future COPD risk: a prospective cohort study from the first to the sixth decade of life. Lancet Respir Med 2018; 6: 535–544.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.85 MB