COPD Positive Screening with Spirometry Increases Motivation to Quit Tobacco Smoking in an Addiction Treatment Center

Abstract

Providing an on-site immediate diagnosis of Chronic Obstructive Pulmonary Disease (COPD) and lung age in tobacco smokers could be a motivational tool for smoking cessation. Our aim was to investigate the effects of an abnormal spirometry results on motivational change and subsequent smoking cessation. We conducted a retrospective analysis of smoking status after 3 months of tobacco counseling. Patients were recruited in an addiction outpatient center. Spirometry results were obtained with a portable device during the first visit. The sample was thus divided in 3 groups: COPD, subthreshold-group (no COPD but abnormal lung age) and normal spirometry. Among the three groups, we compared the immediate motivation change, difference in Q-MAT motivation scale score after minus before spirometry (Kruskal-Wallis test) and the smoking status after 3 months (Fisher test). We included 48 patients (37 males, median age 44 years, median cigarette-per-day 20). Spirometry results divided the sample in COPD (N = 13), subthreshold (N = 11) and normal group (N = 24). Mean Q-MAT score change after spirometry was different between groups (p = 0.019), greater in COPD (4.62 ± 3.38) than normal group (1.46 ± 3.11), and lower in patient with a co-occuring hazardous alcohol use (p = 7.6 × 10⁻³). Three-months smoking status was different between spirometry results groups (p = 0.0021). COPD (5/13, 38.5%) and subthreshold patients (6/10, 60.0%) had stopped more frequently than patients from the normal-group (2/22, 9.1%). The effect of immediate spirometry results on motivation to quit varies according to the screened pulmonary damages and hazardous alcohol use. It could be a useful tool in addiction treatment centers.

Keywords:

• Spirometry
• tobacco smoking
• chronic obstructive pulmonary disease
• motivation
• addiction

Introduction

Tobacco smoking is a major cause of premature death and comorbidities, including respiratory insufficiency and Chronic Obstructive Pulmonary Disease (COPD) [1,2]. Particularly, COPD prevalence is higher among patients in care for a substance use disorder other than tobacco [3]. By a self-adaptation mechanism, clinical respiratory symptoms reported by patient are often mild at the onset of the disease. Some patients may limit, intentionally or unintentionally, their physical exercise below the threshold where subjective symptoms appear [1]. Spirometry has thus been proposed as an opportunity to screen earlier for COPD in asymptomatic patients and to provide motivational support for smoking cessation, but its effectiveness in increasing smoking cessation has not been demonstrated [1,4]. Among spirometry results, the “lung age” has been proposed as an easily communicable health education result [5]. The lung age is the age of a healthy never-smoker individual who would have similar Forced Expiratory Volume in one second (FEV1) that the patient, based on the spirometric references standards, which take into account age, sex, height and ethnicity [5]. The efficacy of spirometry and lung age as motivational tools to promote smoking cessation in patients with smoking and concurrent alcohol or cannabis use has never been tested. Furthermore, knowing that patients suffering from substance use disorders may have lower screening and treatment access [6], we wanted the assessment to be performed “on-site” within the addiction center, and the result to be given immediately during the first tobacco counseling visit. The aim of the study was to investigate the effects of an abnormal spirometry results on motivational change for smoking cessation and subsequent cessation knowing that all patients received standardized counseling, motivational interviewing and the proposal of nicotine replacement therapy (NRT) prescription.

Methods

Subjects and sampling procedure

This retrospective chart review of tobacco counseling consultation performed for consecutive patients was conducted in the addiction medicine department of a Parisian university hospital in 2018. This department receives patients referred by the addiction medicine department and other departments from the same hospital, and is oriented toward smoking cessation. Not all patients had others addictive comorbidities, but the majority presented with alcohol or cannabis use disorders. Inclusion criteria were current tobacco smoking, over 18 years old. Exclusion criteria were cognitive impairment, already known COPD, or previous spirometry screening in the past year, substance use disorders with substances other than cannabis and alcohol.

Assessment

Smoking status

Self-declared smoking status expressed in cigarettes per day was recorded at the first visit as well as at follow up visits. No exhaled carbon monoxide or urinary cotinine measures were performed.

Questionnaires

Patients completed the Alcohol Use Disorders Identification Test (AUDIT), the Cannabis Use Disorders Identification Test (CUDIT), the Fagerström Test for Nicotine Dependence (FTND), and the Q-MAT (“Questionnaire de Motivation à l’Arrêt du Tabac”) [7]. The score range of AUDIT and CUDIT scale is from 0 to 40. A score ≥8 defines, respectively, hazardous use of alcohol or cannabis. The score range of FTND is from 0 to 10. A score ≤2 defines very low, 3–4 low, 5–6 moderate and ≥7 high nicotine dependence. The Q-MAT is a four-questions validated questionnaire assessing the motivation to quit smoking. The score range is from 0 to 20. A score ≤6 defines insufficient motivation, 7–13 medium motivation and ≥14 high motivation [7].

Health education and motivational interviewing

All first tobacco counseling consultation started with a standardized health education and motivational interviewing module. The Motivational interviewing began by questioning the patient’s motivations to stop smoking, investigating whether they were related to health status. The patient was invited to elaborate and build his own treatment plan according to his present motivational status. The medical doctors and practice nurses from this program have been instructed with theoretical and practical training of at least 16 h. The spirometry result was included in motivational interviewing intervention using the figure of Fletcher and Peto [8]. In patients with normal spirometry or subthreshold results we communicated that this assessment was a measure of the current status, and could deteriorate if smoking continued. Furthermore, patients were informed that stopping smoking would maintain the decline of the lung age within the range of natural aging effect and helps prevent COPD. In patients with COPD, smoking cessation has been presented as a key treatment to limit the progression of deterioration in respiratory function and helps prevent the risk of COPD decompensation. The standardized visit ended with for all patients with the proposal of NRT prescription that could be done immediately.

Spirometry procedure

All first tobacco counseling consultation included a spirometry that was performed by the same trained medical doctors or practice nurse than the motivational interviewing with a portable device. The staff was instructed with theoretical and practical training by a pneumologist, following the recommendations from the Société de Pneumologie de Langue Française (SPLF) [9]. The spirometric measurements were performed with the portable Spirobank II Basic^® device and processed by the Winspiro Pro^® software from Medical International Research^® (MIR^®) on a computer connected to the device [10]. FlowMIR^® disposable turbines were used, that do not require calibration, and a monthly “biological” check with a healthy subject was carried out according to the advice of the SPLF [9,10]. Raw data were automatically compared to the Global Lung Function Initiative (GLI-2012) reference standards [11]. Validity criteria for spirometry were checked by a trained medical doctor who was present on site before giving the results to the patients. This validity check included the acceptability of the measurement (sufficient and reproducible effort for at least three measures, expiratory volume/time traces smooth and free from irregularities) and its reproducibility (for FEV1 and forced vital capacity (FVC): it should be observed a difference <150ml between the 2 best measures) [9,12]. Then, the results were communicated to the patient.

COPD was defined when two criteria were met: a ratio of FEV1 to FVC below 70% and the absence of significant reversibility after bronchodilator use (FEV1 change <200ml and <12%) [1]. All abnormal results were expressed using the “lung age” and the figure of Fletcher and Peto [8]. Thus, spirometry results determined 3 groups: (i) COPD group, (ii) subthreshold-group (no COPD but FEV1 below the theoretical value expected in healthy subjects according to (GLI-2012) reference standards, i.e. a lung age higher than the chronological age) and (iii) normal spirometry group.

Outcome measure

The QMAT was reiterated just after the spirometry measure during the first visit to assess the motivation change, expressed as the difference in Q-MAT scale score after minus before spirometry. The primary outcome was the immediate variation on motivation to quit tobacco smoking between groups. The secondary outcome was smoking status after 3 months. The use of prescribed NRT at months 3 was recorded according to patients’ self-declaration.

Statistical analysis

Patients with an initial high motivation to quit smoking, evaluated by the pre-consultation Q-MAT scale (≥14) were excluded from the analysis to avoid a ceiling effect. Their characteristics were compared to those of included patients with appropriate tests.

Primary outcome

The change of QMAT score after minus before spirometry in the whole sample (time effect) was tested with the Wilcoxon test.

The primary outcome was the Q-MAT scale change according to the spirometry group. It was tested by a Kruskal-Wallis and then Mann-Whitney-Wilcoxon tests with FDR correction for pairwise comparisons. A repeated-measure ANOVA was added to test for time x group interaction.

Furthermore, the effect of expected cofounding factors on Q-MAT difference was tested (age (linear regression), sex, AUDIT, CUDIT categories (Mann-Whitney) and FTND categories (Kruskal-Wallis test)). Then, a multivariate linear regression was performed to identify independent factors associated with the Q-MAT change in our sample, including the factor associated with a p-value <0.10 at the univariate analyses step.

Secondary outcome

We compared the 3-months smoking status according to the spirometry group (Fisher test).

We tested the effect of expected associated factors on self-declared month-3 smoking cessation between the three spirometry groups: declared use of prescribed NRT at month 3 (Fisher test), QMAT score change during the first visit (Mann-Whitney-Wilcoxon test). A logistic regression was performed to identify independent factors associated with smoking cessation, including the factor associated with a p-value <0.10 in those univariate analyses. All statistical analyses were performed with R software version 3.2.

Ethics

The study was conducted in adherence to the guidelines of the Declaration of Helsinki and following French laws on biomedical research (Loi Jardé 2014, décrets d’application 2017). This chart review study design was approved by an internal ethics committee. Our hospital has applied for and obtained a specific authorization Number 2017–013 by the CNIL (Commission Nationale Informatique et Liberté, French National Board for Information Systems and Freedom) for the analysis of data collected during routine care. Patients gave a specific consent for the use of their data originating from routine care for research purposes.

Results

Fifty-eight consecutive patients’ files were selected following the inclusion criteria. Among them 10 were excluded for having a high initial motivation (Q-MAT ≥ 14). Thus the analysis was performed on 48 subjects. The median age was 44 years (range 20–67 years). There were 37 men (77.1%). Thirty-four patients had hazardous alcohol use (70.8%) and 11 hazardous cannabis use (22.9%). They smoked a median of 20 cigarettes per day (range 1–50). The FTND was high for 12 patients (25%), medium for 9 (18.8%) and low or very low for 20 (41.7%). The excluded patients were not significantly different than the included one on age, gender, the number of cigarettes per day, the FTND, hazardous alcohol use and hazardous cannabis use (all p-values >0.05). Thirteen patients were defined as COPD (27.1%), 11 were in the subthreshold-group (22.9%) and 24 in normal-spirometry-group (50.0%).

The mean of QMAT scale was 6.38 at the initial evaluation and 9.19 after the communication of spirometry results. The increase was significant in the whole sample (p < 0.001). In univariate analysis, the increase in motivation was significantly different between the three groups (p = 0.019). The repeated measure ANOVA showed significant time effect (p < 0.001), time × group interaction (p = 0.016) but failed to show a significant between-subject effect (p = 0.125). In multivariate analysis, the mean difference was greater in COPD group than normal spirometry group (+3.58, p = 5.1 × 10⁻³). Furthermore, a significantly lower increase in motivation was observed in patient with a co-occuring hazardous alcohol use versus without (−2.87, p = 7.6 × 10⁻³). The results are presented in Table 1. Concerning the month-3 smoking status, available on 45 patients (93.8%), it was also significantly different between the three groups (p = 0.0021). There were 5/13 (38.5%) nonsmoker in the COPD group, 6/10 (60.0%) in subthreshold-group and 2/22 (9.1%) in normal-spirometry-group. Fifteen patients (33.3%) declared that they used the prescribed NRT. This use was associated with tobacco cessation (p = 0.014) and was not statistically different according to the spirometry groups (p = 0.11). QMAT score increase during the first visit was observed to be significantly higher in patients who declared that they did quit smoking at month-3 (p = 0.014; +5.0 [±3.8] versus +1.9 [±3.0]), but not in patients who declared that they did use the prescribed NRT (p = 0.19). In multivariate logistic regression, analysis was limited by the number of tobacco cessation events to two explanatory variables. In this context, the declared use of NRT (OR 5.42 [IC95% 1.18–24.9]) and the observed QMAT score increase during the first visit (OR 1.32 [IC95% 1.04–1.68]) were independently associated with tobacco cessation at 3 months.

Table 1. Univariate analyses of the mean Q-MAT score after minus before spirometry (n = 48).

	Univariate analyses				Multivariate analysis^µ
	Q-MAT scale before (mean [±SD])	Q-MAT scale after (mean [±SD])	Delta Q-MAT scale (mean [±SD])	p-value	Delta Q-MAT scale (β)	p-value
Intercept	–	–	–	–	4.03	–
Spirometry results
No COPD and FEV1 ≥ 100% (n = 24)	6.08 [±4.17]	7.54 [±4.92]	1.46 [±3.11]^a	0.019^$	Ref
No COPD and FEV1 < 100% (n = 11)	6.18 [±3.46]	9.82 [±4.12]	3.64 [±3.29]		1.86	0.11
COPD (n = 13)	7.08 [±1.89]	11.69 [±3.07]	4.61 [±3.38]^a		3.58	5.1 × 10^–Citation3
Age (/10 years)	β = 0.48	β = 0.70	β = 0.22	0.61^µ	−0.12	0.79
Sex
Female (n = 11)	6.82 [±2.71]	10.91 [±3.86]	4,09 [±3.56]	0.22^£	1.27	0.25
Male (n = 37)	6.24 [±3.71]	8.68 [±4.71]	2,43 [±3.38]	MWW U = 153.5	Ref
Nicotine dependence (FTND)					–	–
Low or very low (n = 20)	6.60 [±3.69]	10.15 [±5.40]	3,55 [±3.92]	0.29^$
Moderate (n = 9)	4.89 [±4.14]	5.89 [±3.69]	1,00 [±3.28]
High (n = 12)	6.42 [±2.75]	9.08 [±3.12]	2,67 [±3.20]
Hazardous cannabis use (CUDIT ≥ 8)
Yes (n = 11)	6.64 [±3.64]	7.73 [±4.10]	1.09 [±2.30]	0.060^£	−1.53	0.15
No (n = 37)	6.30 [±3.49]	9.62 [±4.69]	3.32 [±3.60]	MWW U = 127.5
Hazardous alcohol use (AUDIT ≥ 8)
Yes (n = 34)	6.82 [±3.57]	8.79 [±4.36]	1,97 [±3.11]	0.011^£	−2.87	7.6 × 10^–Citation3
No (n = 14)	5.29 [±3.12]	10.14 [±5.14]	4.86 [±3.51]	MWW U =	Ref

^$ Kruskal-Wallis test.

^£ Mann-Whitney Wilcoxon test (MWW).

^µ Linear regression.

QMAT: Questionnaire de Motivation à l’Arrêt du Tabac.

FEV1: Forced Expiratory Volume in 1 s.

^a Significant difference with a p-value adjusted to 0.017 by the False Discovery Rate method and a MWW test.

Discussion

We observed a significant immediate increase of motivation to quit smoking after the realization of an on-site screening spirometry performed during the first tobacco counseling visit. In this standardized program, all patients received motivational interviewing and the proposal of NRT prescription. We also observed a higher quit success at month 3 in patients with abnormal results. This is to the best of our knowledge the first time that such an increase is observed in patients with smoking and co-occurring alcohol or cannabis use disorders. One previous study, performed in tobacco smokers only, found similar results of an observed increase in quit attempts after spirometry measures and communication of the calculated lung age [13]. Of note, this study showing an increase in quit attempts was a randomized study with a one-year follow up and did also demonstrate a higher tobacco cessation rate (13.6% in the group with the use of “lung age” versus 6.4% in the control group with the usual use of FEV1). Other studies did not observe a higher motivation to quit when using the lung age [4,14]. Those discrepancies may emerge from the choice to analyze together patients whatever their initial motivation score [15]. This lesson from previous studies lead us to leave aside patient with already a high motivation score in our study design, despite the risk of reducing power [13]. The use of the lung age to communicate spirometry results may be specifically useful in patients with several co-occuring substance use disorders recruited in addiction outpatient centers where the expected patients with COPD or abnormal lung age is higher than in the general population. Furthermore, patients with multiple substance use disorders are less prone to attend medical visits outside of their current care center [6]. Therefore, the use of portable spirometry device directly in the addiction outpatient center, with a result given on the same day and associated with a specific motivational intervention and NRT prescription can facilitate screening, treatment access and promote smoking cessation.

Motivation increased significantly more in the most pulmonary affected patients, especially after COPD diagnosis, which was previously observed [15]. The sustainability of this motivational change is indirectly confirmed in the following months with the higher self-reported abstinence rate at month-3 in COPD-group and subthreshold-group. The difference between them could be related to several hypotheses: a higher feeling of vulnerability in subthreshold-group, a greater social disposition to change in less impaired patients, a greater sensitivity to motivational interviewing with a higher perceived benefit in preventing the occurrence of COPD than in preventing its progression.

Respiratory damages from tobacco are asymptomatic at first [1]. Emphasizing the pulmonary damage with the easily communicable “abnormal lung age” gives the opportunity to create a “teachable moment” with two mechanisms to promote change: « persuasion » and « enablement » (with NRT), added to standardized motivational interviewing [16,17]. However, our study does not determine whether the diagnosis of COPD increased abstinence or whether it was the normal spirometric result that made the counseling inefficient. Patients did not express a significant decrease in their motivation score after the spirometry when the result was normal in this first smoking cessation visit. But a demotivating effect of an estimated normal lung age cannot be excluded [18].

Our study has several limitations. It was a monocentric, without randomization, blind record, or control group, giving way for placebo effects or nonspecific care setting effects [4]. Only self-reported abstinence was recorded. There is also a limitation inherent to the use of the lung age based on FEV1 standard [18,19]. The FEV1 reference value, established taking into account age, gender, weight and ethnicity, is submitted to a representativeness bias of the populations used to construct its equations (body build, level of physical training and other). And in the absence of repeated measurements in the same individual over time, the inter-individual variability interferes with the estimation of the effect of tobacco smoking [18]. Particularly, the predictability of these equations for determining the lung age in young adults is discussed. In our sample, withdrawing patients below 25 years (N = 4) from the analysis did not change the main results (QMAT change after spirometry and was still significantly associated with the spirometry groups), but we had not enough power left to analyze the secondary outcome of smoking cessation at month-3 (data not shown).

The strengths of the study are the inclusion of patients in the specific context of care for another substance use disorder and the declared 3-month abstinence rate that enables comparative analyses.

These results suggest that on-site, immediate COPD screening with portable devices for smokers recruited in addictology facilities promotes motivation and smoking cessation and should be generalized.

Conclusions

Immediate spirometry result is associated with an increase in motivation to quit, significantly higher in patients with screened pulmonary damages and lower in patient with a co-occuring hazardous alcohol use. It could be a useful tool in addiction treatment centers, receiving patients with nicotine dependence associated with other substance use disorders.

Declaration of interest

The authors report no conflict of interest.