Chronic obstructive pulmonary disease (COPD) is primarily characterized by a persistent airflow limitation that is associated with chronic inflammation involving innate effector cells in the lungs and, in severe cases, at the systemic level as well. The latter is linked to the development of several comorbidities, contributing to the severe burden of COPD on public health.1 Specifically, the World Health Organization (WHO) estimated that in 2012 COPD was the third leading cause of death in the general population.2 In addition, it was classified as the sixth leading cause of disability-adjusted life years (DALYs) in developed countries, the third cause of DALYs in the United States and the fourth in the United Kingdom in 2013.3 A major challenge in reducing COPD morbidity and mortality is identifying modifiable risk factors which could reduce COPD incidence.
Cigarette smoking is the most well studied and dominant risk factor for COPD; however, among individuals with the same smoking history, not all will develop COPD. This indicates that other factors such as genetic predisposition, environmental pollution, or even diet may have an impact on COPD risk.1
It has been proposed that high intake of dietary fiber via its antioxidant and anti-inflammatory properties4–8 may reduce inflammatory processes in the lungs and prevent COPD. Until now, only one prospective cohort study, in the United States, has examined the association between dietary fiber intake and risk of COPD, in a population with low cereal fiber intake.9 In that study, only cereal fiber, but not fruit or vegetable fiber, intake was associated with a decreased risk of COPD.
To further clarify the potential association between dietary fiber intake and incidence of COPD, we examined the associations between total dietary fiber and specific fiber sources and risk of COPD by smoking status in a large prospective cohort of Swedish men with high fiber intake especially from cereal fiber.
METHODS
Study Population
The Cohort of Swedish Men (COSM) was established in central Sweden in 1997, when all men ages 45–79 years (born between 1918 and 1952) residing in Västmanland and Örebro counties (n = 100,303) received written information about the study and a self-administered questionnaire. Among the 48,850 men who returned a completed questionnaire, we excluded those with incorrect or missing national identification numbers (n = 297), those who died before baseline (n = 55), and those with a prior diagnosis of cancer (other than nonmelanoma skin cancer, n = 2,592). Moreover, based on the Swedish health registers, men with a history of COPD (the International Classification of Diseases, 10th Revision; ICD-10 code: J44, n = 281) and those with an extreme total energy intake (3 SDs from the log-transformed mean energy intake, n = 567) were excluded. After the above exclusions, 45,058 men remained and were followed from 1 January 1998 through 31 December 2012. When compared with the Official Statistics of Sweden, the original and analytical cohort well represented the Swedish male population in 1997 in terms of age distribution, educational level, prevalence of overweight and obesity, and smoking status.10 The study was approved by the Regional Ethical Review Board at Karolinska Institutet (Stockholm, Sweden), and response to the questionnaire constituted the participants’ informed consent.
Assessment of Dietary Intake
Data about food consumption were collected by a 96-item food frequency questionnaire (FFQ) in 1997. Men reported how often, on average, they had consumed each type of food during the last year. For most of the food items, including 5 fruits and 13 vegetables, men chose from eight predefined frequency categories, ranging from never to ≥3times per day. For commonly consumed foods, such as bread, men were asked to indicate their daily or weekly frequency of consumption of a standard portion. Dietary fiber intake was calculated by multiplying the frequency of intake of each food item by the fiber content (obtained from the Swedish Food Administration Database)11 of appropriate age-specific portion sizes. Next using the residual method intake of dietary fiber was adjusted for total energy intake to 2,600 kcal/day (mean intake in this population).12
We assessed the relative validity of the FFQ in 248 Swedish men, ages 45–74 years old from the study area, by comparing to fourteen 24-hour recall interviews.13 The Spearman correlation coefficient between the two methods used to collect data was 0.71 for energy-adjusted total dietary fiber intake.
Assessment of Other Variables
Information about education level, weight, height, physical activity, smoking status, and alcohol consumption (type of alcoholic beverages and frequency of consumption) was obtained via the self-administered questionnaire. We calculated body mass index (BMI) by dividing the weight in kg by the square of height in meters. Participants reported their level of activity at work, time spent on home/housework, walking/bicycling, exercise (including gym and other activities), and sedentary time including watching TV/reading and hours of sitting/lying down and sleeping. Daily time spent by the participants on each activity was multiplied by the activity’s typical energy expenditure requirement and expressed in metabolic equivalents. Summarizing all individual activities a metabolic equivalent (MET)-hour per day (24 hours) score was calculated.14 We asked men about their smoking status, including average number of cigarettes smoked per day at different ages (15–20, 21–30, 31–40, 41–50, and 51–60 years, and in the present year), age they started smoking, and years since stopping smoking. Pack-years of smoking were estimated by multiplying the number of years of smoking by the reported number of packs of cigarettes smoked per day within each respective age category.
Case Ascertainment
With the use of the unique personal identification number assigned to each Swedish resident, we identified incident COPD cases through linkage with the Swedish Patient Register (inpatient and outpatient registers) and the Swedish Cause of Death Register. COPD was defined according to the International Classification of Diseases and Related Health Problems, 10th Revision (ICD code J44). In the present study, a COPD case was defined as the first diagnosis of COPD (listed either as the primary diagnosis or at any diagnosis position) in the Swedish Patient Register or in the Cause of Death Register (only the primary position). It has been previously reported that from 1999 to 2009 (a time period overlapping with our study) a higher proportion of patients with COPD were detected and diagnosed in Sweden.15 The proportion of the cases diagnosed in primary care increased from 59% to 81% in those years and patients who received a COPD diagnosis in 2009 were also, on average, seven years younger than those who were diagnosed in 1999 (66 vs. 73 years old, respectively).15 These observations may indicate some degree of COPD underdiagnosis during early years of follow-up of the COSM cohort.
Statistical Analysis
Study participants were followed from baseline (1 January 1998) to the date of COPD diagnosis, death, or the end of follow-up (31 December 2012), whichever occurred first. Total dietary fiber intake and cereal, fruit, and vegetable fiber intake were each categorized into quintiles. Hazard ratios (HRs) and 95% confidence intervals (CIs) of COPD were estimated with Cox proportional hazards regression models. Multivariable HRs were adjusted for the following potential risk factors: age (years, continuous), education (less than high school, high school, or university), BMI (<18.5, 18.5–24.9, 25–29.9, or ≥30 kg/m2), total physical activity (metabolic equivalent-hours/day, quintiles), smoking status, and pack-years of smoking (never; past <20, 20–39, or ≥40 pack-years; or current <20, 20–39, or ≥40 pack-years), alcohol intake (g/day, quintiles), and energy intake (kcal/day, quintiles). Missing data on educational level (1%), smoking status (1%), BMI (5%), and total physical activity score (23%) were handled using the multiple imputation procedure. We applied multiple imputation by chained equations (MICE) to obtain five imputed data sets of the analytical cohorts.16,17 The HRs of the imputed data sets from the MICE procedure were pooled together by using the Rubin’s rule to obtain valid statistical inferences.18 For specific fiber sources, intake of cereal, fruit, and vegetable fiber was mutually adjusted by inclusion in the same multivariable model.
The proportional hazards assumption was assessed by regressing scaled Schoenfeld residuals against survival time, and there was no evidence of departure from the assumption. Tests for linear trend were conducted by using types of fiber intake as continuous variables. Using a likelihood ratio test, the interactions between fiber intake and smoking status, BMI, physical activity, and alcohol consumption in predicting the risk of COPD were tested.
The statistical analyses were performed using SAS version 9.4 (SAS Institute Inc, Cary, NC); all P values ≤0.05 were considered statistically significant and were two-sided.
RESULTS
Characteristics of the Cohort
The mean age of the participants at baseline was 60.2 ± 9.7 years, and 62.9% were ever smokers (24.5% current smokers and 38.4% ex-smokers). The mean energy-adjusted total dietary fiber intake was high (30.4 ± 8.2 g/day), and the median intakes of dietary fiber between the highest and the lowest quintiles differed two-fold. Compared with men in the lowest quintile of total dietary fiber intake, those in the highest quintile were less likely to be current smokers and had higher consumption of whole grains, fruits, and vegetables, while consuming less alcohol and processed red meat (Table 1). Cereal, fruit, and vegetable fiber contributed 67%, 12%, and 9%, respectively, of the total dietary fiber intake. Pearson correlation coefficients between total versus cereal was 0.83, versus fruit was 0.38, and versus vegetable fiber was 0.38.
;)
TABLE 1:
Age-Standardized Characteristics of 45,058 Men in the Cohort of Swedish Men by Quintiles of Energy-Adjusted Total Dietary Fiber Intakea at Baseline 1998
Overall and Time Period–specific Results
During a mean follow-up of 13.1 years (591,944 person-years, 1998–2012), 1,982 incident cases of COPD were ascertained. During the entire follow-up period, daily intake of dietary fiber ≥36.8 g in comparison to intake of <23.7 g was associated with 38% (95% CI = 29%, 47%) lower risk of COPD (Table 2). To examine the impact of a potential underdiagnosis during the early years of follow-up, we calculated HRs stratified by time of diagnosis in 5-year periods (eTable 1; https://links.lww.com/EDE/B265). The absolute number of cases of COPD diagnosed was lowest from 1998 to 2002 (n = 515), with more cases diagnosed in each 5-year period (2003–2007, n = 682; 2008–2012, n = 785). The average number of COPD diagnoses in each of the 5-year periods was 103, 136, and 157 cases per year, respectively. However, the HR for total fiber intake was similar in each 5-year period. In addition, after excluding the first 5 years of follow-up, the results were similar to those obtained for entire follow-up period; the HR between the highest and lowest quintiles of total fiber intake was 0.60 (95% CI = 0.50, 0.71).
TABLE 2:
Hazard Ratios (95% CIs) of Chronic Obstructive Pulmonary Disease by Quintiles of Energy-Adjusted Total Dietary Fiber in 45,058 Swedish Men, 1998–2012
Owing to the potential overlap between COPD and asthma diagnosis, we performed a sensitivity analysis excluding men with an asthma diagnosis before baseline or during follow-up (1,167 cases of asthma) and the results did not appreciably change. The HR of COPD for men in the highest versus lowest quintiles was 0.58 (95% CI = 0.49, 0.68) for total fiber when asthma cases were excluded.
Findings by Smoking Status
We investigated whether the association between dietary fiber intake and risk of COPD differed by cigarette smoking status and other COPD risk factors such as BMI, physical activity, and alcohol consumption. An interaction was found only for smoking status (never vs. ever smokers, P for interaction = 0.04); therefore, we present results stratified by smoking status.
Among current smokers (n = 11,043 participants, 1,033 cases of COPD) and ex-smokers (n = 17,289 participants, 767 cases of COPD), the age-standardized incidence rate (SIR) of COPD was lower among those who consumed ≥36.8 g/day of total fiber (525 and 328 per 100,000 person years, respectively) than among those who consumed <23.7 g/day (1 228 and 563 per 100,000 person years, respectively) (Table 3). The risk of COPD incidence in men in the highest versus those in the lowest quintile of total fiber intake was 46% (95% CI = 33%, 57%) lower in current smokers and 38% (95% CI = 22%, 50%) lower in ex-smokers (Table 3). The HRs comparing the highest to lowest quintiles of cereal fiber intake in relation to COPD incidence were 0.62 (95% CI = 0.51, 0.77) in current and 0.66 (95% CI = 0.52, 0.82) in ex-smokers (Table 3). In current smokers, fruit fiber intake (HR = 0.65; 95% CI = 0.52, 0.81) and vegetable fiber intake (HR = 0.71; 95% CI = 0.57, 0.88) were inversely associated with COPD risk, while in ex-smokers intake of fruit fiber (HR = 0.77; 95% CI = 0.61, 0.98) but not vegetable fiber (HR = 0.92; 95% CI = 0.71, 1.19) was associated. Among men who had never smoked, no associations were observed between total fiber intake and fiber sources and risk of COPD.
TABLE 3:
Hazard Ratios (95% CIs) of Chronic Obstructive Pulmonary Disease by Quintiles of Energy-Adjusted Total Dietary Fiber and Sources of Fiber Intake by Smoking Status in Swedish Men, 1998–2012
In the dose–response analysis (Table 4), each 5 g daily increment in total fiber was associated with 12%, cereal fiber with 9%, fruit fiber with 25%, and vegetable fiber intake with 29% lower risk of COPD in current smokers. In ex-smokers, the decreased risk was observed for total (10%) and cereal fiber (11%).
TABLE 4:
Multivariable Hazard Ratiosa (95% CIs) of Chronic Obstructive Pulmonary Disease by 5 g/day Increment in Total Dietary Fiber Intake and Sources of Dietary Fiber Intake by Smoking Status in Swedish men, 1998–2012
To examine the association of joint effects of total dietary fiber intake with smoking status, we conducted a comprehensive analysis by including current, ex-smokers, and never smokers in the same model (Figure). Compared with never smokers in the highest quintile (≥36.8 g/day) of total dietary fiber intake (the reference group), current smokers in the lowest quintile (<23.7 g/day) had a 17.9-fold and ex-smokers in the lowest quintile had an 8.0-fold higher risk of COPD. The risk of COPD decreased with increasing fiber intake to 8.1-fold in current smokers and 4.4-fold in ex-smokers in the highest quintile of total fiber. There were no observed associations between intake of dietary fiber and risk of COPD in never smokers.
FIGURE:
Hazard ratios (95% confidence intervals) of chronic obstructive pulmonary disease by quintiles of total dietary fiber intake and smoking status (P for interaction = 0.04). Hazard ratios were adjusted for age, education, body mass index, physical activity, intake of energy, and alcohol consumption.
DISCUSSION
In this population-based prospective study of men with high fiber intake, intake of total and cereal fiber was inversely associated with COPD risk in ex- and current smokers, while fruit and vegetable fiber intake reduced risk only in current smokers.
Our results confirm and expand the results from only one previous prospective study based on two US cohorts. That study also observed that current smokers in the highest versus the lowest quintiles of total dietary fiber and cereal fiber intake had a 40% and 34% lower risk of COPD, respectively; in contrast to our results, there was no association with fruit fiber intake.9 Among ex-smokers, also in contrast with our findings, there was no association between total and cereal fiber intake and COPD risk.9 The differences between the results may be explained by different amounts of total dietary fiber and fiber intake from specific sources in the studied populations. Swedish men had higher total fiber intake (medians in Q1–Q5: 20.5–40.9 g/day) than the studied American population (11.2–28.5 g/day); they also had higher cereal fiber intake (12.1–30.1 vs. 2.2–9.0 g/day). In contrast, intake of vegetable fiber was lower in Swedish men (1.0–4.8 g/day) than in the American population (3.5–10.7 g/day); intake of fruit fiber was comparable (1.0–6.9 g/day in Swedish men and 1.4–7.6 g/day in American population). Also the contributions of cereal, fruit, and vegetable fiber to total fiber intake were different between the studies. In our study population, cereal fiber was the major contributor to total fiber intake (67%). The most important food items accounting for cereal fiber intake were crisp bread (37%), whole-grain bread (19%), white bread (9%), cereals/muesli (3%), and oatmeal (3%). Common types of crisp bread and whole-grain bread consumed in this population are mainly based on rye.
Two previous studies have examined the associations between dietary fiber intake and lung function measured by spirometry parameters or symptoms of cough with phlegm.19,20 In a cross-sectional study conducted in 11,897 men and women, those in the highest quintiles of total, cereal, and fruit fiber intake compared to those in the lowest quintiles had higher forced expiratory volume in 1 second (FEV1), forced vital capacity (FVC), and FEV1/FVC ratio, thus had lower spirometry-defined prevalence of COPD.19 A prospective population-based cohort study indicated that a major component of dietary fiber, total nonstarch polysaccharides, and fiber from fruits and fruit juices was associated with reduced incidence of cough with phlegm.20 In both of these studies, the intake of fiber from vegetables was not associated with COPD symptoms.
Potential mechanisms accounting for the observed inverse association of dietary fiber on COPD risk may relate to specific anti-inflammatory effects of dietary fiber.5,8,21 In a cross-sectional study, people with high total fiber intake compared with those with low intake (range of medians between Q1 and Q4: 10.2 vs. 22.4 g/day) had a 63% (95% CI = 13%, 84%) lower risk of systemic inflammation as measured by C-reactive protein (CRP)7; the finding was confirmed in a study of over 23,000 participants (34% decrease).6 In a cross-sectional study conducted among postmenopausal women, total fiber intake was inversely associated with tumor necrosis factor α receptor 2 (TNF-R2) and interleukin 6 (IL-6), but not with CRP concentrations.21 However, in a study of diabetic women, the concentrations of TNF-R2 and CRP were 8% and 18% lower, respectively, in the highest quintile of cereal fiber intake versus the lowest, so the mechanism may vary depending upon metabolic and hormonal status.8
The reason why the inverse association between total fiber consumption and consumption of fiber from specific sources and risk of COPD was restricted to current and ex-smokers could plausibly be owing to higher oxidative stress in these groups and continued endogenous production of reactive oxygen species even after smoking cessation. Among never smokers, the pathogenic mechanisms related to COPD development may be different from those in smokers and ex-smokers; these may relate more to genetic predisposition and environmental exposures causing lower oxidative stress and inflammation than does tobacco smoke.
Strengths of the current study include the population-based prospective design, detailed information on diet, and a large number of incident COPD cases which allowed separate analyses in current smokers and ex-smokers. Nevertheless, owing to lower prevalence of COPD in never smokers, the association between fiber intake and risk of COPD in this group was imprecise. The available data on potential risk factors for COPD also allowed extensive adjustment for confounders. However, as in all observational studies, unmeasured or residual confounding cannot be ruled out, and some degree of misclassification of daily intake of total dietary fiber and cereal, fruit, and vegetable fiber intake is inevitable. Moreover, it could be possible that fiber intake reflects total quality of diet and dietary patterns; people who consume more fiber concomitantly also consume higher levels of other beneficial nutrients and foods. It has been previously reported that a prudent dietary pattern was inverse, while a Western dietary pattern was positively associated with COPD.22,23 Finally, we cannot rule out underascertainment of COPD in the early years of follow-up; however, we observed that risk estimates for the first years of follow-up were not attenuated in comparison to the later period. Moreover, we cannot rule out that some patients classified as having COPD had obtained the diagnosis without the correct spirometry assessment, even though this investigation is formally required for a clinical diagnosis of COPD. Ideally, new studies in this research area should include this type of verifying spirometry that would further strengthen the validity of the case definition.
In this prospective cohort study of men with moderate to high fiber intake, intakes of total fiber and cereal fiber were inversely associated with COPD risk in both ex-smokers and current smokers, while fruit and vegetable fiber intakes were associated with reduced risk only in current smokers. Our findings are in accordance with the previously reported inverse associations between dietary fiber intake and risk of CVD and cancer.24,25
REFERENCES
1. Global Initiative for Chronic Obstructive Lung Disease. Global Strategy for the Diagnosis, Management, and Prevention of Chronic Obstructive Pulmonary Disease (Updated 2016). Global Initiative for Chronic Obstructive Lung Disease. 2016. file:///C:/Users/a/Downloads/WatermarkedGlobal-Strategy-20161.pdf. Accessed 15 August 2016.
2. World Health Organization. Cause-Specific Mortality. Estimates for 2000–2012. Health Statistics and Information Systems. Available at:
http://www.who.int/healthinfo/global_burden_disease/estimates/en/index1.html. Accessed 5 May 2015.
3. GBD 2013 DALYs and HALE Collaborators. Global, regional, and national disability-adjusted life years (DALYS) for 306 diseases and injuries and healthy life expectancy (HALE) for 188 countries, 1990–2013: quantifying the epidemiological transition. Lancet. 2015;386:2145–2191.
4. Young RP, Hopkins RJUpdate on the potential role of statins in chronic obstructive pulmonary disease and its co-morbidities. Expert Rev Respir Med. 2013;7:533–544.
5. Fonseca Wald EL, van den Borst B, Gosker HR, Schols AMDietary fibre and fatty acids in chronic obstructive pulmonary disease risk and progression: a systematic review. Respirology. 2014;19:176–184.
6. Grooms KN, Ommerborn MJ, Pham DQ, Djoussé L, Clark CRDietary fiber intake and cardiometabolic risks among US adults, NHANES 1999-2010. Am J Med. 2013;126:1059.e1–1067.e1.
7. Ma Y, Griffith JA, Chasan-Taber L, et alAssociation between dietary fiber and serum C-reactive protein. Am J Clin Nutr. 2006;83:760–766.
8. Qi L, van Dam RM, Liu S, Franz M, Mantzoros C, Hu FBWhole-grain, bran, and cereal fiber intakes and markers of systemic inflammation in diabetic women. Diabetes Care. 2006;29:207–211.
9. Varraso R, Willett WC, Camargo CA JrProspective study of dietary fiber and risk of chronic obstructive pulmonary disease among US women and men. Am J Epidemiol. 2010;171:776–784.
10. Kaluza J, Larsson SC, Linden A, Wolk AConsumption of unprocessed and processed red meat and the risk of chronic obstructive pulmonary disease: a prospective cohort study of men. Am J Epidemiol. 2016;184:829–836.
11. Bergström L, Kylberg E, Hagman U, et alThe food consumption database KOST: The National Food Administration’s Information System for nutritive values of food. Vår Föda. 1991;43:439–447.
12. Willett W, Stampfer MJTotal energy intake: implications for epidemiologic analyses. Am J Epidemiol. 1986;124:17–27.
13. Messerer M, Johansson SE, Wolk AThe validity of questionnaire-based micronutrient intake estimates is increased by including dietary supplement use in Swedish men. J Nutr. 2004;134:1800–1805.
14. Norman A, Bellocco R, Bergström A, Wolk AValidity and reproducibility of self-reported total physical activity–differences by relative weight. Int J Obes Relat Metab Disord. 2001;25:682–688.
15. Ställberg B, Janson C, Johansson G, et alManagement, morbidity and mortality of COPD during an 11-year period: an observational retrospective epidemiological register study in Sweden (PATHOS). Prim Care Respir J. 2014;23:38–45.
16. van Buuren S, Boshuizen HC, Knook DLMultiple imputation of missing blood pressure covariates in survival analysis. Stat Med. 1999;18:681–694.
17. Royston PMultiple imputation of missing values. Stata J. 2004;4:227–241.
18. Rubin DB, Schenker NMultiple imputation for interval estimation from simple random samples with ignorable nonresponse. J Am Stat Assoc. 1986;81:366–374.
19. Kan H, Stevens J, Heiss G, Rose KM, London SJDietary fiber, lung function, and chronic obstructive pulmonary disease in the atherosclerosis risk in communities study. Am J Epidemiol. 2008;167:570–578.
20. Butler LM, Koh WP, Lee HP, Yu MC, London SJDietary fiber and reduced cough with phlegm: a cohort study in Singapore. Am J Respir Crit Care Med. 2004;170:279–287.
21. Ma Y, Hébert JR, Li W, et alAssociation between dietary fiber and markers of systemic inflammation in the Women’s Health Initiative Observational Study. Nutrition. 2008;24:941–949.
22. Varraso R, Fung TT, Barr RG, Hu FB, Willett W, Camargo CA JrProspective study of dietary patterns and chronic obstructive pulmonary disease among US women. Am J Clin Nutr. 2007;86:488–495.
23. Varraso R, Fung TT, Hu FB, Willett W, Camargo CAProspective study of dietary patterns and chronic obstructive pulmonary disease among US men. Thorax. 2007;62:786–791.
24. Threapleton DE, Greenwood DC, Evans CE, et alDietary fibre intake and risk of cardiovascular disease: systematic review and meta-analysis. BMJ. 2013;347:f6879.
25. Kim Y, Je YDietary fibre intake and mortality from cardiovascular disease and all cancers: a meta-analysis of prospective cohort studies. Arch Cardiovasc Dis. 2016;109:39–54.
