Abstract
Background and Aim
The number of patients with ulcerative colitis (UC) has been increasing in Japan. To elucidate the risk factors for developing UC in Japan, a hospital‐based case–control study was conducted. This study examined the association between smoking/drinking habits and UC onset in detail.
Methods
Cases comprised 132 Japanese patients who had been newly diagnosed with UC between 2008 and 2014 at 38 collaborating hospitals in Japan, and controls comprised 167 patients without UC. Detailed data on smoking and drinking habits were collected using a self‐administered questionnaire.
Results
Ex‐smokers showed an increasing odds ratio (OR) for UC development compared with never smokers (OR 2.42, 95% confidence interval 1.24–4.72). The ORs of ex‐smokers were particularly high among subjects aged less than 40 years, subjects who had smoked more than 10 pack‐years, and subjects who were within 13 years of quitting smoking. Regarding drinking habits, ex‐drinkers also showed a more than twofold higher OR for UC compared to never drinkers. Ex‐drinkers 40 years or older, ex‐drinkers who had consumed more than 364 drink‐years, and subjects who were less than 6 years after quitting drinking showed increased ORs for UC.
Conclusion
These findings suggest the need for careful attention for UC onset among heavy smokers who quit smoking before 40 years of age and heavy drinkers who quit drinking at ≥40 years of age.
Keywords: case–control study, drinking habits, smoking habits, ulcerative colitis
The present multicenter case‐control study using incident (new) cases of UC showed a higher OR for UC among ex‐smokers and ex‐drinkers. Our findings suggest the need for careful attention for UC onset in heavy smokers who quit smoking before 40 years old and in heavy drinkers who quit drinking at or after 40 years old.
Introduction
The incidence of ulcerative colitis (UC) has been showing an increasing trend in many parts of the world. Even in Japan, which was previously thought to have a low incidence rate, UC has been increasing over the past several decades. 1 , 2 , 3 , 4 Many genetic predispositions and environmental factors, such as appendectomy and cigarette smoking, have been suggested as factors affecting the development of UC. 5 , 6 Although the mechanisms underlying the effects of these environmental factors on the onset of UC remain largely unknown, associations between cigarette smoking status and UC development have been investigated in several studies. 7 , 8 Interestingly, the most consistent findings show that current smokers have reduced risk for UC whereas ex‐smokers have increased risk. 9 , 10 , 11 , 12 , 13 However, with respect to ex‐smokers, most associations have been reported from case–control studies in which prevalent cases (individuals previously diagnosed with UC) were used instead of incident cases (newly diagnosed cases). As a result, ex‐smokers who served as cases in those studies might actually have been smokers at the time of UC onset because of reasons such as patients quitting smoking due to gastrointestinal symptoms caused by UC. Thus, apparent risks could potentially have been detected as a result of reverse causality. In addition, if quitting smoking is confirmed as an actual risk factor for UC, identifying groups at higher risk of developing UC after quitting smoking will be important, since smoking cessation is being heavily promoted around the world to prevent various diseases.
On the other hand, the effects of drinking on the risk of UC have been examined in a small number of prior studies, 11 , 14 , 15 but the association remains controversial.
The aim of this study was therefore to explore the details of the association between smoking/drinking habits and UC in a multicenter, collaborative, hospital‐based, case–control study in Japan. In this case–control study, we minimized the problem of reverse causality from patients quitting smoking and/or drinking due to the onset of UC symptoms by prospectively registering newly diagnosed UC cases. In addition, using detailed data, we investigated which groups are strongly affected by smoking/drinking cessation in terms of the development of UC.
Methods
Subjects
The methods used for the present study have been described elsewhere. 16 , 17 , 18 Briefly, from September 2008 to March 2014, a multicenter case–control study was conducted with the cooperation of 38 facilities in Japan.
Cases were prospectively registered among Japanese patients under 80 years of age who were newly diagnosed with UC according to the diagnostic criteria proposed by the Research Committee of Inflammatory Bowel Disease in 2008. Patients referred from other hospitals were allowed to register if they had been diagnosed with UC within the preceding 3 months.
Regarding controls, as far as possible, two patients with matching sex and age within the same 5‐year age category were recruited for each case, from the same hospital as the case. We asked for researchers to select one control from the gastroenterology department and the other one from another department such as orthopedics, ophthalmology, or a general medical department. Exclusion criteria were the presence of malignancy, persistent symptoms of diarrhea and abdominal pain for more than 1 week, or a history of inflammatory bowel disease (IBD).
Collection of information
For cases, the following clinical information was collected by physicians using a standardized questionnaire: date of registration to this study, date of first visit to the hospital, date of symptom onset, date of diagnosis, disease severity at diagnosis (mild, moderate, severe, or fulminant), and disease site at diagnosis (rectum, colon, cecum, or ileum). In addition, the following information was obtained from a self‐administered questionnaire: demographic characteristics, height, weight, history of appendicitis, family history of UC, smoking habit, and drinking habit. Information on smoking habit was collected in the following three categories: non‐smoker, ex‐smoker, or current smoker. For ex‐smokers, the age at quitting smoking and the reasons for quitting were also asked. Furthermore, for ex‐smokers and current smokers, the age at which smoking started and the average number of cigarettes smoked daily were investigated. Information on drinking habits was collected in the following three categories: non‐drinker, ex‐drinker, or current drinker. For ex‐drinkers, the age at quitting drinking and the reasons for quitting were also investigated. Furthermore, for ex‐drinkers and current drinkers, the age at which drinking started and drinking habits during the “peak” period of highest alcohol consumption were investigated. We assessed drinking frequency, period (age at starting, age at stopping), and volume of alcohol intake according to beverage types.
Analysis
Because this was a hospital‐based case–control study, subjects might have quit smoking or drinking because they developed an illness. Subjects who had quit smoking or drinking within 1.5 years before study entry were thus defined as current smokers or current drinkers. Family history of UC was examined in relatives up to the second degree.
The cumulative amount of cigarette smoking was calculated using the following equation: pack‐years of cigarettes smoking (pack‐years) = average number of cigarettes smoked daily/20 (assuming 20 cigarettes per pack) × duration of smoking (years). The cumulative amount of alcohol consumption during the “peak” period was calculated using the following equation: cumulative alcohol consumption (drink‐years) = amount of ethanol consumed each week (g) × duration of “peak” period in which alcohol consumption was highest (years). These variables were classified where possible as quartiles (or dichotomies) according to the distribution of control subjects.
Logistic regression model was used to calculate the odds ratio (OR) and 95% confidence interval (95%CI) of each variable for UC development. Trends for association were assessed by assigning ordinal scores to each category. In multivariate analysis, the model included variables that showed significantly different distributions when comparing cases and controls and variables suggested as risk factors for UC from previous studies.
Sub‐analysis classified subjects into two or three groups according to sex (male/female), age (<40 years/≥40 years), severity (mild/moderate or severe), and disease site (rectum/colon/appendix or ileum). Non‐smokers or non‐drinkers were used as references to investigate the associations between smoking or drinking habits and UC.
Next, the effects of past smoking on the development of UC among ex‐smokers and the effects of past drinking on the development of UC among ex‐drinkers were analyzed in detail. Subjects of analysis were non‐smokers and ex‐smokers, and non‐drinkers and ex‐drinkers, respectively, excluding current smokers and current drinkers.
The level of significance was set at 0.05. Statistical analyses were conducted using SAS version 9.3 (SAS Institute Inc., Cary, NC).
Ethical considerations
The study protocol was approved by the ethics committees at Osaka City University Graduate School of Medicine and all other collaborating hospitals, and the study was performed in accordance with the Declaration of Helsinki. Written informed consent was obtained from all subjects prior to enrolment. In the case of subjects under 20 years of age, written informed consent was obtained from their legal representatives.
Results
Of the 151 cases and 199 controls who met the inclusion criteria, responses to the questionnaire were obtained from 133 cases and 167 controls (response rates: 88% and 84%, respectively). Of these, one case who had not described his drinking habit in the self‐administered questionnaire was excluded. As a result, 132 cases and 167 controls were included in the final analysis. In terms of matching conditions for these 299 participants, 1:2 matching was achieved in 43 pairs (43 cases, 86 controls) and 1:1 matching in 43 pairs (43 cases, 43 controls). The remaining 46 cases and 38 controls were unmatched. When conditional logistic modeling was used, the number of subjects for analysis was reduced to 215. As a result, 299 subjects (132 cases, 167 controls) were analyzed using an unconditional logistic model, which included matching factors (age and sex).
Controls were selected from gastroenterology departments and other departments in approximately 1:1 ratio. The other departments were most frequently orthopedic surgery, followed by general medicine, nephrology, ophthalmology, rheumatology, and others.
Table 1 shows disease characteristics for the cases. Median age at symptom onset was 41.4 years. Median time from symptom onset to enrollment was 2.4 months, and nearly 90% of cases were enrolled within 1 year after symptom onset. As for disease severity at diagnosis, 16% of cases were severe. The site of UC at the time of diagnosis was rectal in 21% and appendix/ileum in about 30%.
Table 1.
Disease characteristics of cases (N = 132)
| n | (%) | ||
|---|---|---|---|
| Age at symptom onset (years) | Median (range) | 41.4 | (8.7–74.8) |
| <30 | 23 | (24) | |
| 30–39 | 22 | (23) | |
| 40–49 | 24 | (26) | |
| ≥50 | 25 | (27) | |
| Missing | 38 | ||
| Time from onset to enrollment (months) | Median (range) | 2.4 | (0–172.8) |
| <4 | 63 | (67) | |
| 4–11 | 21 | (22) | |
| 12–17 | 4 | (4) | |
| ≥18 | 6 | (6) | |
| Unknown | 38 | ||
| Time from initial examination to enrollment (months) | Median (range) | 1.2 | (0–36.0) |
| <4 | 110 | (86) | |
| 4–11 | 14 | (11) | |
| ≥12 | 4 | (3) | |
| Unknown | 4 | ||
| Severity | Mild | 40 | (41) |
| Moderate | 41 | (42) | |
| Severe | 16 | (16) | |
| Fulminant | 0 | (0) | |
| Unknown | 35 | ||
| Disease site | Rectum | 21 | (21) |
| Colon | 44 | (45) | |
| Appendix | 30 | (31) | |
| Ileum | 3 | (3) | |
| Unknown | 34 |
Table 2 shows the factors associated with the development of UC (main analysis). First, we conducted a multivariate analysis (Model 1) in which the model included body mass index (BMI), history of appendicitis, family history of UC, smoking habits, drinking habits, and matching factors (age, sex). Compared to non‐smokers, ex‐smokers showed an increased OR (OR 2.42, 95% CI 1.24–4.72, P = 0.010). OR appeared to be increased in ex‐drinkers compared to non‐drinkers (OR 2.53, 95% CI 0.89–7.19, P = 0.083). Model 1 may have been unstable because of the small number of individuals with a “present” history of appendicitis and a “present” family history of UC. When we examined Model 2 excluding these two variables, the results were almost the same as those for Model 1, so Model 2 was adopted for subsequent analyses.
Table 2.
Factors associated with the development of ulcerative colitis (main analysis)
| Cases (N = 132) | Controls (N = 167) | Model 1 † | Model 2 ‡ | ||||
|---|---|---|---|---|---|---|---|
| Variable | n (%) | n (%) | OR (95%CI) | P‐value | OR (95%CI) | P‐value | |
| Age (years) | <30 | 33 (25) | 38 (23) | 1 | 1 | ||
| 30–39 | 34 (26) | 40 (24) | 0.98 (0.48–1.99) | 0.957 | 1.03 (0.51–2.06) | 0.944 | |
| 40–49 | 31 (23) | 42 (25) | 1.04 (0.50–2.16) | 0.920 | 0.91 (0.44–1.87) | 0.795 | |
| ≥50 | 34 (26) | 47 (28) | 0.83 (0.40–1.73) | 0.622 | 0.70 (0.34–1.44) | 0.335 | |
| (Trend P = 0.67) | (Trend P = 0.30) | ||||||
| Sex | Male | 76 (58) | 84 (50) | 1 | 1 | ||
| Female | 56 (42) | 83 (50) | 0.68 (0.39–1.19) | 0.174 | 0.66 (0.38–1.14) | 0.131 | |
| BMI (kg/m2) | <20.8 | 69 (52) | 53 (32) | 1 | 1 | ||
| 20.8–23.5 | 36 (27) | 57 (34) | 0.39 (0.21–0.72) | 0.003 | 0.42 (0.23–0.76) | 0.004 | |
| ≥23.6 | 27 (20) | 57 (34) | 0.29 (0.15–0.57) | <0.001 | 0.28 (0.14–0.54) | <0.001 | |
| (Trend P < 0.001) | (Trend P < 0.001) | ||||||
| History of appendicitis | Absent | 124 (94) | 140 (84) | 1 | |||
| Present | 8 (6) | 27 (16) | 0.38 (0.16–0.95) | 0.038 | |||
| Family history of ulcerative colitis | Absent | 123 (93) | 162 (97) | 1 | |||
| Present | 9 (7) | 5 (3) | 2.87 (0.82–10.0) | 0.099 | |||
| Smoking habit | No | 67 (51) | 99 (59) | 1 | 1 | ||
| Ex‐smoker | 45 (34) | 29 (17) | 2.42 (1.24–4.72) | 0.010 | 2.65 (1.37–5.13) | 0.004 | |
| Current smoker | 20 (15) | 39 (23) | 0.74 (0.37–1.48) | 0.397 | 0.80 (0.40–1.58) | 0.516 | |
| (Trend P = 0.70) | (Trend P = 0.88) | ||||||
| Drinking habit | No | 37 (28) | 62 (37) | 1 | 1 | ||
| Ex‐drinker | 15 (11) | 9 (5) | 2.53 (0.89–7.19) | 0.083 | 2.20 (0.81–5.98) | 0.121 | |
| Current drinker | 80 (61) | 96 (57) | 1.24 (0.71–2.16) | 0.447 | 1.19 (0.69–2.05) | 0.532 | |
| (Trend P = 0.55) | (Trend P = 0.63) | ||||||
Model 1 includes BMI, history of appendicitis, family history of ulcerative colitis, smoking habit, drinking habit, and matching factors (age, sex).
Model 2 includes BMI, smoking habit, drinking habit, and matching factors (age, sex).
We created subgroups for each sex, age, severity, and disease site, and examined the association between smoking/drinking habits and development of UC (Table 3, subgroup analysis). Variables included in the model were the same as in Model 2 of the main analysis. For ex‐smokers compared with non‐smokers, males in the sex subgroup (OR 2.84, 95% CI 1.17–6.90), younger than 40 years in the age subgroup (OR 9.57, 95% CI 2.67–34.3), moderate or severe in the severity subgroup (OR 4.83, 95% CI 2.02–11.6), and appendix or ileum in the disease site subgroup (OR 7.26, 95% CI 2.46–21.4) showed increased ORs for the development of UC. When comparing current smokers and non‐smokers, no subgroups were associated with the development of UC.
Table 3.
Association between smoking and drinking habits and development of ulcerative colitis (subgroup analyses)
| Smoking habit | Drinking habit | ||||||
|---|---|---|---|---|---|---|---|
| Adjusted OR (95%CI) (ref. Non‐smoker) | Adjusted OR (95%CI) (ref. Non‐drinker) | ||||||
| Ex‐smoker | Current smoker | P for trend | Ex‐drinker | Current drinker | P for trend | ||
| All participants † | (132 cases/167 controls) | 2.65 (1.37–5.13) | 0.80 (0.40–1.58) | 0.88 | 2.20 (0.81–5.98) | 1.19 (0.69–2.05) | 0.63 |
| Sex ‡ | |||||||
| Male | (76 cases/84 controls) | 2.84 (1.17–6.90) | 1.10 (0.44–2.73) | 0.84 | 2.36 (0.54–10.3) | 1.41 (0.64–3.15) | 0.50 |
| Female | (56 cases/83 controls) | 2.67 (0.84–8.54) | 0.39 (0.11–1.34) | 0.41 | 1.98 (0.47–8.30) | 1.03 (0.47–2.24) | 0.96 |
| Age (years) | |||||||
| <40 § | (67 cases/78 controls) | 9.57 (2.67–34.3) | 0.89 (0.33–2.37) | 0.60 | 0.77 (0.13–4.48) | 0.87 (0.40–1.93) | 0.75 |
| ≥40 ¶ | (65 cases/89 controls) | 1.23 (0.51–2.96) | 0.58 (0.21–1.60) | 0.34 | 5.26 (1.32–21.0) | 1.50 (0.68–3.28) | 0.46 |
| Severity † | |||||||
| Mild | (40 cases/167 controls) | 1.55 (0.59–4.07) | 0.42 (0.12–1.44) | 0.26 | 3.48 (0.71–17.1) | 2.13 (0.84–5.40) | 0.15 |
| Moderate/severe | (57 cases/167 controls) | 4.83 (2.02–11.6) | 1.16 (0.46–2.96) | 0.38 | 3.02 (0.91–10.1) | 0.84 (0.40–1.78) | 0.53 |
| Disease site † | |||||||
| Rectum | (21 cases/167 controls) | 1.43 (0.42–4.84) | 0.35 (0.07–1.77) | 0.27 | 1.82 (0.17–19.4) | 2.38 (0.71–8.06) | 0.16 |
| Colon | (44 cases/167 controls) | 2.36 (0.93–6.01) | 0.55 (0.17–1.82) | 0.64 | 1.86 (0.43–8.12) | 0.91 (0.41–2.03) | 0.77 |
| Appendix/ileum | (33 cases/167 controls) | 7.26 (2.46–21.4) | 1.98 (0.64–6.10) | 0.14 | 5.42 (1.38–21.3) | 0.98 (0.37–2.62) | 0.72 |
Adjusted for BMI, smoking habit, drinking habit, and matching factors (age, sex).
Adjusted for BMI, smoking habit, drinking habit, and matching factors (age).
Adjusted for BMI, smoking habit, drinking habit, and matching factors (age: <30 or 30–39, sex).
Adjusted for BMI, smoking habit, drinking habit, and matching factors (age: 40–49 or ≥50, sex).
When comparing ex‐drinkers and non‐drinkers, 40 years and older in the age subgroup (OR 5.26, 95% CI 1.32–21.0) and appendix or ileum in the disease site subgroup (OR 5.42, 95% CI 1.38–21.3) showed increased ORs for the development of UC. When comparing current drinkers with non‐drinkers, no subgroup was associated with development of UC.
Next, we examined association between ex‐smoking and ex‐drinking and the development of UC (Table 4). The variables included in this model were the same as in Model 2 of the main analysis. In ex‐smokers, the OR for the development of UC increased with increasing cumulative number of cigarettes smoked (trend P = 0.001) and increased significantly beyond 10 pack‐years. OR for the time since quitting was significantly higher for 1–5 years (OR 6.46, 95% CI 2.37–17.6) and 5–13 years (OR 3.26, 95% CI 1.18–9.00), but it did not increase for ≥13 years.
Table 4.
Association between ex‐smoking and ex‐drinking and development of ulcerative colitis
| Cases | Controls | ||||
|---|---|---|---|---|---|
| n (%) | n (%) | OR † (95%CI) | P‐value | ||
| Cumulative cigarette smoking (pack‐years) | |||||
| Non‐smoker | 67 (60) | 99 (77) | 1.00 | ||
| Ex‐smoker | <4.0 | 11 (10) | 7 (5) | 2.51 (0.87–7.27) | 0.091 |
| 4.0–9.9 | 7 (6) | 7 (5) | 1.83 (0.56–5.96) | 0.319 | |
| 10.0–21.0 | 10 (9) | 7 (5) | 3.17 (1.00–10.0) | 0.049 | |
| ≥21.1 | 17 (15) | 8 (6) | 5.25 (1.70–16.3) | 0.004 | |
| (Trend P = 0.001) | |||||
| Time since quitting (years) | |||||
| Non‐smoker | 67 (60) | 99 (77) | 1.00 | ||
| Ex‐smoker | 1.5–5.0 | 23 (21) | 7 (5) | 6.46 (2.37–17.6) | <0.001 |
| 5.1–13.0 | 16 (14) | 8 (6) | 3.26 (1.18–9.00) | 0.023 | |
| 13.1–22.0 | 2 (2) | 7 (5) | 0.52 (0.09–2.84) | 0.447 | |
| >22 | 4 (4) | 7 (5) | 0.97 (0.23–4.09) | 0.970 | |
| (Trend P = 0.389) | |||||
| Cumulative alcohol consumption (drink‐years) | |||||
| Non‐drinker | 37 (71) | 62 (89) | 1.00 | ||
| Ex‐drinker | <364 | 3 (6) | 4 (6) | 0.97 (0.16–5.97) | 0.973 |
| ≥364 | 12 (23) | 4 (6) | 5.27 (0.97–28.5) | 0.054 | |
| Unknown | 1 | ||||
| (Trend P = 0.075) | |||||
| Time since quitting (years) | |||||
| Non‐drinker | 37 (71) | 62 (87) | 1.00 | ||
| Ex‐drinker | 1.5–<6 | 11 (21) | 5 (7) | 3.85 (0.88–16.9) | 0.074 |
| ≥6 | 4 (8) | 4 (6) | 1.01 (0.17–6.15) | 0.989 | |
| (Trend P = 0.387) | |||||
Adjusted for BMI, smoking habit or drinking habit, and matching factors (age, sex).
An increased OR for the development of UC was suggested in ex‐drinkers who consumed a large amount of alcohol (≥364 drink‐years in cumulative alcoholic consumption) (OR 5.27, 95% CI 0.97–28.5). In terms of the time since quitting drinking, OR was marginally increased for less than 6 years (OR 3.85, 95% CI 0.88–16.9) but approached 1 for 6 years or more.
Discussion
An association between the development of UC and ex‐smoking/ex‐drinking was suggested in the present analysis. Subgroup analysis showed the effects of previous smoking to be stronger in men or those younger than 40 years and the effects of previous drinking to be stronger in those 40 years or older. The effects of smoking and drinking cessation were more pronounced in participants with more severe or more wide‐ranging disease, indicating that cessation may also impact the severity and extent of UC. A detailed investigation of smoking cessation showed that the ORs for ex‐smokers were particularly high among heavy smokers and those who had quit within the past 5 years. A detailed investigation of the cessation of alcohol consumption showed that the ORs for ex‐drinkers were particularly high among heavy drinkers and those who had quit within the past 5 years. The association between ex‐drinkers and UC onset represents particularly an important result that has not been reported previously.
The higher OR identified in ex‐smokers agrees with findings from previous studies. 9 , 10 , 11 , 12 , 13 Since the present study included only incident (new) cases of UC, higher OR in ex‐smokers was unlikely to have resulted from reverse causality. The OR of participants with a short period after smoking cessation was increased, and the longer the period since smoking cessation, the closer the OR was to that of non‐smokers. These findings suggest that smoking cessation is associated with UC onset, as reported previously. The mechanism may involve nicotine decreasing blood flow during smoking, inhibiting inflammatory mediators from reaching the rectal mucosal surface, and suppressing inflammatory responses, with subsequent smoking cessation transiently increasing the immune function. 10 Therefore, when a person with a large cumulative load of cigarettes smoked quits smoking at <40 years of age, careful follow‐up may be required for around 5 years after quitting. Given the results of previous studies, use of nicotine patches may be beneficial in the treatment of UC when quitting smoking. 19 , 20 Recent reviews have shown that transdermal nicotine, in combination with conventional treatments, is more beneficial in the treatment of UC than conventional treatments alone, but further studies are needed to determine the appropriate doses. 21 On the other hand, a recent study based on smoking status as of 2 years before the diagnosis of UC showed that smoking cessation was not associated with a worse course of illness. 22 The present analysis did not find a lower OR in current smokers.
In this study, the OR of ex‐drinkers ≥40 years was increased. A few studies have examined the association between drinking habits and development of UC, but the conclusions were inconsistent. 11 , 14 , 15 , 23 Since the present study included only incident (new) cases of UC, the higher OR in ex‐drinkers was unlikely to have resulted from reverse causality. The ORs of individuals with a large cumulative alcohol intake or with a short period of abstinence were increased. ORs of those with a small amount of cumulative drinking or a long period of abstinence decreased to the same level as never drinkers. In the future, detailed examinations of the underlying mechanisms will be required, but when an individual with a large cumulative alcohol intake abstains from alcohol above the age of 40 years, follow‐up will be required for around 6 years after quitting drinking.
In our study, in the case of BMI, which we used as an adjusted factor, we identified an association between low BMI and development of UC. BMI information was taken at the time of enrollment, and 89% of cases were enrolled within 1 year of onset, and no association was apparent between BMI and severity or disease site. However, a prospective European cohort study found no association between BMI and development of UC. 24 In a recent systematic review, both underweight and obese status were independently associated with Crohn's disease, but BMI was not significantly associated with UC. 25
The present study has three main strengths. First, the possibility of reverse causality was minimized by including only patients with newly diagnosed UC. Second, rigorous enrollment criteria were established to ensure that only patients with newly diagnosed UC were included. As detailed in previous studies, 16 , 17 , 18 researchers in gastroenterology in this study reached definitive diagnoses according to the diagnostic criteria of this research group, resulting in high diagnostic reliability. Third, the careful study design enabled the use of detailed questions, resulting in the collection of detailed information. Cohort studies of low‐incidence diseases such as UC require very large cohort sizes to ensure outcome occurrence that will allow valid statistical analysis. Time as well as financial and human resource constraints generally make such studies unfeasible. In this study, we could obtain detailed information by using a case–control study design with the cooperation of study subjects and the gastroenterologist in charge.
The present study has four main limitations. The first involves recall bias. The accuracy of the self‐reported information collected for smoking and drinking histories may differ according to how well the participants remembered this information. The degree of recall of smoking and drinking histories between subjects and their controls, however, was likely comparable because both groups were outpatients of comparable age receiving care at medical institutions for a disease (hospital‐based case–control study). Any misclassification of smoking and drinking histories caused by improper recall would likely have balanced out between subjects and their controls (i.e., amounting to nondifferential misclassification) and thus would not have affected the results. Second, controls were selected from the outpatient services of tertiary medical institutions and may have had characteristics differing from those of the general population. This issue was addressed by excluding certain diseases that could have affected the results (malignancies, Crohn's disease) and including patients with a wide range of conditions in addition to patients with a gastroenterologic disease as controls. Additionally, the proportion of current smokers in the control group was 29% for males and 16% for females, which was not considered to differ significantly from distributions in the general population (adult smoking rates in Japan from 2008 to 2013 were 32–38% for men and 8–11% for women 26 ). Third, we may not have been able to detect relevant factors due to the small sample size. Sample size calculations as a post hoc analysis showed that a total of 281 subjects (94 cases and 187 controls) were required to detect a significant association between ex‐smoking and UC development with an α level of 0.05 and a β level of 0.20. To detect any significant association between ex‐drinking and UC development, a total of 440 subjects (147 cases and 293 controls) were required. Therefore, the number of subjects may have been small to investigate the association between ex‐drinking and UC development. Finally, lifestyles such as smoking and drinking habits are thought to be strongly influenced by socioeconomic factors, but such factors could not be included in the model as adjusted factors in this study.
Conclusions
The present multicenter case–control study using incident (new) cases of UC showed a higher OR for UC among ex‐smokers and ex‐drinkers. Our findings suggest the need for careful attention for UC onset in heavy smokers who quit smoking before 40 years of age and in heavy drinkers who quit drinking at or after 40 years.
Acknowledgments
Other members of the Japanese Case–Control Study Group for Ulcerative Colitis are as follows (shown in alphabetical order of affiliation): Masahiro Iizuka (Akita Health Care Center, Akita Red Cross Hospital), Yutaka Kohgo and Yuhei Inaba (Division of Gastroenterology and Hematology/Oncology, Department of Medicine, Asahikawa Medical College), Takashi Hisabe and Toshiyuki Matsui (Department of Gastroenterology, Fukuoka University Chikushi Hospital), Kitaro Futami (Department of Surgery, Fukuoka University Chikushi Hospital), Hiroyuki Hanai (Center for Gastroenterology and IBD Research, Hamamatsu South Hospital), Yoh Ishiguro (Department of Endoscopy, Department of Gastroenterology and Hematology, Hirosaki University Graduate School of Medicine), Shinji Tanaka and Yoshitaka Ueno (Department of Endoscopy, Hiroshima University Hospital), Ken Fukunaga and Takayuki Matsumoto (Division of Lower Gastroenterology, Department of Internal Medicine, Hyogo College of Medicine), Hiroki Ikeuchi (Inflammatory Bowel Disease Center, Hyogo College of Medicine), Hiroshi Fujita and Hirohito Tsubouchi (Digestive and Lifestyle Diseases, Kagoshima University Graduate School of Medical and Dental Sciences), Kazuichi Okazaki (Division of Gastroenterology and Hepatology, Third Department of Internal Medicine, Kansai Medical University), Kazuhiko Yoshioka (Department of Gastroenterology and Hepatology, Kansai Medical University Kouri Hospital), Nagamu Inoue and Toshifumi Hibi (Division of Gastroenterology and Hepatology, Department of Internal Medicine, Keio University School of Medicine), Kiyonori Kobayashi and Kaoru Yokoyama (Department of Gastroenterology, Kitasato University School of Medicine), Hiroshi Yamasaki and Keiichi Mitsuyama (Inflammatory Bowel Disease Center, Division of Gastroenterology, Department of Medicine, Kurume University School of Medicine), Yuji Naito (Molecular Gastroenterology and Hepatology, Graduate School of Medical Science, Kyoto Prefectural University of Medicine), Tsutomu Chiba and Hiroshi Nakase (Department of Gastroenterology and Hepatology, Kyoto University Graduate School of Medicine), Masato Kusunoki (Department of Gastrointestinal and Pediatric Surgery, Mie University Graduate School of Medicine), Haruhiko Inatsu (Department of Internal Medicine, Circulatory and Body Fluid Regulation, Miyazaki University Faculty of Medicine), Shojiro Yamamoto (Division of Gastroenterology and Hematology, Department of Internal Medicine, Miyazaki University Faculty of Medicine), Hisao Fujii (Department of Endoscopy and Ultrasound, Nara Medical University Hospital), Ryota Hokari and Soichiro Miura (Department of Internal Medicine, National Defense Medical College), Kazuhito Sugimura (Department of Gastroenterology and Hepatology, Niigata City General Hospital), Hideki Iijima (Department of Gastroenterology and Hepatology, Osaka University Faculty of Medicine), Satoshi Motoya (IBD Center, Sapporo Kosei General Hospital), Akira Andoh (Department of Medicine, Shiga University of Medical Science), Yoshihide Fujiyama (Shiga University of Medical Science), Shunji Ishihara (Department of Internal Medicine II, Shimane University Faculty of Medicine), Shin‐Ei Kudo and Noriyuki Ogata (Digestive Disease Center, Showa University, Northern Yokohama Hospital), Naoki Yoshimura (Department of Internal Medicine, Social Insurance Chuo General Hospital), Toshiaki Watanabe (Department of Surgical Oncology, Tokyo University), Kazuo Ohtsuka and Masakazu Nagahori (Department of Gastroenterology and Hepatology, Tokyo Medical and Dental University), Shingo Kameoka and Michio Itabashi (Department of Surgery II, Tokyo Women's Medical University), Yuji Funayama (Department of Colorectal Surgery, Tohoku Rosai Hospital), Fukunori Kinjo (Department of Endoscopy, University Hospital, University of the Ryukyus), Atsuo Kitano (Department of Gastroenterology, Wakakusa First Hospital), Atsushi Nakajima, Hirokazu Takahashi, and Takuma Higurashi (Division of Gastroenterology, Yokohama City University School of Medicine), and Akira Sugita (Department of Surgery, Yokohama Municipal Citizen's Hospital).
Other members of the Japanese Case‐Control Study Group for Ulcerative Colitis are listed in Acknowledgments.
Financial support: This work was supported by a research grant for Research on Intractable Disease, Health and Labor Sciences Research Grants from the Ministry of Health, Labor and Welfare, Japan (http://stf/stf/seisakunitsuite/bunya/hokabunya/kenkyujigyou/). The funder had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Declaration of conflict of interest: Dr. Kenji Watanabe has received grants or contracts from EA Pharma, Takeda and EP CRSU and has also received payment or honoraria for lectures, presentations, speakers’ bureaus, manuscript writing or educational events from JIMRO, EA Pharma, Takeda and AbbVie Japan. The other authors have nothing to disclose regarding funding or conflicts of interest with regard to this manuscript.
References
- 1. Cosnes J, Gower‐Rousseau C, Seksik P et al. Epidemiology and natural history of inflammatory bowel diseases. Gastroenteroloy. 2011; 140: 1785–94. [DOI] [PubMed] [Google Scholar]
- 2. Murakami Y, Nishiwaki Y, Oba MS et al. Estimated prevalence of ulcerative colitis and Crohn's disease in Japan in 2014: an analysis of a nationwide survey. J. Gastroenterol. 2019; 54: 1070–7. [DOI] [PubMed] [Google Scholar]
- 3. Murakami Y, Nishiwaki Y, Oba MS et al. Correction to: Estimated prevalence of ulcerative colitis and Crohn's disease in Japan in 2015: an analysis of a nationwide survey. J. Gastroenterol. 2020; 55: 131. [DOI] [PubMed] [Google Scholar]
- 4. Asakura K, Nishiwaki Y, Inoue N, Hibi T, Watanabe M, Takebayashi T. Prevalence of ulcerative colitis and Crohn's disease in Japan. J. Gastroenterol. 2009; 44: 659–65. [DOI] [PubMed] [Google Scholar]
- 5. Piovani D, Danese S, Peyrin‐Biroulet L, Nikolopoulos GK, Lytras T, Bonovas S. Environmental Risk factors for inflammatory bowel diseases: an umbrella review of meta‐analyses. Gastroenterology. 2019; 157: 647–59. [DOI] [PubMed] [Google Scholar]
- 6. Loftus EV. Clinical epidemiology of inflammatory bowel disease: incidence, prevalence, and environmental influences. Gastroenterology. 2004; 126: 1504–17. [DOI] [PubMed] [Google Scholar]
- 7. Mahid SS, Minor KS, Soto RE, Hornung CA, Galandiuk S. Smoking and inflammatory bowel disease: a meta‐analysis. Mayo Clin. Proc. 2006; 81: 1462–71. [DOI] [PubMed] [Google Scholar]
- 8. Salih A, Widbom L, Hultdin J, Karling P. Smoking is associated with risk for developing inflammatory bowel disease including late onset ulcerative colitis: a prospective study. Scand. J. Gastroenterol. 2018; 53: 173–8. [DOI] [PubMed] [Google Scholar]
- 9. Park S, Chun J, Han KD et al. Dose‐response relationship between cigarette smoking and risk of ulcerative colitis: a nationwide population‐based study. J. Gastroenterol. 2019; 54: 881–90. [DOI] [PubMed] [Google Scholar]
- 10. Abraham N, Selby W, Lazarus R et al. Is smoking an indirect risk factor for the development of ulcerative colitis? An age‐ and sex‐matched case‐control study. J. Gastroenterol. Hepatol. 2003; 18: 139–46. [DOI] [PubMed] [Google Scholar]
- 11. Nakamura Y, Labarthe DR. A case‐control study of ulcerative colitis with relation to smoking habits and alcohol consumption in Japan. Am. J. Epidemiol. 1994; 140: 902–11. [DOI] [PubMed] [Google Scholar]
- 12. Boyko EJ, Koepsell TD, Perera DR, Inui TS. Risk of ulcerative colitis among former and current cigarette smokers. N. Engl. J. Med. 1987; 316: 707–10. [DOI] [PubMed] [Google Scholar]
- 13. Takahashi H, Matsui T, Hisabe T et al. Second peak in the distribution of age at onset of ulcerative colitis in relation to smoking cessation. J. Gastroenterol. Hepatol. 2014; 29: 1603–8. [DOI] [PubMed] [Google Scholar]
- 14. Boyko EJ, Perera DR, Koepsell TD, Keane EM, Inui TS. Coffee and alcohol use and the risk of ulcerative colitis. Am. J. Gastroenterol. 1989; 84: 530–4. [PubMed] [Google Scholar]
- 15. Bergmann MM, Hernandez V, Bernigau W et al. No association of alcohol use and the risk of ulcerative colitis or Crohn's disease: data from a European Prospective cohort study (EPIC). Eur. J. Clin. Nutr. 2017; 71: 512–8. [DOI] [PubMed] [Google Scholar]
- 16.Ohfuji S, Fukushima W, Watanabe K et al. Pre‐illness isoflavone consumption and disease risk of ulcerative colitis: a multicenter case‐control study in Japan. PLoS One. 2014; 9: e110270. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Kobayashi Y, Ohfuji S, Kondo K et al. Association between dietary iron and zinc intake and development of ulcerative colitis: a case‐control study in Japan. J. Gastroenterol. Hepatol. 2019; 34: 1703–10. [DOI] [PubMed] [Google Scholar]
- 18.Kobayashi Y, Ohfuji S, Kondo K et al. Association of dietary fatty acid intake with the development of ulcerative colitis: a multicenter case‐control study in Japan. Inflamm. Bowel Dis. 2021; 27: 617–28. [DOI] [PubMed] [Google Scholar]
- 19.Pullan RD, Rhodes J, Ganesh S et al. Transdermal nicotine for active ulcerative colitis. N. Engl. J. Med. 1994; 330: 811–5. [DOI] [PubMed] [Google Scholar]
- 20. Thomas GAO, Rhodes J, Mani V et al. Transdermal nicotine as maintenance therapy for ulcerative colitis. N. Engl. J. Med. 1995; 332: 988–92. [DOI] [PubMed] [Google Scholar]
- 21.Kannichamy V, Antony I, Mishra V et al. Transdermal nicotine as a treatment option for ulcerative colitis: a review. Cureus. 2020; 12: e11096. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Blackwell J, Saxena S, Alexakis C et al. The impact of smoking and smoking cessation on disease outcomes in ulcerative colitis: a nationwide population‐based study. Aliment. Pharmacol. Ther. 2019; 50: 556–67. [DOI] [PubMed] [Google Scholar]
- 23.Georgiou AN, Ntritsos G, Papadimitriou N, Dimou N, Evangelou E. Cigarette smoking, coffee consumption, alcohol intake, and risk of Crohn's disease and ulcerative colitis: a Mendelian randomization study. Inflamm. Bowel Dis. 2021; 27: 162–8. [DOI] [PubMed] [Google Scholar]
- 24.Chan SS, Luben R, Olsen A et al. Body mass index and the risk for Crohn's disease and ulcerative colitis: data from a European Prospective Cohort Study (The IBD in EPIC Study). Am. J. Gastroenterol. 2013; 108: 575–82. [DOI] [PubMed] [Google Scholar]
- 25.Bhagavathula AS, Clark CCT, Rahmani J, Chattu VK. Impact of body mass index on the development of inflammatory bowel disease: a systematic review and dose‐response analysis of 15.6 million participants. Healthcare (Basel). 2021; 9: 35. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Adult smoking rate (Ministry of Health, Labor and Welfare National Health and Nutrition Survey) [Translated from Japanese.] https://www.health-net.or.jp/tobacco/statistics/kokumin_kenkou_eiyou_report.html