Skip to main content

Table 2 Characteristics of included studies reporting on the association between smoking and colectomy in ulcerative colitis

From: Smoking influences the need for surgery in patients with the inflammatory bowel diseases: a systematic review and meta-analysis incorporating disease duration

Study

Country

Source of Patients

Definition of Smoking

Timing of Smoking

Duration of Follow-up

Sample Size

Adjusted Estimates

Beaugerie 2001 [15]

France

Tertiary care

Current: >7 cigarettes/week for ≥6 months after diagnosis

Former: Quit for >12 months at time of evaluation

At colectomy or study enrolment

Mean:

Current smokers: 10 years

Former smokers: 16 years

Never smokers: 14 years

96a

Crudeb

Boyko 1988 [16]

USA

HMO

Current: Smoked >100 cigarettes and continued to smoke or initiated smoking after diagnosis

Former: Smoked >100 cigarettes and quit prior to diagnosis and did not restart

Never: Smoked <100 cigarettes

At diagnosis

Median (range): 8 years (1, 35)

206

Age; sex

Frolkis 2016 [14]

UK

THIN database

Patient coded in EMR as Current, Former, or Never Smoker within one year of index (diagnosis date)

At diagnosis

Median (Q1, Q3): 5 years (3, 8)

3600

Age at diagnosis; sex; use of immunosuppressants; steroid use within 90 days of diagnosis

Hoie 2007 [17]

Europec

Population-basedd

Current: Maintained same smoking behaviour throughout follow-up

Former: Dates when patient started and stopped smoking were compared with year of disease onset and colectomy

Never: Patients who had never been daily cigarette smokerse

At diagnosis

Median (range): 10 (9, 12) years

771

Crude

  1. HMO health maintenance organization
  2. a32 former smokers were matched with 32 current smokers and 32 never smokers
  3. bCrude hazard ratios were estimated from Kaplan-Meier curves using the method proposed by Guyot et al.[25]
  4. cGreece, Israel, Italy, Spain, Denmark, the Netherlands, Norway
  5. dAdditional information on study methodology obtained from Shivananda et al.[32]
  6. eStudy also reports a hazard ratio for patients with unknown smoking status relative to individuals who never smoked