Action to stop smoking in suspected tuberculosis (ASSIST) in Pakistan: a cluster randomized, controlled trial

Ann Intern Med. 2013 May 7;158(9):667-75. doi: 10.7326/0003-4819-158-9-201305070-00006.


Background: Tobacco use is responsible for a large proportion of the total disease burden from tuberculosis. Pakistan is one of the 10 high-burden countries for both tuberculosis and tobacco use.

Objective: To assess the effectiveness of a behavioral support intervention and bupropion in achieving 6-month continuous abstinence in adult smokers with suspected pulmonary tuberculosis.

Design: Cluster randomized, controlled trial. (Current Controlled Trials: ISRCTN08829879)

Setting: Health centers in the Jhang and Sargodha districts in Pakistan.

Patients: 1955 adult smokers with suspected tuberculosis.

Intervention: Health centers were randomly assigned to provide 2 brief behavioral support sessions (BSS), BSS plus 7 weeks of bupropion therapy (BSS+), or usual care.

Measurements: The primary end point was continuous abstinence at 6 months after the quit date and was determined by carbon monoxide levels in patients. Secondary end points were point abstinence at 1 and 6 months.

Results: Both treatments led to statistically significant relative risks (RRs) for abstinence compared with usual care (RR for BSS+, 8.2 [95% CI, 3.7 to 18.2]; RR for BSS, 7.4 [CI, 3.4 to 16.4]). Equivalence between the treatments could not be established. In the BSS+ group, 275 of 606 patients (45.4% [CI, 41.4% to 49.4%]) achieved continuous abstinence compared with 254 of 620 (41.0% [CI, 37.1% to 45.0%]) in the BSS group and 52 of 615 (8.5% [CI, 6.4% to 10.9%]) in the usual care group. There was substantial heterogeneity of program effects across clusters.

Limitations: Imbalances in the urban and rural proportions and smoking habits among treatment groups, and inability to confirm adherence to bupropion treatment and validate longer-term abstinence or the effect of smoking cessation on tuberculosis outcomes.

Conclusion: Behavioral support alone or in combination with bupropion is effective in promoting cessation in smokers with suspected tuberculosis.

Primary funding source: International Development Research Centre.

Publication types

  • Multicenter Study
  • Randomized Controlled Trial
  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adolescent
  • Adult
  • Bayes Theorem
  • Behavior Therapy* / economics
  • Bupropion / adverse effects
  • Bupropion / economics
  • Bupropion / therapeutic use*
  • Drug Costs
  • Female
  • Humans
  • Male
  • Medication Adherence
  • Middle Aged
  • Pakistan
  • Rural Population
  • Smoking / adverse effects
  • Smoking Cessation / methods*
  • Smoking Prevention*
  • Tuberculosis, Pulmonary / complications*
  • Tuberculosis, Pulmonary / prevention & control
  • Urban Population
  • Young Adult


  • Bupropion

Associated data

  • ISRCTN/ISRCTN08829879