Date of Award

January 2022

Document Type


Degree Name

Medical Doctor (MD)



First Advisor

Lori A. Bastian


In people with chronic pain, smoking is associated with greater pain intensity and pain-related functional interference. Researchers have modeled a reciprocal relationship in which pain and smoking exacerbate each other, resulting in greater pain and increased smoking. The Pain and Smoking Inventory (PSI) measures patients’ perceived interrelations of their pain and smoking behavior, and comprises three conceptually-distinct domains: smoking to cope with pain (PSI-Cope), pain as a motivator of smoking (PSI-Motivate), and pain as a barrier to cessation (PSI-Barrier). Within a smoking cessation trial targeting cognitive behavioral interventions to promote smoking cessation and pain self-management, this study aimed to (1) evaluate PSI scores effects on pain interference and self-efficacy to quit smoking, and (2) qualitatively examine patient perspectives on the association between pain and smoking.

This was a secondary analysis of baseline data from the Pain and Smoking Study (PASS), a randomized clinical trial of a telephone-delivered, cognitive behavioral intervention among Veterans with chronic pain who smoke cigarettes. For the PSI analysis, a stepwise logistic regression model was used to examine associations between the three PSI domains and pain interference/self-efficacy. For the qualitative analysis, a retrospective thematic analysis of smoking cessation counseling notes was conducted using Atlas.ti. Coders participated in a consensus-forming exercise with salient themes validated among the wider research team.

The 371 participants were 88% male, with a median age of 60 years (range 24-82) and smoked a median of 15 cigarettes per day. Participants were mainly white (61%) and unemployed (70%), and 59% had a Patient Health Questionnaire-9 score ≥10 for depressive symptoms. Overall, mean pain intensity in past week was 5.2 (Standard Deviation (SD) 1.6), and 59% had high pain interference (≥5/10), while 20% had low self-efficacy to quit smoking (≤2/6). PSI-Cope scores were negatively associated with pain interference (OR: 0.72, 95% CI: 0.55, 0.95) and PSI-Motivate scores were positively associated with pain interference (OR: 1.38, 95% CI: 1.08, 1.78). Higher PSI-Barrier sub-scores were associated with low self-efficacy (OR: 1.41, 95% CI: 1.11, 1.79). Among a subset of Veterans (n = 136) with counseling notes, in our qualitative analysis we found that (1) pain motivates smoking and helps manage pain-related distress, as a coping strategy and through cognitive distraction, and (2) pain motivates smoking but smoking does not offer pain relief. Concerns about managing pain without smoking was a notable barrier to cessation.

Our study suggests that individuals who hold maladaptive perceptions of pain-smoking interrelations may be more likely to endorse high pain interference and low self-efficacy, two established predictors of cessation outcomes. Many patients with chronic pain who smoke readily identified pain as a motivator of their smoking behavior and are reluctant to quit for this reason. Future research should evaluate the clinical utility of assessing each PSI domain in relation to cessation outcomes, and integrated interventions for smokers with pain should address cognitive behavioral mood management and promote uptake of more adaptive self-management strategies for pain.


This thesis is restricted to Yale network users only. This thesis is permanently embargoed from public release.