Subscription is FREE for qualified healthcare professionals in the US.

Smoking and Aberrant Behavior in Chronic Pain Patients

A narrative review of smoking and aberrant drug-taking behavior in chronic pain patients together with emerging clinical implications.
Page 1 of 2

Smoking rates in the United States among adults with persistent pain are high (30% to 64%) compared to the general population (21%). Furthermore, current evidence demonstrates that pain patients who smoke are at higher risk for aberrant drug-taking behavior (ADTB) on opioid therapy compared to non-smokers. Aberrant drug-taking behaviors involve a spectrum of nonadherent behaviors with opioid therapy that range from “rare, highly aberrant” to “common, not aberrant” related to addiction-related outcomes. The goal of this narrative review was to qualitatively examine studies on smoking and ADTB in patients with persistent pain and discuss emerging clinical implications.

Understanding the role of specific risk factors that predict aberrant drug-taking behaviors with opioid therapy is critical to improving pain management outcomes. Several studies suggest a relationship between cigarette smoking and aberrant drug-taking behaviors. This narrative review evaluates the evidence for a connection between smoking and aberrant drug-taking behavior with opioid therapy for persistent pain. Further, we propose a theoretical framework for understanding empirical linkages between smoking and aberrant drug-taking behaviors and critically evaluate the clinical significance of smoking when treating patients with persistent pain.


A literature search was conducted using PubMed, PsychInfo, and Medline with the four concepts: smoking, pain, aberrant behaviors, and opioids. A search for each concept was conducted separately and then combined.


Six studies showed that current smoking is associated with higher levels of analgesic use (Antonov et al., 1996;1 Antonov et al.,2 1998; Corrigall et al., 1992;3 Friedman et al., 2003;4 Jamison et al., 19915) and more ADTB than nonsmoking (Michna et al., 20046). Smokers with more severe nicotine dependency are more likely to misuse opioid medication than smokers with lower nicotine dependency.6 Michna and colleagues (2004)6 showed that current smoking and an index of tobacco dependence (e.g., smoking within 1 hour of waking) were among the most useful predictors of problems with opioid use among persistent pain patients. In this study, 61% of patients at low-risk for aberrant drug behaviors and 82% of patients at high-risk were current smokers.

Additional evidence for the smoking and ADTB connection is demonstrated by the development of two screening instruments that assess a patient’s risk for nonadherence with opioid use. These instruments assess current smoking status in the determination of risk potential for opioid misuse and include the Screening Instrument for Substance Abuse (SISAP)7 and the Screener and Opioid Assessment for Patients in Pain (SOAP).8 When developing the SISAP, Coambs and colleague7 retrospectively analyzed a large epidemiologic dataset and showed that current smoking and previous smoking status were significant risk factors for substance abuse. Specifically, the item, “Have you ever smoked cigarettes?” accurately identified 76% of patients who had substance abuse problems.7 While this important study established a fundamental link between smoking and risk of ADTB, the study aims and design do not allow for elaboration on the potential mechanisms of the association or a clinical explanation. Table 1 highlights selected study findings from the review results.

Table 1: Summary of Selected Studies on the Relationship between Smoking and Specific Analgesic Use Behaviors
Study Study Design Sample Outcome
Michna et al., 20046 Cross-sectional interview-based survey 145 chronic low back pain patients 82.2% of patients at high risk for ADTB and 61.0% of patients at low-risk for ADTB were current smokers.
Friedman et al., 20034 Cross-sectional survey 48 chronic pain patients on opioid therapy Current smoking status differentiated pain patients who had substance abuse disorders from patients without a substance abuse history.
Coambs et al., 19967 Retrospective population-based survey 9,915 adults I Canada Current and former smoking status were strong predictors of substance abuse.
Antonov et al., 19961 Population-based survey 13,295 adults in Sweden Current smoking status (26.0%) was positively associated with current analgesic use in women, but not men.
Antonov et al., 19982 Population-based survey 11,996 adults in Sweden Current smoking status was significantly associated with prescription analgesic use in men and with nonprescription analgesic use in women.

Theoretical Framework for Smoking-ADTB and Potential Clinical Implications

Much research identifying the association between smoking and the risk for ADTB during opioid therapy has been preliminary in nature. We propose three different explanations or profiles for understanding linkages between smoking and ADTB. As described in the section below, these profiles or “groups” involve the hypothesized roles of pseudoaddiction; substance abuse disorders, and chemical coping/negative affect reduction.

Group 1: Pseudo-addicted Group

While nicotine is clearly analgesic in studies of acute pain (often in opioid naïve subjects, e.g., Flood & Daniel, 2004;9 Jamner et al., 1998;10 Perkins et al., 199411), it is unclear whether smoking produces analgesic effects in patients with persistent pain. In fact, behaviors displayed by smokers that are perceived as aberrant may be motivated by the undertreatment of pain or inadequate analgesia. Weissman & Haddox (1989)12 proposed the iatrogenic phenomenon of pseudo-addiction wherein patients who have inadequate pain relief can be confused for addicts in their desperate attempts to seek relief. While Jamison et al. (1991)5 reported that many patients smoke to relieve severe pain intensity, other studies showed that nicotine-opioid interactions may lead to diminished pain relief (Zevin & Benowitz, 1999).13 These findings include, but are not limited to, the decreased efficacy of propoxyphene in smokers compared to non-smokers (Anon, 1973);14 the increased metabolic clearance of pentazocine in smokers (Vaughan et al., 1976);15 the need for higher doses of pentazocine in smokers compared to non-smokers (Keeri-Szanto & Pomeroy, 1971),16 and evidence for smoking inducing the metabolism of codeine (Yue et al., 1994).17 Thus, the smoking–ADTB connection could be the result of smokers receiving less analgesia from “routine” doses of opioids than do non-smokers, and subsequent pseudo-addictive behaviors that are in pursuit of enhanced pain control.

Last updated on: February 26, 2013
First published on: September 1, 2007