The trusted source for
healthcare information and
Abstract & Commentary
Source: Breslau N, et al. Smoking and panic attacks. Arch Gen Psychiatry 1999;56:1141-1147.
Further clarification of the relationship between smoking and panic attacks is important for patient education. For example, daily smoking might cause panic attacks, panic attacks might increase the risk for daily smoking, or, if there is an association between panic attacks and smoking, it might be noncausal: smoking and panic attacks might be linked by a shared etiology. Although observational studies cannot definitively test causal hypotheses, they can dampen the plausibility of some hypotheses and suggest the plausibility of others.
To address this issue, Breslau and colleagues analyzed data from two epidemiological studies: the Epidemiologic Study of Young Adults (ESYA) in southwest Michigan, and the National Comorbidity Survey Tobacco Supplement (NCSTS). Both studies used a structured, lay-administered form of the DSM-III-R for diagnosis. The ESYA (n = 1007) consisted of 21- to 30-year-old members of an HMO, with baseline interviews in 1989 and follow-up interviews in 1990, 1992, and 1994. The study group was 62% female, 80% Caucasian, and 45% married. The NCSTS (n = 4411) interviewed 15- to 54-year-olds at one point. Hazard models with time-dependent covariates were used to estimate the risk for the onset of panic attacks associated with current and prior daily smoking and vice versa. The study controlled for sex and history of major depression, since the latter is associated with both panic attacks and smoking. The influence of heavy drinking was analyzed, given its potential role in panic attacks and its association with smoking, but its inclusion in the model did not alter estimates. The role of lung disease was assessed as a confounding variable that could cause panic attacks.
Baseline characteristics of the population in the ESYA and NCSTS (respectively) were: 42% and 47% smoked daily; 21% and 18% had a history of depression; 12% and 7% had panic attacks; 5% and 3.5% had panic disorder; 13% and 11% smoked daily and had a history of depression; 7% and 3% had panic attacks and a history of depression; and 4% and 2% had panic disorder and a history of depression. In the ESYA database, the lifetime association between panic attacks and daily smoking was similar in both men and women, with odds ratios (ORs) of 3.13 and 2.61, respectively. The hazard ratio (HR) of panic attack associated with prior daily smoking was 3.96. The HR of panic attacks associated with depression alone was 12.98, compared to 12.77 for depression with daily smoking; analyses indicated that if smoking followed the onset of depression, there was an additive effect for the risk of panic attacks. The HR of the first panic attack in daily smokers who continued to smoke was 4.71, compared to 0.21 in those who quit smoking. The HR of smoking after the onset of panic attacks was 1.0. The NCSTS data were more limited in scope. The HR of the first panic attack in daily smokers who continued to smoke was 2.08, compared to 1.85 in those who quit smoking, which is not statistically different. The HR of smoking after the onset of panic attacks was only 1.37. For both data sets, the ORs for panic attacks associated with lung disease, alone and with daily smoking, were higher than for daily smoking alone. The ORs of first panic attack associated with lung disease alone was 9.2, for daily smoking alone was 1.7 (insignificant), and together was 10.7.
The results suggest the possibility that the relationship between smoking and panic attack or disorder might flow primarily in one direction (i.e., from smoking to subsequent onset of panic attacks or disorder), particularly in active daily smokers. Lung disease is associated with an increased lifetime prevalence of panic attacks in nonsmokers and smokers (with or without lung disease). By increasing the risk for lung disease, smoking might indirectly increase the risk for panic attacks.
Comment by Donald M. Hilty, MD
The relationship between depression and smoking is believed to be noncausal, mediated largely or entirely through genetic factors that influence the liability to both smoking and depression.1 The current study best supports a hypothesis that smoking predisposes a patient to panic attacks, but does not rule out a noncausal relationship similar to that of smoking and depression. One interesting hypothesis posits that panic attacks may represent a suffocation false alarm.2 Smokers, who often develop pulmonary problems, may be more prone to react to suffocation signals as manifest panic attacks. Carbon monoxide in cigarette smoke might affect the suffocation alarm threshold and/or the asphyxiation monitor for the suffocation alarm system (which may be the carotid body).3
1. Kendler KS, et al. Smoking and major depression: A causal analysis. Arch Gen Psychiatry 1993;50:36-43.
2. Klein DF. False suffocation alarms, spontaneous panics, and related conditions: An integrative hypothesis. Arch Gen Psychiatry 1993;50:306-317.
3. Preter M, Klein DF. Panic disorder and the suffocation false alarm theory. In: Bellodi L, Perna G, eds., The Panic Respiration Connection. Milan, Italy: MDM Medical Media Srl; 1998:1-24.