Abstract
Through the end of the 20th century, it was widely assumed that physical dependence on tobacco developed only after years of moderate-to-heavy daily smoking. A string of adolescent studies now establishes that physical dependence typically begins soon after the onset of very light nondaily smoking. Case series and surveys reveal that certain symptoms of physical dependence develop in a set sequence that appears to be characteristic of all dependent smokers. This sequence of symptom development provides the basis for an objective clinical approach to staging the progression of dependence. The identification of a set sequence in the development of the symptoms of physical dependence suggests that the as-yet-unidentified neurological cause of physical dependence also develops in a set sequence. This insight provides hope for identifying the cause of physical dependence and its cure.
Physical Dependence on Tobacco
The Historical View
A description of the historical view will provide a context for recent developments. To the end of the 20th century, it was widely assumed that physical dependence required heavy daily smoking over many years.8–11 Based on the presumption that nicotine affected the brain only so long as it was actually present in high concentrations, it was presumed that physically dependent smokers had to maintain nicotine in the brain throughout the waking hours lest nicotine withdrawal rapidly ensue.12,13 It was believed that from the outset, smoking was driven by the pursuit of pleasure, and that tolerance triggered an increase in the frequency of smoking to achieve the same level of reward.8–11 The recent focus of research on the development of nicotine addiction in adolescents reveals that this historical view was wrong in every detail. 14
The Evidence-Based View
In the year 2000, the first longitudinal study of the development of tobacco dependence reported that 22% of adolescents experienced a symptom of dependence within 1 month of initiating occasional smoking. 15 Sixty-three percent of the adolescents who had initiated occasional smoking went on to develop symptoms of dependence. Among those with symptoms, 62% had either experienced their first symptom before initiating daily smoking, or initiated daily smoking upon experiencing their first symptom. 15 The median frequency of smoking at the onset of symptoms was two cigarettes per week. 16 Symptoms of physical dependence were among the most common symptoms reported. The predictive validity of symptom reports was established by the fact that they predicted continued smoking over the 3-year study with an odds ratio of 44. 16
A replication study involving 1,293 adolescents in Quebec demonstrated that symptoms of physical dependence typically made their appearance after the onset of occasional smoking, but before the onset of smoking at least once per week. 17 A study of 353 adolescent smokers in Chicago found that 27% of those reporting symptoms of dependence had experienced these symptoms within1month of initiating smoking. 18 Again, symptoms of physical dependence were reported by some subjects within days of initiating smoking. A fourth study, involving 5,692 adolescents, also reported that symptoms of dependence developed rapidly in some individuals and were present before the onset of daily smoking. 19 Among 594 adolescents who had not smoked more than 100 cigarettes in their lifetime, early symptoms of dependence were found to predict continued smoking 2 years later. 20
Symptoms of dependence have been reported by nondaily adolescent smokers in many cross-sectional studies as well. 21 The National Household Survey on Drug Abuse included 1,734 youth who had started smoking within 24 months of assessment. Nicotine dependence was diagnosed in 36% of youth who had smoked at least 12 cigarettes in their lifetime and in 6% of those who had smoked fewer than 12 cigarettes. 22 Daily smoking was not a prerequisite for dependence, as one third of dependent smokers had never smoked daily. A survey of nearly 25,000 adolescent smokers in New Zealand found symptoms of dependence in 66% of those who were smoking monthly and 82% of those who were smoking weekly, but not yet daily. 23 In the Global Youth Tobacco Surveys in Greece and Cyprus, symptoms of addiction were reported by 41% of youth who were smoking 1 or 2 days per month. 24
In sum, all available evidence indicates that nicotine is like the vast majority of prescription medications: its pharmacological actions start with the first dose. Symptoms of dependence appear so quickly after the onset of intermittent smoking that half of the youth are addicted before they have finished their first pack. 23 The consistent reports of physical dependence among nondaily smokers made it clear that physically dependent smokers could cope without smoking frequently throughout the day. This disproved a long held theory, that physical dependence requires smokers to maintain the constant presence of nicotine in the blood through frequent dosing throughout the day. Subsequent case reports confirmed that beginning smokers can keep withdrawal in check by smoking just 1 or 2 cigarettes per week. 25
The Latency to Withdrawal
The time lapse between finishing the last cigarette and the appearance of withdrawal symptoms is termed the latency to withdrawal.26–30 Smokers' reports of their latency to withdrawal are quite reliable. 30 When smokers first experience physical dependence, the latency to withdrawal may exceed a week. 31 In other words, smoking 1 cigarette will keep all withdrawal symptoms at bay for a week or more.
The Latency to Withdrawal Shortens
With continued tobacco use, the latency to withdrawal becomes shorter and shorter.14,31 A smoker who was initially able to keep withdrawal symptoms away by smoking 1 cigarette per week eventually finds that this is no longer possible. As the latency to withdrawal shortens, staving off withdrawal might require smoking a cigarette once every 5 days, then once every 3 days, then every other day, and so on. When the latency to withdrawal shortens to 1 day, daily smoking ensues. Once daily smoking has been initiated, further shortening of the latency to withdrawal prompts a gradual increase in the number of cigarettes smoked per day. This trajectory of gradually increasing frequency of smoking in terms of days per month, followed, after the onset of daily smoking, by an increase in the number of cigarettes smoked per day, has been documented in the National Youth Tobacco Survey. 21 This trajectory is driven by the form of tolerance that manifests as a shrinking of the latency to withdrawal, and not by tolerance to the pleasure of smoking as previously believed. 14
The Desire to Use Tobacco as a Symptom of Withdrawal
Withdrawal syndromes can follow the abrupt discontinuation of addictive drugs like nicotine, as well as nonaddictive drugs such as beta-blockers or selective serotonin reuptake inhibitors. A key difference in withdrawal from addictive and nonaddictive drugs is that withdrawal from nonaddictive drugs is not accompanied by an urge, craving, or desire to use the drug. Patients who abruptly discontinue anti-depressants may feel terrible, but they never experience craving for their anti-depressant. The most important feature of withdrawal for addictive drugs may therefore be the desire to use the drug. This is the only withdrawal symptom that is common to all addictive drugs.
Wanting, Craving, and Needing
New research indicates that the desire to use nicotine that is triggered by withdrawal progresses through 3 levels of intensity that smokers commonly call wanting, craving, and needing.25,31 Wanting is described as a mild desire to smoke that is short-lived and easily ignored. It does not intrude upon the smoker's thoughts. Craving is a stronger urge to smoke that is more persistent and difficult to ignore. By definition, craving intrudes upon smokers' thoughts, interrupting their concentration. It feels like their brain is telling them that it is time to smoke. 31 Needing is an intense and urgent desire to smoke that is unpleasant and unremitting. Smokers cannot concentrate on anything other than the urgency of needing to smoke, and understand that they will not feel and function normally until the urge is satisfied.
During withdrawal it is typical for a smoker to experience wanting, followed by craving, and eventually needing. Smokers have a typical latency to wanting, a latency to craving, and a latency to needing. Left to their own devices, smokers will typically smoke a cigarette when they begin to experience wanting; therefore, the latency to wanting correlates best with the frequency of smoking (DiFranza, unpublished data).
Stages of Physical Dependence
Symptoms of biological conditions often develop in a characteristic sequence that provides insight into disease mechanisms. For example, primary, secondary, and tertiary syphilis produce characteristic symptoms that appear in the same order and signal localized, systemic, and neurologic infection, respectively. Breaking research indicates that physical dependence to nicotine also produces a characteristic set of symptoms that always develops in the same order.25,31
Smokers who are at stage zero of physical dependence (no dependence) can forgo smoking without experiencing any desire to smoke.
Smokers at stage 1 of physical dependence always experience wanting anytime they go too long without smoking. How long is too long depends on their latency to wanting, but this would typically be measured in units of days or weeks at the onset of dependence. The experience of wanting is mild and transient, and in stage 1 no matter how long the patient remains abstinent, the desire to smoke never gets any stronger than wanting.
When smokers at stage 2 of physical dependence forgo smoking for too long, they first experience wanting and then craving. The time it takes to progress to craving will depend on the smoker's latency to craving, but this will typically be measured in terms of hours or a few days for a smoker in stage 2.
When smokers reach stage 3 of physical dependence, abstinence from smoking will trigger wanting, followed by craving, which in turn will be followed by needing. Because the latencies shorten as addiction progresses, a smoker in stage 3 might experience wanting within minutes of extinguishing a cigarette. Craving could follow a few minutes after that, and needing could be experienced in an hour or 2.
Case studies suggest that the symptoms of wanting, craving, and needing always develop in this sequence.25,31 This statement is also supported by a cross-sectional study of 349 adolescent smokers of whom 99.4% reported symptom combinations that were consistent with this sequence. 32 These studies provide the first evidence that all smokers progress through a common sequence of symptom development.
Significance
Despite the fact that some neurophysiologic correlates of withdrawal have been identified, the biological mechanism underlying tobacco withdrawal is unknown.3,4 While several medications that target withdrawal symptoms are available, there is a need for more safe and effective medications.33–35 As smoking causes more than 400,000 premature deaths in the United States each year, 36 it is important to develop a better understanding of the neurobiology of physical dependence so that more effective drugs can be developed. The discovery that physical dependence develops in a characteristic sequence implies that the neural mechanisms that underlie physical dependence also develop in a characteristic sequence. Hopefully, this insight brings us a step closer to developing an effective cure for addiction.
The historical view of the onset of tobacco dependence was wrong because it was based solely on observations of adult smokers who had reached the end-stage of the progression of dependence. Based on these observations and some erroneous assumptions, a picture was painted of the onset of dependence during adolescence that had no basis in reality, but was accepted by the scientific community for 3 decades. 10 The critical contribution that adolescent research has made to the discovery of how tobacco dependence develops underlines the importance of this population in research.
Footnotes
Acknowledgment
This work was supported by internal funding by the Department of Family Medicine and Community Health at the University of Massachusetts Medical School.
Author Disclosure Statement
No competing financial interests exist.
