A number of groups have been identified as being at greater risk of developing cannabis dependence and include adolescent populations, Aboriginal and Torres Strait Islanders (in Australia) and people suffering from mental health conditions.

As demonstrated by the chart below, numerous studies have examined factors which mediate substance abuse or dependence. In these examples, the predictor variables lead to the mediator which in turn leads to the outcome, which is always substance abuse or dependence. For example, research has found that being raised in a single-parent home can lead to increased exposure to stress and that increased exposure to stress, not being raised in a single-parent home, leads to substance abuse or dependence. The following are some, but by no means all, of the possible mediators of substance abuse.

As demonstrated by the chart below, numerous studies have examined factors which moderate substance abuse or dependence. In these examples, the moderator variable impacts the level to which the strength of the relationship varies between a given predictor variable and the outcome of substance abuse or dependence. For example, there is a significant relationship between psychobehavioral risk factors, such as tolerance of deviance, rebelliousness, achievement, perceived drug risk, familism, family church attendance and other factors, and substance abuse and dependence. That relationship is moderated by familism which means that the strength of the relationship is increased or decreased based on the level of familism present in a given individual.

Examples of mediators and moderators can be found in several empirical studies. For example, Pilgrim et al.’s hypothesized mediation model posited that school success and time spent with friends mediated the relationship between parental involvement and risk-taking behavior with substance use (2006). More specifically, the relationship between parental involvement and risk-taking behavior is explained via the interaction with third variables, school success and time spent with friends. In this example, increased parental involvement led to increased school success and decreased time with friends, both of which were associated with decreased drug use. Another example of mediation involved risk-taking behaviors. As risk-taking behaviors increased, school success decreased and time with friends increased, both of which were associated with increased drug use.

A second example of a mediating variable is depression. In a study by Lo and Cheng (2007), depression was found to mediate the relationship between childhood maltreatment and subsequent substance abuse in adulthood. In other words, childhood physical abuse is associated with increased depression, which in turn, in associated with increased drug and alcohol use in young adulthood. More specifically, depression helps to explain how childhood abuse is related to subsequent substance abuse in young adulthood.

A third example of a mediating variable is an increase of externalizing symptoms. King and Chassin (2008) conducted research examining the relationship between stressful life events and drug dependence in young adulthood. Their findings identified problematic externalizing behavior on subsequent substance dependency. In other words, stressful life events are associated with externalizing symptoms, such as aggression or hostility, which can lead to peer alienation or acceptance by socially deviant peers, which could lead to increased drug use. The relationship between stressful life events and subsequent drug dependence however exists via the presence of the mediation effects of externalizing behaviors.

An example of a moderating variable is level of cognitive distortion. An individual with high levels of cognitive distortion might react adversely to potentially innocuous events, and may have increased difficulty reacting to them in an adaptive manner (Shoal & Giancola, 2005). In their study, Shoal and Giancola investigated the moderating effects of cognitive distortion on adolescent substance use. Individuals with low levels of cognitive distortion may be more apt to choose more adaptive methods of coping with social problems, thereby potentially reducing the risk of drug use. Individuals with high levels of cognitive distortions, because of their increased misperceptions and misattributions, are at increased risk for social difficulties. Individuals may be more likely to react aggressively or inappropriately, potentially alienating themselves from their peers, thereby putting them at greater risk for delinquent behaviors, including substance use and abuse. In this study, social problems are a significant risk factor for drug use when moderated by high levels of cognitive distortions.

Young people are at greater risk of developing cannabis dependency because of the association between early initiation into substance use and subsequent problems such as dependence, and the risks associated with using cannabis at a developmentally vulnerable age. In addition there is evidence that cannabis use during adolescence, at a time when the brain is still developing, may have deleterious effects on neural development and later cognitive functioning.

Withdrawal is the body's reaction to abstaining from a substance upon which a person has developed a dependence syndrome. When dependence has developed, cessation of substance use produces an unpleasant state, which promotes continued drug use through negative reinforcement; i.e., the drug is used to escape or avoid re-entering the associated withdrawal state. The withdrawal state may include physical-somatic symptoms (physical dependence), emotional-motivational symptoms (psychological dependence), or both. Chemical and hormonal imbalances may arise if the substance is not introduced. Psychological stress may also result if the substance is not re-introduced.

Infants also suffer from substance withdrawal, known as Neonnatal Abstinence Syndrome (NAS) which has severe and life-threatening effects on growing fetus. Addiction to drugs and alcohol in expecting mothers does not only cause NAS but also an array of other issues which can continually affect the infant throughout his/her lifetime. The type of drug which was abused during the months of pregnancy has many different effects on the child which can affect the infant in many ways throughout his/her life.

The dependence potential of a drug varies from substance to substance, and from individual to individual. Dose, frequency, pharmacokinetics of a particular substance, route of administration, and time are critical factors for developing a drug dependence.

An article in "The Lancet" compared the harm and dependence liability of 20 drugs, using a scale from zero to three for physical dependence, psychological dependence, and pleasure to create a mean score for dependence. Selected results can be seen in the chart below.

There are approximately 976 million smokers in the world. Estimates are that half of smokers (and one-third of former smokers) are dependent based on DSM criteria, regardless of age, gender or country of origin, but this could be higher if different definitions of dependence were used. Recent data suggest that, in the United States, the rates of daily smoking and the number of cigarettes smoked per day are declining, suggesting a reduction in population-wide dependence among current smokers. However, there are different groups of people who are more likely to smoke than the average population, such as those with low education or low socio-economic status and those with mental illness. There is also evidence that among smokers, some subgroups may be more dependent than other groups. Men smoke at higher rates than do women and score higher on dependence indices; however, women may be less likely to be successful in quitting, suggesting that women may be more dependent by that criterion. Higher nicotine dependence has also been linked with mental illness, including anxiety and depression.

Traditional cigarettes are the most common delivery device for nicotine. However, electronic cigarettes are becoming more popular. Nicotine can also be delivered via other tobacco products such as chewing tobacco, snus, pipe tobacco, hookah, all of which can produce nicotine dependence. The risk for developing dependence is related to how much nicotine is in a product and how fast it can get to the user’s brain. Smoke from products like cigarettes and hookah is absorbed in the lungs and then nicotine gets to the brain in about 6 seconds. Cigarettes and hookah also have large amounts of nicotine in them, and therefore people who use them have a higher risk of developing dependence. Products where nicotine is absorbed into the blood through the mouth (e.g., snus, chewing tobacco), and therefore takes longer to get to the brain, have lower dependence risk. Nicotine replacement medications like the nicotine patch, nicotine lozenges, and nicotine gum have minimal to no dependence risk because they do not contain nearly as much nicotine as tobacco products and they don’t get into the brain very quickly. This is why nicotine replacement medications are safe.

Some medical systems, including those of at least 15 states of the United States, refer to an Addiction Severity Index to assess the severity of problems related to substance use. According to DARA Thailand, the index assesses potential problems in seven categories: medical, employment/support, alcohol, other drug use, legal, family/social, and psychiatric.

A study consisting of 1,081 U.S. residents who had first used cocaine within the previous 24 months was conducted. It was found that the risk of becoming dependent on cocaine within two years of first use was 5–6%. The risk of becoming dependent within 10 years of first use increased to 15–16%. These were the aggregate rates for all types of use considered, such as smoking, snorting, and injecting. Among recent-onset users individual rates of dependency were higher for smoking (3.4 times) and much higher for injecting. Women were 3.3 times more likely to become dependent, compared with men. Users who started at ages 12 or 13 were four times as likely to become dependent compared to those who started between ages 18 and 20.

However, a study of non-deviant users in Amsterdam found a "relative absence of destructive and compulsive use patterns over a ten year period" and concluded that cocaine users can and do exercise control. "Our respondents applied two basic types of controls to themselves: 1) restricting use to certain situations and to emotional states in which cocaine's effects would be most positive, and 2) limiting mode of ingestion to snorting of modest amounts of cocaine, staying below 2.5 grams a week for some, and below 0.5 grams a week for most. Nevertheless, those whose use level exceeded 2.5 grams a week all returned to lower levels".

In Europe, the rate of problem gambling is typically 0.5 to 3 percent. The "British Gambling Prevalence Survey 2007", conducted by the United Kingdom Gambling Commission, found approximately 0.6 percent of the adult population had problem gambling issues—the same percentage as in 1999. The highest prevalence of problem gambling was found among those who participated in spread betting (14.7%), fixed odds betting terminals (11.2%) and betting exchanges (9.8%). In Norway, a December 2007 study showed the amount of present problem gamblers was 0.7 percent.

In the United States, cocaine use results in about 5,000–6,000 deaths annually.

In the United States, the percentage of pathological gamblers was 0.6 percent, and the percentage of problem gamblers was 2.3 percent in 2008. Studies commissioned by the National Gambling Impact Study Commission Act has shown the prevalence rate ranges from 0.1 percent to 0.6 percent. Nevada has the highest percentage of pathological gambling; a 2002 report estimated 2.2 to 3.6 percent of Nevada residents over the age of 18 could be called problem gamblers. Also, 2.7 to 4.3 percent could be called probable pathological gamblers.

According to a 1997 meta-analysis by Harvard Medical School's division on addictions, 1.1 percent of the adult population of the United States and Canada could be called pathological gamblers. A 1996 study estimated 1.2 to 1.9 percent of adults in Canada were pathological. In Ontario, a 2006 report showed 2.6 percent of residents experienced "moderate gambling problems" and 0.8 percent had "severe gambling problems". In Quebec, an estimated 0.8 percent of the adult population were pathological gamblers in 2002. Although most who gamble do so without harm, approximately 6 million American adults are addicted to gambling.

Signs of a gambling problem include:

- Using income or savings to gamble while letting bills go unpaid

- Repeated, unsuccessful attempts to stop gambling

- Chasing losses

- Losing sleep over thoughts of gambling

- Arguing with friends or family about gambling behavior

- Feeling depressed or suicidal because of gambling losses

Research studies have come to different conclusions on the number of therapeutic dose users who develop a physical dependence and withdrawal syndrome. Estimates by researchers of the number of people affected range 20–100% of patients prescribed benzodiazepines at therapeutic dosages long term are physically dependent and will experience withdrawal symptoms.

Benzodiazepines can be addictive and induce dependence even at low doses, with 23% becoming addicted within 3 months of use. Benzodiazepine addiction is considered a public health problem. Approximately 68.5% of prescriptions of benzodiazepines originate from local health centers, with psychiatry and general hospitals accounting for 10% each. A survey of general practitioners reported that the reason for initiating benzodiazepines was due to an empathy for the patients suffering and a lack of other therapeutic options rather than patients demanding them. However, long-term use was more commonly at the insistence of the patient, it is presumed, because physical dependence or addiction had developed.

Approximately twice as many women as men are prescribed benzodiazepines. It is believed that this is largely because men typically turned to alcohol to cope with stress and women to prescription drugs. Biased perception of women by male doctors may also play a role in increased prescribing rates to women; however, increased anxiety features in women does not account for the wide gap alone between men and women.

A study published in the British Journal of General Practice in July 2017 found that in a sample taken from a survey conducted in 2014–2015 in Bradford a mean of 0.69% of registered patients had been prescribed benzodiazepines for more than a year. This would suggest that there were around 300,000 long-term users of diazepine in the UK.

According to the CDC in 2017, in the US, "the age-adjusted drug poisoning death rate involving opioid analgesics increased from 1.4 to 5.4 deaths per 100,000 population between 1999 and 2010, decreased to 5.1 in 2012 and 2013, then increased to 5.9 in 2014, and to 7.0 in 2015. The age-adjusted drug poisoning death rate involving heroin doubled from 0.7 to 1.4 deaths per 100,000 resident population between 1999 and 2011 and then continued to increase to 4.1 in 2015."

In 2012 it was estimated that 9.2 percent of the population over the age of 12 years old had used an illicit drug in the previous month. In 2015, it was estimated the 20.5 million Americans had a substance use disorder. Of these 20.5 million, two million used prescribed pain medications (most of those being opioids) and one-half of a million were using heroin.

In 2015, in the US there were 33,000 deaths due to drug overdose that involved opioid use. Of these, about 15,000 were from prescribed opioids and 13,000 were from heroin use.

Non-medical consumption of opioids peaked around 2010 and then started to decrease between 2011 and 2013.

Among adults, the rate of inpatient hospital stays in the United States related to opioid overuse increased by an average of 5% annually from 1993–2012. The percentage of inpatient stays due to opioid overuse admitted from the emergency department increased from 43% in 1993 to 64% in 2005, but have remained relatively constant since 2005.

The prevalence of opioid use and opioid or opiate dependency varies by age and gender, among a myriad of other factors. Men are at higher risk for opioid use and dependency than women, and men also account for more opioid overdoses than women, although this gap is closing. Women are more likely to be prescribed pain relievers, be given higher doses, use them for longer durations, and may become dependent upon them faster.

Deaths due to opioid use also tend to skew at older ages than deaths from use of other illicit drugs. This does not reflect opioid use as a whole, which includes individuals in younger age demographics. Overdoses from opioids are highest among individuals who are between the ages of 40 and 50, in contrast to heroin overdoses, which are highest among individuals who are between the ages of 20 and 30. 21 to 35-year olds represent 77% of individuals who enter treatment for opioid use disorder, however, the average age of first-time use of prescription painkillers was 21.2 years of age in 2013.

Most research has focused on adult population or on college students, but little is known about epidemiology of behavioral addictions in adolescence. A study conducted by Villella "et al" looked at a group of students and the prevalence of various addictions. His results showed exercise addiction was the second most prevalent, after compulsive buying.

High risk groups that appear to be addicted to exercise include athletes in sports encouraging thinness or appearance standards, young and middle-age women, and young men.

Street children in many developing countries are a high risk group for substance misuse, in particular solvent abuse. Drawing on research in Kenya, Cottrell-Boyce argues that "drug use amongst street children is primarily functional – dulling the senses against the hardships of life on the street – but can also provide a link to the support structure of the ‘street family’ peer group as a potent symbol of shared experience."

In order to maintain high-quality performance, some musicians take chemical substances. Some musicians take drugs or alcohol to deal with the stress of performing. As a group they have a higher rate of substance abuse. The most common chemical substance which is abused by pop musicians is cocaine, because of its neurological effects. Stimulants like cocaine increase alertness and cause feelings of euphoria, and can therefore make the performer feel as though they in some ways ‘own the stage’. One way in which substance abuse is harmful for a performer (musicians especially) is if the substance being abused is aspirated. The lungs are an important organ used by singers, and addiction to cigarettes may seriously harm the quality of their performance. Smoking causes harm to alveoli, which are responsible for absorbing oxygen.

Opioid use disorders resulted in 122,000 deaths worldwide in 2015, up from 18,000 deaths in 1990. Deaths from all causes rose from 47.5 million in 1990 to 55.8 million in 2013.

As with many human diseases and disorders, animal models are sometimes used to study addiction. For example, voluntary wheel running by rodents, viewed as a model of human voluntary exercise, has been used to study withdrawal symptoms, such as changes in blood pressure, when wheel access is removed from mice.

A conceptual model of PIU has been developed based on primary data collected from addiction researchers, psychologists, and health providers as well as older adolescents themselves. That study identified seven concepts, or clusters, that make up PIU using a concept mapping approach. These seven clusters are: psychosocial risk factors; physical impairment; emotional impairment; social and functional impairment; risky Internet use; impulsive Internet use; and Internet use dependence. The last three constructs have not been previously identified. Risky Internet use are behaviors that increase risks of adverse consequences. It is not just the amount of time spent on the Internet that puts an adolescent at risk; how the time is spent is also an important consideration. The impulsive use construct describes an inability to maintain balance or control of internet use in relation to everyday life. Finally, the dependent use construct reflects the more severe symptoms that are typically associated with addictions, such as withdrawal symptoms. Thus, internet addiction may represent a severe form of PIU.

Other research also stresses the fact that the Internet addiction disorder is not a unidimensional but a multidimensional construct. Various facets of Internet use must be differentiated because of their differential predictors, mechanisms and consequences. Online activities which, if done in person, would normally be considered troublesome, such as compulsive gambling, or shopping, are sometimes called "net compulsions".

Sending a letter to patients warning of the adverse effects of long-term use of benzodiazepines and recommending dosage reduction has been found to be successful and a cost-effective strategy in reducing benzodiazepine consumption in general practice. Within a year of the letter's going out, there was found to be a 17% fall in the number of benzodiazepines being prescribed, with 5% of patients having totally discontinued benzodiazepines. A study in the Netherlands reported a higher success rate by sending a letter to patients who are benzodiazepine-dependent. The results of the Dutch study reported 11.3% of patients discontinuing benzodiazepines completely within a year.

Addiction is defined by Webster Dictionary as a "compulsive need for and use of a habit-forming substance characterized by tolerance and by well-defined physiological symptoms upon withdrawal; broadly: persistent compulsive use of a substance known by the user to be harmful".

Problematic Internet use is also called compulsive Internet use (CIU), Internet overuse, problematic computer use, or pathological computer use (PCU), problematic Internet use (PIU), or Internet addiction disorder (IAD)). Another commonly associated pathology is video game addiction, or Internet gaming disorder (IGD).

Another growing area is social media addiction. Psychology researchers surveyed 253 undergraduate students at the University of Albany and found that not only is social media (particularly Facebook) itself potentially addictive, those who use it may also be at greater risk for substance abuse.

Psychological dependence is a form of dependence that involves emotional–motivational withdrawal symptoms (e.g., a state of unease or dissatisfaction, a reduced capacity to experience pleasure, or anxiety) upon cessation of drug use or engagement in certain behaviors. Physical and psychological dependence are sometimes classified as a facet or component of addiction, such as in the DSM-IV-TR; however, some drugs which produce dependence syndromes do not produce addiction, and vice versa, in humans. Addiction and psychological dependence are both mediated through reinforcement, a form of operant conditioning, but are associated with different forms of reinforcement. Addiction is a compulsion for rewarding stimuli that is mediated through positive reinforcement. Psychological dependence, which is mediated through negative reinforcement, involves a desire to use a drug or perform a behavior to avoid the unpleasant withdrawal syndrome that results from cessation of exposure to it.

Psychological dependence develops through consistent and frequent exposure to a stimulus. Behaviors which can produce observable psychological withdrawal symptoms (i.e., cause psychological dependence) include physical exercise, shopping, sex and self-stimulation using pornography, and eating food with high sugar or fat content, among others. Behavioral therapy is typically employed to help individuals overcome psychological dependence upon drugs or maladaptive behaviors that produce psychological dependence.

The rate of people who have problems of shopping addiction is a very controversial matter, because the dividing line between pathological behaviours and those behaviours which, even if excessive, are socially accepted, is very difficult to determine. However, shopping addiction and other manifestations of the lack of self-control on spending are widespread problems which are constantly expanding. Studies using samples of the general population show that between 8% and 16% of the people have problems with excessive or uncontrolled purchases. Clinical studies give much lower figures, however, between 2% and 5%. According to the European Report on the programme for the prevention and treatment of personal problems related to consumer addiction, personal purchasing habits and over-indebtedness, 3% of European adults and 8% of European young people have a level of shopping addiction which could be considered as pathologic, that is, which seriously affects the life of the people who suffer from this. Other estimates for the prevalence of compulsive buying range from a low of 2 percent to 12 percent or more (in the U.S. population.

Most of people who have these problems neither receive nor ask for treatment. Those that ask for help only do it after years of suffering, when the addiction has caused very serious economic repercussion and has harmed the relationship with their family and social environment. For this reason and due to the lack of social consciousness about this problem, the unrecorded figure of people who suffer from these problems is very high. In addition to the severe cases of shopping addiction, an important part of consumers (between 30% and 50% of the population) have deficiencies with spending self-control or excessive purchases. According to the European Report, 33% of European adults and 46% of the European young people have minor or moderate problems with shopping addiction or lack of economic self-control.