Internet Mental Health


  • Tobacco use disorder is the continued use of tobacco despite clinically significant distress or impairment.

  • It typically includes a strong desire to use tobacco, difficulties in controlling its use, persisting in its use despite harmful consequences, a higher priority given to drug use than to other activities and obligations, increased tolerance, and a physical withdrawal state.


    Lasts for years



  • Almost half of ALL cancer cases in the US are linked to tobacco use

  • Tobacco use is the leading cause of death in the developed world (due to cancers, cardiovascular diseases, chronic obstructive pulmonary disease); "second hand smoke" causes cancer for those close by (family, coworkers)

  • Tobacco smoking is the most common cause of preventable death and disease in the developed world

  • The leading causes of death in USA in 2000 were tobacco (435 000 deaths; 18.1% of total US deaths), poor diet and physical inactivity (365 000 deaths; 15.2%), and alcohol consumption (85 000 deaths; 3.5%)

  • Smoking kills one in every 2 smokers; smokers die, on average, 10 years younger

  • US vice president Mike Pence said he doesn't believe that smoking kills

Back to top

Click Here For Free Diagnosis

Back to top

Dependence Syndrome Due To Use Of Tobacco F17 - ICD10 Description, World Health Organization
Repeated use of tobacco that typically includes a strong desire to take the drug, difficulties in controlling its use, persisting in its use despite harmful consequences, a higher priority given to drug use than to other activities and obligations, increased tolerance, and a physical withdrawal state.
Tobacco Use Disorder - Diagnostic Criteria, American Psychiatric Association

An individual diagnosed with tobacco use disorder needs to meet all of the following criteria:

  • A problematic pattern of tobacco use leading to clinically significant impairment or distress, as manifested by at least two of the following, occurring within a 12-month period:

    • Tobacco is often taken in larger amounts or over a longer period than was intended.

    • There is a persistent desire or unsuccessful efforts to cut down or control tobacco use.

    • A great deal of time is spent in activities necessary to obtain or use tobacco.

    • Craving, or a strong desire or urge to use tobacco.

    • Recurrent tobacco use resulting in a failure to fulfill major role obligations at work, school, or home (e.g., interference with work).

    • Continued tobacco use despite having persistent or recurrent social or interpersonal problems caused or exacerbated by the effects of tobacco (e.g., arguments with others about tobacco use).

    • Important social, occupational, or recreational activities are given up or reduced because of tobacco use.

    • Recurrent tobacco use in situations in which it is physically hazardous (e.g., smoking in bed).

    • Tobacco use is continued despite knowledge of having a persistent or recurrent physical or psychological problem that is likely to have been caused or exacerbated by tobacco.

    • Tolerance, as defined by either of the following:

      • A need for markedly increased amounts of tobacco to achieve the desired effect.

      • A markedly diminished effect with continued use of the same amount of tobacco.

    • Withdrawal, as manifested by either of the following:

      • The characteristic withdrawal syndrome for tobacco:

        • Daily use of tobacco for at least several weeks.

        • Abrupt cessation of tobacco use, or reduction in the amount of tobacco used, followed within 24 hours by four (or more) of the following signs and symptoms:

          • Irritability, frustration, or anger.

          • Anxiety.

          • Difficulty concentrating.

          • Increased appetite.

          • Restlessness.

          • Depressed mood.

          • Insomnia.

      • Tobacco (or closely related substance, such as nicotine) is taken to relieve or avoid withdrawal symptoms.

    • Specify if:

      • In early remission: After full criteria for tobacco use disorder were previously met, none of the criteria for tobacco use disorder have been met for at least 3 months but for less than 12 months (with the exception that the criterion, "Craving, or a strong desire or urge to use tobacco," may be met).

      • In sustained remission: After full criteria for tobacco use disorder were previously met, none of the criteria for tobacco use disorder have been met at any time during a period of 12 months or longer (with the exception that the criterion, "Craving, or a strong desire or urge to use tobacco," may be met).

    • Specify if:

      • On maintenance therapy: The individual is taking a long-term maintenance medication, such as nicotine replacement medication, and no criteria for tobacco use disorder have been met for that class of medication (except tolerance to, or withdrawal from, the nicotine replacement medication).

      • In a controlled environment: This additional specifier is used if the individual is in an environment where access to tobacco is restricted.

Back to top

Diagnostic Features

The leading causes of death in USA in 2000 were tobacco (435 000 deaths; 18.1% of total US deaths), poor diet and physical inactivity (365 000 deaths; 15.2%), and alcohol consumption (85 000 deaths; 3.5%). Thus cigarette smoking is the most common cause of preventable death and disease. Smoking kills one in every 2 smokers. Smokers die, on average, 10 years younger.

Tobacco Use Disorder is a condition characterized by the harmful consequences of repeated tobacco use, a pattern of compulsive tobacco use, and (sometimes) physiological dependence on tobacco (i.e., tolerance and/or withdrawal). This disorder is only diagnosed when tobacco use becomes persistent and causes significant occupational, social or medical impairment. Use of tobaco rarely causes interference with work or doing home obligations, but it can cause social problems (e.g., arguments about smoking), or use that is physically hazardous (e.g., smoking in bed, smoking around flammable chemicals). Although smoking may not cause significant occupational or social impairment; tobacco use is the leading cause of death in the developed world.

Tobacco Intoxication is very rare since tolerance rapidly develops; hence the disappearance of the initial dizziness and nausea.

Tobacco Withdrawal is characterized by four or more of the following: irritability/frustration/anger, anxiety, difficulty concentrating, increased appetite, restlessness, depressed mood, or insomnia.


Medical complications of tobacco use often begins in the 40s, and become more severe over time. One-half of smokers who do not quit die early from a tobacco-related illness. The longer the person smokes, the less reversible the medical harm. Secondhand smoke increases the risk of heart disease and cancer by 30%. Long-term use of nicotine medications does not appear to cause medical harm.


The most common diseases from smoking are cardiovascular illnesses, chronic obstructive pulmonary disease, and cancers. Smoking during pregnancy increases the risk of low birth weight and miscarriage. Children with other Substance Use Disorders, Major Depressive Disorder, Bipolar Disorders, Anxiety Disorders, Personality Disorders, and Attention Deficit - Hyperactivity Disorder are at a higher risk of developing Tobacco Use Disorder.

Associated Laboratory Findings

Carbon monoxide in the breath, and nicotine or its metabolite cotinine in blood, saliva, or urine can be used to quantify the amount of current tobacco use.


In America, 57% of adults have never smoked, 22% are former smokers, and 21% are current smokers. Smokeless tobacco use is less than 5%, and smoking pipes/cigars is less than 1%. The 12-month prevalence is 13% among adults aged 18 years and older. Prevalence declines from 17% among 18- to 29-year-olds to 4% among adults aged 65 years and older. In developed nations the male:female ratio is 1:1, but in developing nations smoking is much greater in males than females.


In America, tobacco smoking starts in adolescence and Tobacco Use Disorder starts in late adolescence. Initiation of smoking after age 21 years is rare. Most smokers make multiple attempts at quitting, and one-half eventually abstain (usually after age 30).

Familial Pattern

About 50% of the risk of developing Tobacco Use Disorder is genetic. Many individuals with Tobacco Use Disorder have genetic behavioral disinhibition (a highly heritable tendency towards impulsivity, novelty-seeking, dangerous risk-taking and rule-breaking).

Effective Therapies

Nicotine replacement therapy increases the chances of stopping smoking by 50 to 70%. Adding behavioural support (in person or via telephone for at least four sessions of contact) to nicotine replacement therapy increases the chances of stopping smoking by a further 10 to 25%. Group programs are more effective for helping people to stop smoking than being given self-help materials without face-to-face instruction and group support. Medications such as nicotine replacement and bupropion are ineffective with adolescents. Bupropion and nortriptyline aid long-term smoking cessation in adults, but other antidepressant medications do not. However, adding bupropion or nortriptyline to nicotine replacement therapy does not provide an additional long-term benefit. One experiment with 878 smokers achieved a high and long-lasting smoking cessation rate. Instead of running a typical smoking cessation program; this very successful experiment put all its resources into a program of large cash rewards (up to US$750) for non-smoking.

Top 20 Most Harmful Drugs In Britain In 2008

Professor David Nutt published in the Lancet the following rating of Britain's most dangerous drugs. They are listed in descending order from the most harmful.

1. Heroin

Class A drug. Originally used as a painkiller and derived from the opium poppy. There were 897 deaths recorded from heroin and morphine use in 2008 in England and Wales, according to the Office of National Statistics (ONS). There were around 13,000 seizures, amounting to 1.6m tonnes of heroin.

2. Cocaine

Class A. Stimulant produced from the South American coca leaf. Accounted for 235 deaths -- a sharp rise on the previous year's fatalities. Nearly 25,000 seizures were made, amounting to 2.9 tonnes of the drug.

3. Barbiturates

Class B. Synthetic sedatives used for anaesthetic purposes. Blamed for 13 deaths.

4. Street methadone

Class A. A synthetic opioid, commonly used as a substitute for treating heroin patients. Accounted for 378 deaths and there were more than 1,000 seizures of the drug.

5. Alcohol

Subject to increasing concern from the medical profession about its damage to health. According to the ONS, there were 8,724 alcohol deaths in the UK in 2007. Other sources claim the true figure is far higher.

6. Ketamine

Class C. A hallucinogenic dance drug for clubbers. There were 23 ketamine-related deaths in the UK between 1993 and 2006. Last year there were 1,266 seizures.

7. Benzodiazepines

Class C. A hypnotic relaxant used to treat anxiety and insomnia. Includes drugs such as diazepam, temazepam and nitrazepam. Caused 230 deaths and 1.8m doses were confiscated in more than 4,000 seizure operations.

8. Amphetamine

Class B. A psychostimulant that combats fatigue and suppresses hunger. Associated with 99 deaths, although this tally includes some ecstasy deaths. Nearly 8,000 seizures, adding up to almost three tonnes of confiscated amphetamines.

9. Tobacco

A stimulant that is highly addictive due to its nicotine content. More than 100,000 people a year die from smoking and tobacco-related diseases, including cancer, respiratory diseases and heart disease.

10. Buprenorphine

An opiate used for pain control, and sometimes as a substitute to wean addicts off heroin. Said to have caused 43 deaths in the UK between 1980 and 2002.

11. Cannabis

Class B. A psychoactive drug recently appearing in stronger forms such as "skunk". [Since this video was made; there is now conclusive proof that cannabis causes a 6.7 fold increase in the risk of developing schizophrenia.] Caused 19 deaths and there were 186,000 seizures, netting 65 tonnes of the drug and 640,000 cannabis plants.

12. Solvents

Fumes inhaled to produce a sense of intoxication. Usually abused by teenagers. Derived from commonly available products such as glue and aerosol sprays. Causes around 50 deaths a year.

13. 4-MTA

Class A. Originally designed for laboratory research. Releases serotonin in the body. Only four deaths reported in the UK between 1997 and 2004.

14. LSD

Class A. Hallucinogenic drug originally synthesised by a German chemist in 1938. Very few deaths recorded.

15. Methylphenidate

Class B drug. Brand name of Ritalin. A psychostimulant sometimes used in the treatment of attention deficit disorders.

16. Anabolic steroids

Class C. Used to develop muscles, notably in competitive sports. Also alleged to induce aggression. Have been blamed for causing deaths among bodybuilders. More than 800 seizures.

17. GHB

Class C drug. A clear liquid dance drug said to induce euphoria, also described as a date rape drug. Can trigger comas and suppress breathing. Caused 20 deaths and 47 seizures were recorded.

18. Ecstasy

Class A. Psychoactive dance drug. Caused 44 deaths, with around 5,000 seizures made.

19. Alykl nitrites

Known as "poppers". Inhaled for their role as a muscle relaxant and supposed sexual stimulant. Reduce blood pressure, which can cause fainting and in some cases death.

20. Khat

A psychoactive plant, the leaves of which are chewed in east Africa and Yemen. Also known as qat. Produces mild psychological dependence. Its derivatives, cathinone and cathine, are Class C drugs in the UK.

Ineffective therapies

Anxiolytic medication, acupuncture, acupressure, laser therapy and electrical stimulation are all ineffective for the treatment of this disorder.

Legalizing Illicit Drugs

Some people argue that illicit drugs should be legalized to decrease the crime associated with these drugs. Historically, tobacco and alcohol were once illegal drugs. Tobacco smoking is now the leading cause of death in America, and alcoholism is the third leading cause of death. Thus legalizing illicit drugs does not make them any less medically and socially harmful. In fact the opposite is true; legalizing illicit drugs increases their use and the harm they cause. The Government of Finland is passing legislation that will gradually ban all tobacco use by 2040.

Back to top

Back to top

  • Tobacco use is the leading cause of death in the developed world. All patients with COPD, regardless of severity, appear to benefit from quitting smoking. Similarly, patients with recent cardiovascular disease events also benefit from quitting smoking. The risk of total mortality and rate of recurrence of lung cancer is substantially lower in smokers who manage to quit smoking following the diagnosis of early stage lung cancer or small cell lung cancer. Together, these data suggest that quitting smoking is effective both as a primary and a secondary intervention in reducing total morbidity and mortality related to COPD, CVD, and lung cancer.
  • Effectiveness of the Electronic Cigarette: An Eight-Week Flemish Study with Six-Month Follow-up on Smoking Reduction, Craving and Experienced Benefits and Complaints (2014) This study followed 48 smokers who were unwilling to quit smoking, but willing to add second-generation nicotine e-cigarettes to their tobacco smoking. At the end of a 8-month trial, 21% of all participants had completely stopped smoking. [Note: This study started with 48 subjects, but ended with only 36 - thus lost 25% of its subjects.]
  • Electronic cigarettes for smoking cessation: a randomised controlled trial. (2013) - At 6 months, verified abstinence was 7.3% with nicotine e-cigarettes, 5.8% with nicotine patches, and 4.1% with placebo e-cigarettes. [Note: This study used first-generation, and now obsolete, e-cigarettes. Subsequent studies using second-generation e-cigarettes have proven to be much more effective in helping smokers quit.]
  • Pharmacotherapeutic management of nicotine dependence in pregnancy. (2011) Smoking in pregnancy can cause serious adverse antenatal and postnatal morbidities, and a significant number of women continue to smoke in pregnancy despite these consequences. Early intervention in the form counseling from their physicians, pregnancy-specific self-help materials, counseling sessions with a health educator, and/or continued follow-up can result in better pregnancy outcomes and possibly long-term cessation. If a woman cannot quit despite these measures, pharmacotherapy can be considered. Currently, nicotine replacement therapy (NRT), transdermal patches, and bupropion are used in pregnancy, but data on the safety and efficacy are largely lacking.
  • Pharmacotherapy for smoking cessation: present and future. (2011) Tobacco dependence is a chronic disease that often requires repeated interventions and multiple attempts to quit. To date, three medications are FDA-approved for smoking cessation: nicotine replacement therapy, sustained-release bupropion, and varenicline. These treatments are effective across a broad range of populations, and are recommended for all smokers, including those with psychiatric or addictive comorbidity. Less is known however concerning the benefit-risk profile of these medications in pregnant women and adolescents. With these limitations in mind, clinicians should encourage and offer counseling and a prescription of pharmacotherapy to every patient willing to make a quit attempt. Despite the relative efficacy of first-line medications, many smokers relapse after one given quit attempt, and alternative pharmacotherapies are needed. Clonidine and nortriptyline have been proposed as second-line medications.
  • Effectiveness of over-the-counter nicotine replacement therapy: a qualitative review of nonrandomized trials. (2011) Some lines of evidence appear to confirm the effectiveness of OTC NRT, but others do not.
  • Adverse effects and tolerability of medications for the treatment of tobacco use and dependence. (2010)Tobacco use is the leading cause of preventable death and disability in the world. Although gradually declining in most developed countries, the prevalence of tobacco use has increased among developing countries. Treatment for tobacco use and dependence is effective, although long-term abstinence rates remain disappointingly low. In response, new treatments continue to be developed. In addition, many of the pharmacotherapies that have been available for years have found new applications with the use of medication combinations, higher doses and a longer duration of therapy for approved medications. There are now seven medications (nicotine patch, nicotine gum, nicotine lozenge, nicotine inhaler, nicotine nasal spray, bupropion sustained release and varenicline) approved for tobacco dependence treatment in most countries, and many national and professional society practice guidelines recommend their use. Although each of the medications used for tobacco dependence treatment has been rigorously tested for efficacy and safety, broader experience in clinical trials and in observational population-based studies suggests that adverse events associated with these medications are relatively common. Since 2008, two of the medications (varenicline and bupropion) have come under increasing scrutiny because of reports of unexplained serious adverse events (SAEs), including behavior change, depression, self-injurious thoughts and suicidal behavior. To date, this association has not been shown to be caused by these medications, but concerns about medication safety continue. Prescribers require a working knowledge of the common adverse effects for all of these medications as well as a more detailed knowledge of the SAE potential. Nicotine replacement therapy (NRT) has been rigorously tested in clinical trials for over 30 years. A number of adverse effects are commonly associated with NRT use, although SAEs are rare. The adverse effects associated with NRT are due to the pharmacological action of nicotine as well as the mode and site of the NRT application. Bupropion has been tested in over 40 controlled clinical trials and has been associated with higher rates of treatment discontinuation due to adverse events than NRTs. A number of SAEs are associated with bupropion and new warnings were recently added to bupropion prescribing information because of observed neuropsychiatric symptoms including suicidal thoughts and behaviours. Varenicline is the most recently approved medication for tobacco dependence treatment and, although proven safe in clinical testing, new safety concerns have arisen based on post-marketing reports. Warnings have been added to the prescribing information for varenicline because of neuropsychiatric symptoms also including suicidal thoughts and behaviours. Informed decision making regarding the use of pharmacotherapy for the treatment of tobacco dependence requires knowledge about the risks of drug treatment that is weighed against the risks of continued tobacco use and the benefits of treatment. Over half of all long-term smokers will die of a tobacco-related disease and the risk of a serious or life-threatening adverse event with tobacco cessation pharmacotherapy is vanishingly small. Pharmacotherapy for tobacco dependence is also among the most cost-effective preventive health interventions. Given these factors, the benefit?:?risk ratio is strongly in favour of pharmacotherapy for tobacco dependence treatment in virtually all smokers who are motivated to quit.
  • Motivational interviewing for smoking cessation: a meta-analytic review. (2010) Motivational interviewing (MI) is a treatment approach that has been widely examined as an intervention for tobacco dependence and is recommended in clinical practice guidelines. (However research for) MI for smoking cessation noted effects that were modest in magnitude (with only) 2.3% difference in abstinence rates between MI and comparison groups. (Editor's note: thus motivational interviewing for smoking cessation is ineffective.)
  • Does dual use jeopardize the potential role of smokeless tobacco in harm reduction? (2010) Dual use of smokeless tobacco and cigarettes may result in reduction in smoking-related harm as smoking intensity is decreased and smoking cessation increases.
  • A tobacco reconceptualization in psychiatry: toward the development of tobacco-free psychiatric facilities. (2010) Tobacco dependence is the leading cause of death in persons with psychiatric and substance use disorders. The creation of smoke-free psychiatric hospitals was mandated by the Joint Commission for Accreditation of Health Organizations (JCAHO) in 1992. This review article addresses the reasons why tobacco should be excluded from psychiatric and addictions treatment settings, and strategies that can be employed to initiate and maintain tobacco-free psychiatric settings.
  • Comparison of available treatments for tobacco addiction. (2010) Effective behavioral and pharmacologic treatments for smoking cessation are available. Behavioral treatments including brief (< 3 min) counseling by physicians are effective. Seven first-line pharmacologic treatments are currently available: five nicotine replacement therapies, bupropion, and varenicline. In addition, clonidine and nortriptyline are second-line treatments for smoking cessation. These treatments increase the chances of quitting smoking by two- to threefold, supporting their use in smokers who are motivated to quit. However, effective treatments for many subpopulations, including smokers with psychiatric comorbidities as well as adolescent, pregnant, or postpartum smokers, remain to be developed and represent an important challenge.
  • Nicotine dependence and psychiatric disorders in the United States: results from the national epidemiologic survey on alcohol and related conditions. (2010) Nicotine-dependent individuals made up only 12.8% (95% confidence interval, 12.0-13.6) of the population yet consumed 57.5% of all cigarettes smoked in the United States. Nicotine-dependent individuals with a comorbid psychiatric disorder made up 7.1% (95% confidence interval, 6.6-7.6) of the population yet consumed 34.2% of all cigarettes smoked in the United States.

Back to top

How to Quit Smoking

Click Here For More Self-Help

Back to top

    "In physical science a first essential step in the direction of learning any subject is to find principles of numerical reckoning and practicable methods for measuring some quality connected with it. I often say that when you can measure what you are speaking about and express it in numbers you know something about it; but when you cannot measure it, when you cannot express it in numbers, your knowledge is of a meagre and unsatisfactory kind: it may be the beginning of knowledge, but you have scarcely, in your thoughts, advanced to the stage of science, whatever the matter may be."

    Lord Kelvin (1824 – 1907)

  • The best summary on bad research is given by Laura Arnold in this TEDx lecture. If you read nothing else about research, you owe it to yourself to watch this short video - it is excellent!

  • Canadian researchers who commit scientific fraud are protected by privacy laws: There are criminals in every community - even in the scientific research community (especially if a lot of money is at stake). Criminal researchers can hide their fraud behind outdated privacy laws.

  • The power of asking "what if?"

  • The active placebo effect: 2300 years ago, the Greek Stoic philosophers taught that it is not the objective event, but our subjective judgment about the event, that determines our behavior. The active placebo effect bears witness to this ancient wisdom.

  • Criteria For High Quality Research Studies

  • It is troubling that a recent study found that two-thirds of important psychological research studies couldn't be replicated. High quality research must meet the following criteria:

    • Randomized Controlled Trial:
      Ask: Was the trial randomized? Was the randomization procedure described and was it appropriate? The best research design is to have research subjects randomly assigned to an experimental or control group. It is essential that confounding factors be controlled for by having a control group or comparator condition (no intervention, placebo, care as usual etc.).

    • Representative Sample:
      Ask: Do the research subjects represent a normal cross-section of the population being studied? Many psychological research studies using university students are flawed because their subjects are not representative of the normal population since they are all W.E.I.R.D. (White, Educated, Intelligent, Rich, and living in a Democracy).

    • Single Blind Trial:
      Ask: Was the treatment allocation concealed? It is essential that the research subjects are kept "blind" as to whether they are in the experimental or control group (in order to control for any placebo effects).

    • Double Blind Trial (Better Than Single Blind Trial):
      Ask: Were blind outcome assessments conducted? In a double blind study, neither the research subjects nor the outcome assessors know if the research subject is in the experimental or control group. This controls for both the placebo effect and assessor bias.

    • Baseline Comparability:
      Ask: Were groups similar at baseline on prognostic indicators? The experimental and control groups must be shown to be comparable at the beginning of the study.

    • Confounding Factors:
      Ask: Were there factors, that weren't controlled for, that could have seriously distorted the study's results? For example, research studies on the effectiveness of mindfulness cognitive therapy in preventing depressive relapse forgot to control for whether the research subjects were also simultaneously receiving antidepressant medication or other psychological treatments for depression.

    • Intervention Integrity:
      Ask: Was the research study protocal strictly followed? The research subjects must be shown to be compliant (e.g., taking their pills, attending therapy) and the therapists must be shown to be reliably delivering the intervention (e.g., staying on the research protocol).

    • Statistical analysis:
      Ask: Was a statistical power calculation described? The study should discuss its statistical power analysis; that is whether the study size is large enough to statistically detect a difference between the experimental and control group (should it occur) and usually this requires at least 50 research subjects in the study.

      Ask: Are the results both statistically significant and clinically significant? The results should be both statistically significant (with a p-value <0.05) and clinically significant using some measure of Effect Size such as Standardized Mean Difference (e.g., Cohen's d >= 0.33). The summary statistics should report what percentage of the total variance of the dependent variable (e.g., outcome) can be explained by the independent variable (e.g., intervention). In clinical studies, the study should report the number needed to treat for an additional beneficial outcome (NNTB), and the number needed to treat for an additional harmful outcome (NNTH).

        Number Needed To Benefit (NNTB): This is defined as the number of patients that need to be treated for one of them to benefit compared with a control in a clinical trial. (It is defined as the inverse of the absolute risk reduction.) Note: Statistically, the NNTB depends on which control group is used for comparison - e.g., active treatment vs. placebo treatment, or active treatment vs. no treatment.

        Number Needed To Harm (NNTH): This is defined as the number of patients that need to be treated for one of them to be harmed compared with a control in a clinical trial. (It is defined as the inverse of the absolute increase in risk of harm.)

        Tomlinson found “an NNTB of 5 or less was probably associated with a meaningful health benefit,” while “an NNTB of 15 or more was quite certain to be associated with at most a small net health benefit.”

      Ask: Does the researcher accept full responsibility for the study's statistical analysis? The researcher should not just hand over the study's raw data to a corporation (that may have $1,000 million invested in the study) to do the statistical analysis.

    • Completeness of follow-up data:
      Ask: Was the number of withdrawals or dropouts in each group mentioned, and were reasons given for these withdrawals or dropouts? Less than 20% of the research subjects should drop out of the study. The intervention effect should persist over an adequate length of time.

    • Handling of missing data:
      Ask: Was the statistical analysis conducted on the intention-to-treat sample? There must be use of intention-to-treat analysis (as opposed to a completers-only analysis). In this way, all of the research subjects that started the study are included in the final statistical analysis. A completers-only analysis would disregard those research subjects that dropped out.

    • Replication of Findings:
      Ask: Can other researchers replicate this study's results? The research study's methodology should be clearly described so that the study can be easily replicated. The researcher's raw data should be available to other researchers to review (in order to detect errors or fraud).

    • Fraud:
      Ask: Is there a suspicion of fraud? In a research study, examine the independent and dependent variables that are always measured as a positive whole number (e.g., a variable measured on a 5-point Likert-type scale ranging from "1 = definitely false to 5 = definitely true" etc.). For each of these variables, look at their sample size (n), mean (M) and standard deviation (SD) before they undergo statistical analysis. There is a high suspicion of fraud in a study's statistics:

      • If the M is mathematically impossible (online calculator): This is one of the easiest ways to mathematically detect fraud. The mean (M) is defined as "the sum (Sum) of the values of each observation divided by the total number (n) of observations". So: M = Sum/n. Thus: (Sum) = (M) multiplied by (n). We know that, if a variable is always measured as a positive whole number, the sum of these observations always has to be a whole number. For these variables to test for fraud: calculate (M) multiplied by (n). This calculates the Sum which MUST be a positive whole number. If the calculated Sum isn't a positive whole number; the reported mean (M) is mathematically impossible - thus the researcher either cooked the data or made a mistake. A recent study of 260 research papers published in highly reputable psychological journals found that 1 in 2 of these research papers reported at least one impossible value, and 1 in 5 of these research papers reported multiple impossible values. When the authors of the 21 worst offending research papers were asked for their raw data (so that its reliability could be checked) - 57% angrily refused. Yet such release of raw data to other researchers is required by most scientific journals. (Here is an example of a research paper filled with mathematically impossible means.)

      • If the SD is mathematically impossible (online calculator): When researchers fraudulently "cook" their data, they may accidently give their data a mean and standard deviation that is mathematically impossible.

      • If the SD/M is very small (i.e., the variable's standard deviation is very small compared to the mean suggesting data smoothing).

      • If the SD's are almost identical (i.e., the variables have different means but almost identical standard deviations).

      • If the 4th digit of the values of the variables aren't uniformly distributed - since each should occur 10% of the time (Benford's Law).

      • If the researcher is legally prevented from publishing negative findings about a drug or therapy because that would violate the "nondisclosure of trade secrets" clause in the research contract (i.e., it is a "trade secret" that the drug or therapy is ineffective - hence this can not be "disclosed"). Approximately half of all registered clinical trials fail to publish their results.

      • If the researcher refuses to release his raw data to fellow researchers (so that they can check its validity). In order to be published in most scientific journals, a researcher must promise to share his raw data with fellow researchers. Thus a researcher's refusal to do so is almost a sure indicator of fraud.

      • If the research study's data contradicts the study's own conclusions - surprisingly, this often occurs.

  • Calling Bullshit In The Age of Big Data - "Bullshit is language, statistical figures, data graphics, and other forms of presentation intended to persuade by impressing and overwhelming a reader or listener, with a blatant disregard for truth and logical coherence." Reading the syllabus of this university course should be required reading for every student of mental health. This syllabus is absolutely fantastic!

  • Statistical Methods in Psychology Journals: Guidelines and Explanations - American Psychologist 1999

  • Not All Scientific Studies Are Created Equal - video

  • The efficacy of psychological, educational, and behavioral treatment

  • Estimating the reproducibility of psychological science

  • Psychologists grapple with validity of research

  • Industry sponsorship and research outcome (Review) - Cochrane Library

  • 'We've been deceived': Many clinical trial results are never published - (text and video)

  • Junk science misleading doctors and researchers

  • Junk science under spotlight after controversial firm buys Canadian journals

  • Medicine with a side of mysticism: Top hospitals promote unproven therapies - Are some doctors becoming modern witchdoctors?

  • When Evidence Says No, But Doctors Say Yes

  • Cochrane Reviews (the best evidence-based, standardized reviews available)

Research Topics

Back to top

Internet Mental Health © 1995-2018 Phillip W. Long, M.D.