Table 1.
All Mendelian randomization (MR) studies included for qualitative synthesis, with their identifying information, description of the exposure and outcome variable(s), whether the study used individual-level and/or summary-level data, the total quality rating, and a brief summary of their findings
| ID | First author (Year) | Type of substance | Individual- /summary-level data | Exposure variable(s) | Outcome variable(s) | Quality | Finding(s) |
|---|---|---|---|---|---|---|---|
| 1 | Zhou et al. (2019a) | Smoking | Summary level | Educational attainment (self-report) | Smoking initiation, cigarettes per day, smoking cessation, age at smoking initiation (all self-report) | –a | Evidence for causal, decreasing effects of educational attainment on smoking initiation and cigarettes per day. Evidence for causal, increasing effects of educational attainment on smoking cessation and age at smoking initiation |
| 2 | Zeng et al. (2019) | Smoking | Summary level | Educational attainment (self-report) | Smoking initiation (self-report) | – + | Evidence for causal, decreasing effects of educational attainment on smoking initiation. This effect was only modestly attenuated when adjusting for years spent in school |
| 3 | Gage et al. (2018) | Smoking | Summary level | Educational attainment (self-report) | Smoking initiation (self-report), cigarettes per day (self-report), smoking cessation (self-report), cotinine levels (measured in the blood) | – + | Evidence for causal, decreasing effects of education on smoking initiation and cigarettes per day. Evidence for causal, increasing effects of education on smoking cessation. Weak evidence for causal, decreasing effects of education on cotinine levels |
| 4 | Tillmann et al. (2017) | Smoking | Summary level | Educational attainment (self-report) | Smoking initiation (self-report) | – + | Evidence for causal, decreasing effects of education on smoking initiation |
| 5 | Carter et al. (2019) | Smoking | Both | Educational attainment (self-report) | Lifetime smoking (self-report) | – + | Evidence for causal, decreasing effects of education on lifetime smoking |
| 6 | Sanderson et al. (2019) | Smoking | Both | Educational attainment (self-report), cognitive functioning (multivariable MR) | Smoking initiation, smoking cessation (all self-report) | + | Evidence for causal, decreasing effects of education on smoking initiation, and evidence for causal, increasing effects of education on smoking cessation. These effects of education are independent of cognitive functioning |
| 7 | Gage et al. (2020) | Smoking | Summary level | Smoking initiation (self-report), lifetime smoking (self-report) | Cognitive functioning (fluid intelligence), educational attainment (self-report) | + | Evidence for causal, decreasing effects of smoking initiation and lifetime smoking on educational attainment and cognitive functioning. Results for educational attainment were more robust than results for cognitive functioning |
| 8b | Fu et al. (2019) | Smoking | Individual level | Smoking initiation (self-report) | Cognitive functioning (a composite measure of cognitive tests) | – | Evidence for causal, decreasing effects of current smoking on cognitive functioning |
| 9 | North et al. (2015) | Smoking | Individual level | Cigarettes per day (self-report) | Cognitive functioning (a general fluid cognitive ability score, derived from a range of different cognitive functioning tests), cognitive decline (% change in continuous cognitive measures from baseline to last available wave) | – + | Overall no consistent evidence for causal effects. Very weak evidence for causal, increasing effects of cigarettes per day on cognitive decline (higher odds of being in the top 25% of cognitive decliners). In never smokers, weak evidence for causal, decreasing effects of smoking on search speed |
| 10b | Adams (2019) | Smoking | Summary level | Lifetime smoking (self-report), cognitive functioning (fluid intelligence), neuroticism (self-report) | Lifetime smoking (self-report), cognitive functioning (fluid intelligence), neuroticism (self-report) | – + | Evidence for causal, increasing effects of lifetime smoking on neuroticism. Evidence for causal, decreasing effects of cognitive functioning on lifetime smoking |
| 11 | Østergaard et al. (2015) | Smoking | Summary level | Smoking initiation, cigarettes per day (all self-report) | Alzheimer's disease (diagnosis) | – + | Evidence for causal, decreasing effects cigarettes per day on Alzheimer's disease. No clear evidence for causal effects of smoking initiation on Alzheimer's disease |
| 12 | Gibson et al. (2019) | Smoking | Both | Smoking initiation, cigarettes per day, smoking cessation, sleep duration, chronotype, insomnia (all self-report) | Smoking initiation, cigarettes per day, smoking cessation; sleep duration, chronotype, insomnia (all self-report) | – + | Evidence for causal, decreasing effects of cigarettes per day on the odds of being a morning person. Weak evidence for causal, increasing effects of insomnia on cigarettes per day, and weak evidence for causal, decreasing effects of insomnia on smoking cessation. No clear evidence for causal effects in any of the other tested relationships |
| 13 | Millard et al. (2019) | Smoking | Individual level | Cigarettes per day (self-report) | Chronotype (self-report) | – + | Evidence for causal, decreasing effects of cigarettes per day on the odds of being a morning person |
| 14 | Jansen et al. (2019) | Smoking | Summary level | Insomnia (self-report), cigarettes per day (self-report) | Insomnia (self-report), cigarettes per day (self-report) | – + | Evidence for causal, increasing effects of insomnia on cigarettes per day. No clear evidence for causal effects of cigarettes per day on insomnia |
| 15 | Lane et al. (2019) | Smoking | Summary level | Insomnia (self-report) | Smoking initiation, age at smoking initiation, cigarettes per day, smoking cessation (all self-report) | – + | No clear evidence for causal effects |
| 16 | Bjorngaard et al. (2013) | Smoking | Individual level | Current smoking, cigarettes per day (all self-report) | Anxiety and depression (self-reported Hospital Anxiety and Depression Scale) | – + | Conflicting findings: in the whole sample evidence for causal, increasing effects of smoking on anxiety, but when stratified, effects were very weak in smokers (stronger in former and never smokers). No clear evidence for causal effects of smoking on depression |
| 17 | Lewis et al. (2011) | Smoking | Individual level | Smoking status (current, former, never smoker), cigarettes per day (all self-report) | Postnatal depression (self-reported Edinburgh Postnatal Depression Scale) | – | Weak evidence for causal, decreasing effects of smoking status and cigarettes per day on depressed mood during pregnancy |
| 18 | Taylor et al. (2014a) | Smoking | Individual level | Cigarettes per day (self-report) | Psychological distress (composite score derived from a range of self-report or symptom scale or diagnosis measures) | – + | No clear evidence for causal effects |
| 19 | Skov-Ettrup et al. (2017) | Smoking | Individual level | Cigarettes per day, pack years of cigarettes (all self-report) | Psychological distress (3 questions on stress, fatigue & hopelessness – all self-report) | – + | No clear evidence for causal effects |
| 20 | Wootton et al. (2019) | Smoking | Summary level | Lifetime smoking (self-report), smoking initiation (self-report); schizophrenia (diagnosis); major depression (diagnosis) | Lifetime smoking (self-report), smoking initiation (self-report), schizophrenia (diagnosis), major depression (diagnosis) | + | Evidence for causal, increasing effects of smoking initiation and of lifetime smoking on schizophrenia and on depression. Evidence (less strong) for causal, increasing effects of depression on smoking initiation and lifetime smoking, and of schizophrenia on lifetime smoking |
| 21 | Vermeulen et al. (2019) | Smoking | Summary level | Lifetime smoking (self-report), smoking initiation (self-report), bipolar disorder (diagnosis) | Lifetime smoking (self-report), smoking initiation (self-report), cigarettes per day (self-report), smoking cessation (self-report); bipolar disorder (diagnosis) | + | Evidence for causal, increasing effects of smoking initiation and of lifetime smoking on bipolar disorder. No clear evidence for causal effects of bipolar disorder on smoking |
| 22b | Barkhuizen et al. (2020) | Smoking | Summary level | Smoking initiation (self-report), psychotic experiences (self-report), schizophrenia (diagnosis), major depression (diagnosis), bipolar disorder (diagnosis) | Smoking initiation (self-report); psychotic experiences (self-report), schizophrenia (diagnosis); major depression (diagnosis); bipolar disorder (diagnosis) | – + | Evidence for causal, increasing effects of smoking initiation on major depression, bipolar disorder and cognitive disorganization. Weak evidence for causal, increasing effects of smoking initiation on schizophrenia and on negative symptoms. Weak evidence for causal, increasing effects of schizophrenia on smoking initiation. No clear evidence for causal effects in any of the other tested relationships |
| 23 | Wium-Andersen et al. (2015a) | Smoking | Individual level | Cigarettes per day (self-report) | Antipsychotic medication use (national health records), schizophrenia (diagnosis); antidepressant use (national health records), major depression (diagnosis) | – + | Weak evidence for causal, increasing effects of cigarettes per day on antipsychotic medication use and schizophrenia, but no clear evidence for causal effects on depression |
| 24 | Byrne et al. (2019) | Smoking | Summary level | Cigarettes per day (self-report) | Schizophrenia (diagnosis) | – + | Evidence for causal, increasing effects of cigarettes per day on schizophrenia |
| 25 | Gage et al. (2017b) | Smoking | Summary level | Smoking initiation (self-report), schizophrenia (diagnosis) | Schizophrenia (diagnosis); smoking initiation (self-report), cigarettes per day (self-report), smoking cessation (self-report) | – + | Evidence for causal, increasing effects of smoking initiation on schizophrenia. No clear evidence for causal effects of schizophrenia on smoking |
| 26 | Fluharty et al. (2018) | Smoking | Summary level | Childhood aggression (parental report; meta-analysis of continuous study-specific scales) and attention-deficit hyperactivity disorder (diagnosis) | Smoking initiation, age at onset smoking (all self-report) | – + | Evidence for causal, increasing effects of attention-deficit hyperactivity disorder on smoking initiation. No clear evidence for causal effects of attention-deficit hyperactivity disorder on age at onset smoking nor of aggression on smoking |
| 27 | Sallis et al. (2019) | Smoking | Both | Extraversion, neuroticism, smoking initiation, cigarettes per day, smoking cessation (all self-report) | Extraversion, neuroticism; smoking initiation, cigarettes per day, smoking cessation (all self-report) | – + | Evidence for causal, increasing effects of neuroticism on cigarettes per day and for causal, increasing effects of extraversion on smoking initiation No clear evidence for causal effects of smoking on extraversion or neuroticism |
| 28b | Leppert et al. (2019) | Smoking | Summary level | attention-deficit hyperactivity disorder (diagnosis) | Lifetime smoking (self-report) | –a | Evidence for causal, increasing effects of attention-deficit hyperactivity disorder on lifetime smoking |
| 29 | Harrison et al. (2020a) | Smoking | Both | Smoking initiation, cigarettes per day, lifetime smoking (all self-report) | Suicidal ideation (self-report) | – + | No clear evidence for causal effects |
| 30 | Rosoff et al. (2019) | Alcohol | Summary level | Educational attainment (self-report); frequency of alcohol use (self-report), alcohol drinks per week (self-report), specific alcohol types in drinks per week (self-report), problematic alcohol use (self-reported alcohol use disorders identification test), alcohol use disorder (diagnosis), individual alcohol use disorder symptoms (self-report) | Educational attainment (self-report); frequency of alcohol use (self-report), alcohol drinks per week (self-report), specific alcohol types in drinks per week (self-report), problematic alcohol use (self-reported alcohol use disorders identification test), alcohol use disorder (diagnosis), individual alcohol use disorder symptoms (self-report) | – + | Evidence for causal, decreasing effects of education on total drinks per drinking day, weekly spirits intake, binge drinking, and alcohol use disorder. Evidence for causal, increasing effects of education on alcohol intake frequency, weekly wine intake. In the other direction, evidence for causal, increasing effects of weekly wine and champagne intake and frequency of alcohol use on education, and evidence for causal, decreasing effects of weekly beer and cider intake on education |
| 31 | Zhou et al. (2019b) | Alcohol | Summary level | Years of education (self-report) | Alcohol use frequency (self-report), frequency of different types of alcohol use (self-report) | – + | Evidence for causal, increasing effects of educational attainment on alcohol use frequency, frequency of red wine use, and frequency of white wine/champagne use. Evidence for causal, decreasing effects of educational attainment on frequency of beer/cider and spirits |
| 32 | Kumari et al. (2014) | Alcohol | Individual level | Alcohol initiation, frequency of alcohol use (all self-report) | Cognitive functioning (word recall, verbal fluency, processing speed tasks) | – | No clear evidence for causal effects |
| 33 | Almeida et al. (2014a) | Alcohol | Individual level | Frequency of alcohol use (self-report) | Cognitive impairment (in-person mini-mental state examination) | – | No clear evidence for causal effects |
| 34 | Ritchie et al. (2014) | Alcohol | Individual level | Alcohol use in gram per day (self-report) | Cognitive functioning (in-person Moray House Test No. 12) | – | No clear evidence for causal effects of alcohol on cognitive functioning, However, there was an interaction such that individuals with higher genetic ability to process alcohol showed relative improvements in cognitive ability with more consumption, whereas those with low processing capacity showed a negative relationship |
| 35 | Au Yeung et al. (2012) | Alcohol | Individual level | Alcohol drinks per day (in-person interview) | Cognitive functioning (10-word list learning task + in-person mini-mental state examination) | – | No clear evidence for causal effects |
| 36 | Mahedy et al. (2020) | Alcohol | Summary level | Alcohol drinks per week (self-report) | Working memory, response inhibition, emotion recognition (all in-clinic test assessments) | – | No clear evidence for causal effects |
| 37 | Andrews et al. (2020) | Alcohol | Summary level | Alcohol drinks per week (self-report), problematic alcohol use (self-reported Alcohol Use Disorders Identification Test), alcohol use disorder (diagnosis) | Alzheimer's disease (diagnosis), Alzheimer's disease age of onset (diagnosis) | – + | No clear evidence for causal effects of alcohol on Alzheimer's disease diagnosis. Evidence that higher number of alcohol drinks per week causes earlier Alzheimer's disease onset. Contradicting, there was evidence that alcohol use disorder caused later disease onset |
| 38 | Nishiyama et al. (2019) | Alcohol | Individual level | Alcohol drinking days per week; cups of coffee per day (all self-report) | Hours of sleep per night (self-report) | – | Evidence that alcohol causes longer sleep duration, no clear evidence for causal effects of coffee on sleep |
| 39 | Almeida et al. (2014b) | Alcohol | Individual level | Frequency of alcohol use (self-report) | Depression (self-report on receiving treatment for or being diagnosed with depression; for a subgroup diagnosis obtained from national health records) | – | No clear evidence for causal effects |
| 40 | Wium-Andersen et al. (2015b) | Alcohol | Individual level | Alcohol drinks per week (self-report) | Depression (diagnosis obtained from national health records), psychological distress (self-report) | – + | No clear evidence for causal effects |
| 41 | Polimanti et al. (2019) | Alcohol | Summary level | Major depression (diagnosis); alcohol use disorder (diagnosis), alcohol use frequency (self-report), alcohol drinks per week (self-report) | Major depression (diagnosis); alcohol use disorder (diagnosis), alcohol use frequency (self-report), alcohol drinks per week (self-report) | + | Evidence for causal, increasing effects of major depression on alcohol use disorder. No clear evidence for causal effects of major depression on the other alcohol use variables, nor for causal effects in the other direction |
| 42 | Zhou et al. (2020) | Alcohol | Summary level | Major depression (diagnosis), schizophrenia (diagnosis), bipolar disorder (diagnosis), depressed effect (self-report), neuroticism (self-report), worrying (self-report), insomnia (self-report), cognitive functioning (fluid intelligence), educational attainment (self-report), alcohol use disorder (diagnosis) | Major depression (diagnosis), schizophrenia (diagnosis), bipolar disorder (diagnosis), depressed affect (self-report), neuroticism (self-report), worrying (self-report), insomnia (self-report), cognitive functioning (fluid intelligence), educational attainment (self-report), alcohol use disorder (diagnosis) | + | Evidence for causal, increasing effects of worrying and neuroticism on alcohol use disorder. Evidence for causal, decreasing effects of cognitive functioning and educational attainment on alcohol use disorder. Evidence for causal, decreasing effects of alcohol use disorder and education. No clear evidence for causal effects in any of the other tested relationships |
| 43 | Irons et al. (2007) | Alcohol | Individual level | Alcohol initiation, past year use of alcohol, past year drinking index, past year drunkenness index (all self-report), alcohol use disorder (clinical, in-person interview) | Antisocial personality disorder, delinquent behavior inventory (all clinical, in-person interview), exposure to bad peer models (self-report) | – | No clear evidence for causal effects |
| 44 | Chao et al. (2017) | Alcohol | Individual level | Alcohol use frequency, alcohol drinks per typical drinking occasion, desire to drink (all self-report) | Externalizing problems (Youth Self-Report), internalizing problems (self-report on Children's Depression Inventory and State-trait Anxiety Inventory) | – | Evidence for causal, increasing effects of alcohol on aggression and attention problems but no clear evidence for effects on delinquency, anxiety, or depression |
| 45 | Hodgson et al. (2020) | Cannabis | Summary level | Cannabis initiation (self-report), major depression (diagnosis) | Cannabis initiation (self-report), major depression (diagnosis) | – + | No clear evidence for causal effects |
| 46 | Soler Artigas et al. (2019) | Cannabis | Summary level | attention-deficit hyperactivity disorder (diagnosis), cannabis initiation (self-report) | Attention-deficit hyperactivity disorder (diagnosis), cannabis initiation (self-report) | – + | Evidence for causal, increasing effects of attention-deficit hyperactivity disorder on cannabis initiation. No clear evidence causal effects of cannabis initiation on attention-deficit hyperactivity disorder |
| 47 | Pasman et al. (2018) | Cannabis | Summary level | Cannabis initiation (self-report), schizophrenia (diagnosis) | Cannabis initiation (self-report), schizophrenia (diagnosis) | – + | Evidence for causal, increasing effects of schizophrenia on cannabis initiation. Weak evidence for causal, increasing effects of cannabis initiation on schizophrenia |
| 48 | Vaucher et al. (2018) | Cannabis | Summary level | Cannabis initiation (self-report) | Schizophrenia (diagnosis) | – + | Evidence for causal, increasing effects of cannabis initiation on schizophrenia |
| 49 | Gage et al. (2017a) | Cannabis | Summary level | Cannabis initiation (self-report); schizophrenia (diagnosis) | Cannabis initiation (self-report); schizophrenia (diagnosis) | – + | Evidence for causal, increasing effects of schizophrenia on cannabis initiation. Weak evidence for causal, increasing effects of cannabis initiation on schizophrenia |
| 50 | Zhou et al. (2018) | Coffee | ndividual level | Cups of coffee per day (self-report) | Cognitive functioning (composite global cognition & memory scores, derived from a range of different cognitive functioning tests) | – + | No clear evidence for causal effects |
| 51 | Treur et al. (2018) | Coffee | Summary level | Cups of coffee (self-report), plasma caffeine (measured in blood), caffeine metabolic ratio (measured in blood), sleep duration (self-report), chronotype (self-report), insomnia (self-report) | Cups of coffee (self-report), plasma caffeine (measured in blood), caffeine metabolic ratio (measured in blood); sleep duration (self-report), chronotype (self-report), insomnia (self-report) | – + | Weak evidence for causal, decreasing effects of higher plasma caffeine levels on the odds of being a morning person. No clear evidence for causal effects in any of the other tested relationships |
| 52 | Kwok et al. (2016) | Coffee | Summary level | Cups of coffee per day (self-report) | Major depression (diagnosis), Alzheimer's disease (diagnosis) | – + | No clear evidence for causal effects |
| 53 | Ding et al. (2019) | Multiple: smoking alcohol | Individual level | Years of education (self-report) | Current smoking (self-report); alcohol drinking days per week (self-report) | – | Evidence for causal, decreasing effects of educational attainment on current smoking. No clear evidence for causal effects of educational attainment on alcohol drinking days per week |
| 54b | Yuan et al. (2020a) | Multiple: smoking alcohol : | Summary level | Educational attainment (self-report); cognitive functioning (fluid intelligence) | Age at smoking initiation, cigarettes smoked per day, alcohol drinks per week (all self-report) | –a | Evidence for causal, increasing effects of educational attainment on age at onset smoking and decreasing effect on cigarettes per day, the effect remained the same when adjusted for cognitive functioning. Evidence for causal, increasing effects of cognitive functioning on age at onset smoking and decreasing effect on cigarettes per day, but when adjusted for educational attainment, the effect was largely attenuated Evidence for causal, increasing effects of educational attainment on alcohol drinks per week, when adjusted for cognitive functioning, this effect was attenuated. Evidence for causal, increasing effects of cognitive functioning on alcohol drinks per week, the effect remained the same when adjusted for educational attainment |
| 55 | Davies et al. (2019) | Multiple: smoking alcohol | Both | Years of school (self-report), cognitive functioning (fluid intelligence) | Smoking initiation, current smoking; alcohol use frequency (all self-report) | + | Evidence for causal, increasing effects of cognitive functioning on alcohol use frequency. Evidence for causal, decreasing effects of years of school on smoking initiation and current smoking |
| 56b | Davies et al. (2018a) | Multiple: smoking alcohol | Both | Years of education (self-report) | Alcohol use frequency; smoking initiation, current smoking (all self-report) | + | Evidence for causal, increasing effects of more years of education on alcohol use frequency, and evidence for causal, decreasing effects of more years of education on smoking initiation and current smoking |
| 57 | Harrison et al. (2020b) | Multiple: smoking alcohol | Both | Alcohol drinks per week; smoking initiation, lifetime smoking (all self-report) | Education (self-reported university degree status), loneliness (self-report) | – + | Evidence for causal, decreasing effects of lifetime smoking and smoking initiation on education. No clear evidence for causal effects in any of the other tested relationships |
| 58 | Mahedy et al. (2021) | Multiple: smoking cannabis | Both | Smoking initiation (self-report); cannabis initiation (self-report) | Working memory, response inhibition, emotion recognition (all in-clinic test assessments) | – | No clear evidence for causal effects |
| 59 | Andrews et al. (2021) | Multiple: smoking alcohol | Summary level | Alcohol drinks per week (self-report), problematic alcohol use (self-reported alcohol use disorders identification test); smoking initiation (self-report), cigarettes per day (self-report) | Alzheimer's disease (diagnosis), Alzheimer's disease age at onset (disorder) | + | No clear evidence for causal effects |
| 60 | Larsson et al. (2017) | Multiple: smoking alcohol coffee | Summary level | Smoking initiation, cigarettes per day, smoking cessation; alcohol drinks per week; cups of coffee per day (all self-report) | Alzheimer's disease (diagnosis) | – + | Weak evidence for a causal, decreasing effect of cigarettes smoked per day on Alzheimer's disease. Weak evidence for a causal, increasing effect of coffee on Alzheimer's disease. No clear evidence for causal effects of smoking initiation, smoking cessation or alcohol |
| 61 | Wootton et al. (2020) | Multiple: smoking alcohol | Summary level | Alcohol drinks per week (self-report), alcohol use disorder (diagnosis), smoking initiation (self-report), loneliness (self-report) | Alcohol drinks per week (self-report), alcohol use disorder (diagnosis), smoking initiation (self-report), cigarettes per day (self-report), smoking cessation (self-report); loneliness (self-report) | – + | Weak evidence for causal, increasing effects of loneliness on smoking initiation and cigarettes per day, weak evidence for causal, decreasing effects of loneliness on smoking cessation. Strong evidence for an effect such that smoking initiation increases loneliness. No clear evidence for causal effects in any of the other tested relationships |
| 62 | Lim et al. (2020) | Multiple: alcohol cannabis | Summary level | Cannabis initiation (self-report), alcohol use disorder (diagnosis) | Non-suicidal self-harm (NSSH), suicidal self-harm (SSH) (self-report) | – + | No clear evidence for causal effects |
| 63 | Treur et al. (2019) | Multiple: smoking alcohol cannabis coffee | Summary level | Smoking initiation (self-report); alcohol drinks per week (self-report), problematic alcohol use (self-reported Alcohol Use Disorders Identification Test), alcohol use disorder (diagnosis); cannabis initiation (self-report); cups of coffee per day (self-report), attention-deficit hyperactivity disorder (diagnosis) | Smoking initiation (self-report), cigarettes per day (self-report) smoking cessation (self-report), lifetime smoking (self-report), alcohol drinks per week (self-report), problematic alcohol use (self-reported alcohol use disorders identification test), alcohol use disorder (diagnosis), cannabis initiation (self-report), cups of coffee per day (self-report), attention-deficit hyperactivity disorder (diagnosis in adulthood) | + | Evidence for causal, increasing effects of attention-deficit hyperactivity disorder on smoking initiation, cigarettes per day, smoking cessation and cannabis initiation. Weak evidence for causal, increasing effects of attention-deficit hyperactivity disorder on alcohol use disorder. No clear evidence for causal effects of attention-deficit hyperactivity disorder on the other alcohol measures nor on cups of coffee per day In the other direction, weak evidence for causal, increasing effects of smoking initiation on attention-deficit hyperactivity disorder risk |
a
This score pertains to the relationship that is of interest to the current systematic review, and not necessarily the whole study. For instance, it may be that in the study as a whole (more) extensive MR sensitivity methods were performed but for the causal estimate of interest no sensitivity methods were applied (e.g. when smoking is merely used as a mediator in a multivariable MR study).
b
Pre-print publication (not peer-reviewed) obtained from bioRxiv.org, medRxiv.org or arXiv.org.
Note that the quality rating is based on a number of key indices, the most important being: phenotype measurement (sample size, quality of the exposure measurement, quality of the outcome measurement), instrument strength (p value threshold used to select genetic variants, number of genetic variants included, biological knowledge, F statistic for instrument strength, % variance that the instrument explains), and analytical factors (type of main analysis, whether or not basic sensitivity analyses were applied, whether or not additional sensitivity analyses were applied). Combined, these indices were weighted to come to a complete quality score (see Supplementary Table S1). A few important notes regarding this weighting of the evidence: (1) where absolute thresholds were used to judge the quality of a particular aspect of the study (e.g. sample size), it should be noted that these are somewhat arbitrary and were merely used to provide an indication of quality. (2) With regard to ‘phenotype measurement,’ a very well measured phenotype in a moderate sample size may be just as powerful as a more superficially measured phenotype in a very large sample. However, in case of very small sample sizes (e.g. n = 180 such as in the study by Irons et al., 2007) even an extremely thoroughly measured phenotype will not lead to a high total score. (3) With regard to ‘instrument strength,’ when a study uses a single genetic variant that explains a relatively large amount of the variance and for which there is good biological knowledge, the fact that only one SNP was used is not necessarily problematic. For example, this is the case for SNP rs1051730 in the nicotinic acetylcholine receptor CHRNA5/A3/B4 gene cluster – each additional risk allele increases smoking heaviness with one additional cigarette smoked per day (Katikireddi, Green, Taylor, Davey Smith, and Munafò, 2018).