Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 Jan 1.
Published in final edited form as: Drug Alcohol Depend. 2021 Nov 27;230:109200. doi: 10.1016/j.drugalcdep.2021.109200

Effects of cannabis use on physical health using a co-twin control design

J Megan Ross 1, Jarrod M Ellingson 1, Maia J Frieser 2, Robin C Corley 3, Christian J Hopfer 1, Michael C Stallings 3,4, Sally J Wadsworth 3, Chandra A Reynolds 5, John K Hewitt 3,4
PMCID: PMC8714702  NIHMSID: NIHMS1761784  PMID: 34871975

Abstract

Background.

Research on the influence of cannabis use on anthropometrics, cardiovascular and pulmonary functioning, and other indicators of physical health has reported mixed results. We examined whether cannabis frequency is associated with physical health outcomes phenotypically and after controlling for shared genetic and environmental factors via a longitudinal co-twin control design.

Methods.

We tested the phenotypic associations of adolescent, young adult, and adult cannabis frequency with adult physical health. Next, we ran multilevel models to test if significant phenotypic associations remained at the between-family and within-twin pairs levels. Participants include 677 individual twins (308 twin pairs) aged 20–35.

Results.

At the phenotypic level, adolescent cannabis use was associated with less adult exercise engagement (b=−0.846 minutes, p=.000). Adult cannabis use was associated with a lower heart rate (HR; b=−0.170 bpm, p=.001) and more frequent appetite loss (b=0.018, p=.000). Only between-family effects were significant for adolescent cannabis use and exercise engagement (b=−1.147 minutes, p=.000) and adult cannabis use and appetite loss frequency (b=0.041, p=.002). The total within-twin (b=−0.184, p=.014), MZ only (b=−0.304, p=.003), and between-family effects (b=−0.164, p=.025) were significant between adult cannabis use and a lower HR, which persisted after controlling for familial confounds and other substance use.

Conclusions.

The associations between cannabis use with exercise engagement and frequency of appetite loss are explained by familial confounding while the association between cannabis use and resting HR was not. These results do not support a causal association between cannabis use once a week and poorer physical health effects among adults aged 20–35.

Keywords: cannabis, physical health, co-twin control, body mass index

1. INTRODUCTION

There are disparate findings about whether cannabis negatively affects physical health. The consequences of cannabis use on physical health are not well understood, specifically, if and how cannabis use influences body mass index (BMI), cardiovascular function, pulmonary/respiratory function, and other indicators of physical health. It is well-documented that tobacco use has numerous negative impacts on health, including a strong link with lung cancer and chronic obstructive pulmonary diseases (Sethi & Rochester, 2000). Although cannabis and tobacco smoke have some similarities regarding chemical compositions, like carcinogens (Owen et al., 2014), there are differences in the main psychoactive compounds and how each substance is smoked (Simmons & Tashkin, 1995); thus, it is unknown whether cannabis affects physical health in similar ways as tobacco.

There is inconsistent evidence regarding the long-term effects of cannabis on BMI. Some studies have reported no association between cannabis use and BMI (Barry & Petry, 2009; Jin et al., 2017; Levendal et al., 2012; Rooke et al., 2013), while other cross-sectional studies have found a positive association between cannabis use and BMI (Liemburg et al., 2016; Ross et al., 2016), abdominal fat (Muniyappa et al., 2013), and metabolic syndrome (Yankey et al., 2016). However, the most consistent evidence has been a negative association between cannabis use and BMI when comparing users and non-users (Danielsson et al., 2016; Gerberich et al., 2003; Penner et al., 2013) and when examining a dose-dependent association (Hayatbakhsh et al., 2010; Meier et al., 2016; Meier et al., 2019; Ross, Pacheco-Colón, et al., 2020). Cannabis use is also associated with other factors that are commonly associated with lower BMI including lower rates and risk factors of diabetes (Alshaarawy & Anthony, 2015; Ngueta et al., 2015; Penner et al., 2013) and cardiometabolic syndrome diagnoses (Meier et al., 2019; Waterreus et al., 2016).

The long-term effects of cannabis use on cardiovascular health are unclear. One experimental study reported associations between higher saliva THC levels and increased heart rate (HR; Menkes et al., 1991). Case reports have also been published on rare cardiovascular deaths among young adults who recently used (Bachs & Mørland, 2001; Mittleman et al., 2001). One study found that HR variability significantly increased among those who tested positive for THC compared to controls (Schmid et al., 2010). Research on the impact of cannabis use on blood pressure (BP), another indicator of cardiovascular health, is currently inconclusive. One study of over 12,000 adults found that recently active cannabis users had higher systolic BP compared to non-users and a dose-dependent association between past 30-day use and systolic BP (Alshaarawy & Elbaz, 2016). Contrary to the previous study, two studies reported that greater cannabis use is associated with decreases in systolic/diastolic BP (Meier et al., 2016; Meier et al., 2019).

Pulmonary/respiratory function is another area of physical health with conflicting results regarding an association with cannabis use. A systematic review of experimental studies on short-term cannabis exposure reported an increase in forced expiratory volume in 1 second (FEV1; total amount of air exhaled in 1-second) by 0.15–0.25 L (Tetrault et al., 2007). Other studies have found a positive dose-dependent association with cannabis joint-years and elevated forced vital capacity (FVC; total amount of air exhaled forcefully; Meier et al., 2016; Pletcher et al., 2012). Other studies have reported that cannabis use is associated with reduction in FEV1 and FEV1/FVC ratio; however, these associations were negligible after controlling for tobacco use (Hancox et al., 2010; Moore et al., 2005; Taylor et al., 2002). A recent review found that cannabis smoking is not associated with measures of pulmonary function (Owens et al., 2014). Although long-term cannabis use is associated with increased cough, phlegm, and wheezing (Tetrault et al., 2007), there is inconsistent evidence for whether long-term use is associated with changes in FEV1, FVC, or FEV1/FVC.

The literature on associations between cannabis use and physical health is inconsistent. To clarify the role of cannabis use on physical health, we used a co-twin control design, which controls for genetic and environmental factors shared by members of a family, allowing us to conduct a controlled natural experiment. Monozygotic (MZ) twins share 100% of their genetic makeup and shared environmental factors (e.g., grew up in the same household) and dizygotic (DZ) twins share 50% of their genetic makeup and 100% of shared environmental factors. By comparing twins who are discordant for their cannabis use, we can make stronger inferences about the effects of cannabis use on physical health. For example, numerous cross-sectional and longitudinal studies have found support for a lower IQ among individuals who use cannabis (e.g., Meier et al., 2012; Scott et al., 2018). However, recent co-twin control studies have not found support for a causal association between cannabis use and lower IQ (Jackson et al., 2016; Lyons et al., 2004; Meier et al., 2017; Ross et al., 2020). Co-twin control studies suggest that this association found across numerous populations is related to familial confounding. More specifically, those who use cannabis and use cannabis more frequently are more likely to have a lower IQ, independent of cannabis use (Jackson et al., 2016; Lyons et al., 2004; Meier et al., 2017; Ross et al., 2020).

If within-twin pair effects are significant across MZ and DZ pairs and are not attenuated relative to the phenotypic effect, then the observed effect is not due to familial confounds (i.e., supports a causal association). If the DZ within-twin pair effect is about ½ the magnitude of the phenotypic effect and is negligible among MZ twins, this suggests complete confounding by familial factors (i.e., does not support a causal association). If the DZ within-twin pair effect is about ¾ the magnitude of the phenotypic effect and is about ¼ the magnitude among MZ twins, this suggests some confounding by familial factors (i.e., partial support for a causal association; McGue et al., 2010).

We hypothesize that cannabis use will be related to various physical health measures based on prior research like lower anthropometrics, higher resting HR, lower systolic and diastolic BP, more frequent gum disease, better diet/nutrition, and exercising longer while there will be no association between cannabis use and pulmonary function. Given the paucity of research on several of our outcomes, we do not have hypotheses for how cannabis use is associated with right/left hand grip and frequency of loss of appetite, nausea, chronic pain, weight problems, skin problems, rapid HR, headaches, and injuries. Analyses were not pre-registered, and results should be considered exploratory.

2. METHODS

2.1. Participants

Participants are part of an ongoing study, the Colorado Adoption/Twin Study of Lifespan behavioral development and cognitive aging (CATSLife; Wadsworth et al., 2019) which includes twins from the foundational Longitudinal Twin Study (LTS), examining cognitive, emotional, and behavioral development. The original LTS sample consists of same-sex twins assessed from infancy to adulthood totaling 483 twin pairs (Corley et al., 2019; Rhea et al., 2006; 2013). The CATSlife study collected data from 677 individual LTS twins (N=308 complete twin pairs, 164 MZ twin pairs, 144 DZ twin pairs) with data from the adult assessment and the sample size ranges for each analysis depending on whether data is also present from the adolescent and/or young adult assessment. Missing data for outcome variables ranged from 0%–7%. Consistent with the race/ethnic distribution at the time of recruitment, the sample is 53.9% females and 91.6% Caucasian (including 9.7% Hispanic/Latinx), 1.3% American Indian/Alaska Native, 5.6% multiracial, and 1.5% other/unknown/not reported (Corley et al., 2019; Wadsworth et al., 2019). During adolescence, young adulthood, and adulthood 12, 47, and 62 participants, respectively, endorsed daily cannabis use. Around 60% of the sample continues to reside in Colorado. Assessments for the project occurred between 2015–2019, after recreational cannabis was legalized in Colorado. Past month prevalence rates of cannabis, alcohol, and tobacco use in this sample are higher than the national average (Centers for Disease Control and Prevention, 2020), but consistent with Colorado state averages (with the exception of alcohol use which is higher in this study sample; Centers for Disease Control and Prevention, 2020). A more detailed description of the participant characteristics and total N for each measure are included in Table 1 and 2.

Table 1.

Descriptive statistics of substance use

Variable Adolescence Young Adulthood Adulthood
N M (SD) N M (SD) N M (SD)
Age (years) 641 17.24 (0.63) 640 22.79 (127) 677 29.30 (1.24)
Cannabis
 Frequency past month (days) 640 1.15 (4.51) 639 3.15 (8.06) 671 3.94 (8.82)
 Endorsed use in past six months (%, past month for adulthood) 640 23.44 639 31.77 671 28.17
Tobacco
 Frequency past month (days) 624 3.38 (8.85) 639 5.96 (11.09) 672 3.97 (9.63)
 Endorsed use in past six months (%, past month for adulthood) 624 25.80 639 38.18 672 17.71
Alcohol
 Frequency past month (days) 637 0.97 (2.16) 640 5.06 (6.26) 672 7.79 (7.73)
 Endorsed use in past six months (%, past month for adulthood) 637 54.95 640 91.72 672 81.15
Other drugs
 Frequency past month (days) 639 0.30 (2.18) 637 0.79 (3.96) 671 0.36 (2.37)
 Endorsed use in past six months (%, past month for adulthood) 639 10.64 637 19.47 671 6.71
Open in a new tab

Note: M = mean and SD=standard deviation.

Table 2.

Descriptive statistics of adult physical health outcomes

Variable N Mean/Percent SD Range
Age 677 29.30 yrs 1.24 28.05–34.55 yrs
BMI 649 26.34 kg/m2 5.96 16.31–54.93 kg/m2
Waist circumference 649 90.50 cm 15.99 27.00–163.00 cm
Hips circumference 649 104.43 cm 11.85 37.00–159.50 cm
Systolic BP 651 112.60 mmHG 12.32 78.00–148.00 mmHG
Diastolic BP 651 68.74 mmHG 8.81 43.00–101.00 mmHG
Resting HR 651 70.34 beats/min 11.43 39.00–108.00 beats/min
FVC 625 4.56 L 0.99 0.40–7.40 L
FEV1 625 3.63 L 0.77 0.30–6.40 L
FCV/FEV (%) 625 80.09% 7.42 35.48–97.44%
Left hand grip 646 78.80 24.75 20.90–154.50
Right hand grip 648 87.49 26.94 32.70–187.70
Chronic pain (%) 670 62.54% - -
Gum disease (%) 670 17.46% - -
Loss of appetite (%) 670 29.56% - -
Nausea (%) 670 57.01% - -
Weight problems (%) 670 21.64% - -
Problems breathing (%) 670 30.00% - -
Skin problems (%) 670 28.81% - -
Rapid HR (%) 670 37.31% - -
Headaches (%) 670 75.22% - -
Injuries (%) 670 58.21% - -
Unhealthy diet 664 3.77 1.93 0.00–16.00
Healthy diet 661 9.69 4.54 0.00–24.00
Fast food 667 1.21 1.20 0.00–7.00
Exercise engagement in past 24 hours (mins.) 649 21.47 mins. 46.83 0.00–480.00 mins.
Open in a new tab

Note: SD = standard deviation, BMI = Body mass index, BP = blood pressure, HR = heart rate, FVC = forced vital capacity, and FEV1 = forced expiratory volume in one second. Chronic pain, gum disease, loss of appetite, nausea, weight problems, problems breathing, skin problems, rapid HR, headaches, and injuries are calculated as the percent who reported >1 (i.e. less than once a year, about once year, about once a month, once a week, or daily).

The Institutional Review Board at the University of Colorado Boulder approved all study protocols and procedures. During adolescence, participant assent and parental consent were obtained. Once participants turned 18 years old, participant consent was obtained.

2.2. Measures

2.2.1. Substance use.

Participants completed self-report questionnaires that asked the number of days that cannabis, tobacco, alcohol, and other drug use frequency were used in the past six months during the adolescent and young adult assessments (Salomonsen-Sautel et al., 2012) and in the past month during the adult assessment (Hamilton et al., 2011). Participants who said they did not use that substance in the past six months or past month were given a zero. To ease the interpretation of results, we divided the substance frequency variables from adolescence and young adulthood by 6. A one-unit increase corresponds to an increase of one day/month of use. To investigate whether associations were specific to a particular age or from persistent use, we examined whether outcomes were also associated with average lifetime cannabis frequency and average lifetime alcohol, tobacco, and other drug use frequency (see Ellingson et al., 2020). We calculated the mean number of days in the past month that each substance was used across all three assessments as a measure of average lifetime frequency for each substance.

2.2.2. Physical health.

2.2.2.1. Objective measures of physical health.

Research staff administered physical health functioning measures following PhenX Toolkit standard protocols (Hamilton et al., 2011) and only collected health data at the adult measurement wave via the CATSLife assessment (M=29.30 years). Additional details about objective measures of physical health are included in the supplemental materials.

Anthropometrics.

Staff measured height/weight using a balance beam scale with a height rod (ABCO Health-o-meter) and waist/hip circumference using a tape measure in centimeters.

Cardiovascular function.

BP and HR were measured three times at one-minute intervals using a digital BP monitor (Omron IntelliSense machine HEM-907XL). The median was used as the outcome.

Pulmonary function.

Pulmonary function was assessed using a spirometer (NDD EasyOne Plus Frontline Spirometer) via three trials at one-minute intervals. Median score (in liters) was used for forced vital capacity (FVC), forced expiratory volume in 1-second (FEV1), and FEV1/FVC ratio. Forced vital capacity (FVC) is the total air exhaled, while forced expiratory volume in 1-second (FEV1) is the total air exhaled in the first second. FEV1/FVC ratio represents the proportion of vital capacity exhaled in the first second.

Grip Strength.

A hand-held digital dynamometer (Jamar Smart Hand Dynamometer) measured hand grip strength via three trials completed in 20-second intervals. The highest number achieved across the trials was the outcome variable in pounds.

2.2.2.2. Self-reported measures of physical health.

Participants completed questions regarding frequency of chronic pain, gum disease, nausea, weight problems (i.e., large changes in weight), difficulty breathing, skin problems, rapid HR, headaches, and injuries. Response options are 0=never to 5=daily. All self-reported measures were previously developed for the Colorado Adoption Project (Rhea, Bricker, et al., 2013; Svedberg et al., 2005) except for the question about frequency of gum disease which was adapted from the AddHealth study (Chantala & Tabor, 1999). One self-report question developed for the CATSLife study asked if participants exercised in the past 24 hours, and if so, for how long (minutes; Wadsworth et al., 2019). Participants were given a zero if they reported not exercising in the past 24 hours.

Diet/nutrition.

Three questions each were added together to create two separate variables for healthy and unhealthy diet. These questions asked about frequency of consuming healthy (i.e. fruit/vegetables/leafy greens) and unhealthy food (i.e. french fries, doughnuts/muffins/sweet rolls, and cookies/cake/pie/brownies) in the past 30 days. Response options ranged from 0=never to 9=5 times/day. Although all questions were not included in the assessment, these questions were taken from the Nutrition and Dietary Supplement of the PhenX Toolkit (Hamilton et al., 2011), which used the National Cancer Institute’s Five-Factor Screener (National Cancer Institute, 2005). There was one question about frequency of fast food consumption in the past seven days (response options range from 0=0 to 7=>20) which was taken from the AddHealth study (Chantala & Tabor, 1999).

2.3. Analytic Procedures

We estimated regression and multilevel models using Mplus 8 (Muthén & Muthen, 2017). Adolescent, young adult, adult, and lifetime average cannabis frequency were separately run as predictors with each adult physical health variable as the outcome. For each analysis, we estimated the associations with and without controlling for the parallel measures of tobacco, alcohol, and other drug use frequency. Sex was controlled for in all models (grand-mean centered). We only report unstandardized coefficients because standardized coefficients are not recommended for multilevel models (Nezlek, 2012). To correct for multiple testing in phenotypic models, we used the p.adjust function from the stats package in R which applies the Hochberg’s correction (i.e., provides adjusted p-values; Hochberg, 1988). Multiple testing correction was applied as a family-wise correction to each drug category and at each wave. Only effects that are significant after correcting for multiple testing are followed-up with co-twin control analyses. This is the first study (in the searchable literature) to examine phenotypic and between- and within-twin pair associations between cannabis use and physical health. Thus, our aims being exploratory guided our decision to examine all potential associations between cannabis use and physical health across all analyses. We also estimated univariate and bivariate biometrical twin models to confirm co-twin control results and a description of the analytic procedures, results, and table are in the supplemental materials.

2.3.1. Model 1: Phenotypic Associations.

Phenotypic associations were estimated between cannabis exposure during adolescence, young adulthood, adulthood, and lifetime average cannabis frequency and measures of adult physical health. The “type=complex” option was used for phenotypic analyses because it accounts for the non-independence of twin pairs by using a sandwich estimator, providing the same estimates as a standard regression but corrects the standard errors for non-independence.

2.3.2. Model 2: Within-Family Exposure.

We estimated multilevel regression models between cannabis frequency and physical health between- (first-level) and within-families (second-level) only if cannabis frequency was significantly associated with a specific physical health outcome in the phenotypic analyses. Linear and logistic multilevel models were conducted for continuous and categorical outcomes, respectively. Cannabis frequency was averaged in twin pairs, which was used to estimate between-family effects. Each twin’s deviation from the twin pair average of cannabis frequency was used to estimate twin-specific risk (i.e., within-family effect). This within-family index assessed the effect of differential exposure in each family, which takes into account the unmeasured shared environmental and genetic factors that make twins alike (Neuhaus & McCulloch, 2006). Random between-family intercepts were included for health outcomes while fixed effects were estimated for each substance use variable. Lastly, we included an interaction term for the within-family effect by zygosity. We calculated MZ within-family effects, DZ within-family effects, and the average within-family effect for both MZ and DZ twins (Carlin et al., 2005). In addition, when covariates were added to the models (tobacco, alcohol, and other drug use frequency), we calculated average within-family effects of these variables for both MZ and DZ twins. We grand-mean centered all independent variables, except for individual within-family effects which were centered within each twin pair. More details about model 2 are included in the supplemental material.

3. RESULTS

3.1. Model 1: Phenotypic Associations

Table 3 displays the associations between cannabis frequency and measures of physical health while controlling for sex and controlling for sex, tobacco, alcohol, and other drug use frequency.

Table 3.

Unstandardized coefficients for the phenotypic analyses of the association between cannabis frequency with physical health

Cannabis only Controlling for tobacco, alcohol, and other drug use
Cannabis effect (95% CI) p-value Cannabis effect (95% CI) p-value Tobacco effect (95% CI) p-value Alcohol effect (95% CI) p-value Other drug effects (95% CI) p-value
Adolescent cannabis use
 BMI −0.040 (−0.150, 0.070) .472 −0.092 (−0.234, −0.008) .038 0.151 (0.022, 0.178) .011 −0.015 (−0.316, 0.233) .768 0.029 (−0.365, 0.541) .705
 Waist circumference 0.087 (−0.168, 0.343) .502 −0.117 (−0.385, 0.152) .395 0.266 (0.058, 0.474) .012 −0.172 (−0.865, 0.522) .628 0.155 (−0.999, 1.309) .792
 Hip circumference −0.113 (−0.288, 0.061) .204 −0.205 (−0.409, −0.001) .049 0.130 (−0.039, 0.299) .132 −0.133 (−0.606, 0341) .583 0.073 (−0.778, 0.924) .866
 Systolic BP 0.067 (−0.096, 0.230) .418 0.000 (−0.178, 0.177) .997 0.056 (−0.059, 0.171) .340 0.214 (−0.207, 0.634) .319 −0.209 (−0.551, 0.132) .229
 Diastolic BP 0.072 (−0.065, 0.209) .301 −0.021 (−0.181, 0.139) .798 0.102 (0.014, 0.190) .023 −0.021 (−0.357, 0.314) .902 0.170 (−0.192, 0.533) .357
 Resting HR 0.064 (−0.205, 0.333) .643 −0.103 (−0.380, 0.174) .468 0.183 (0.064, 0.302) .003 0.115 (−0.352, 0.582) .631 −0.135 (−0.539, 0.268) .511
 FCV 0.016 (−0.003, 0.035) .092 0.004 (−0.022, 0.030) .762 −0.002 (−0.009, 0.005) .543 0.012 (−0.022, 0.047) .490 −0.001 (−0.021, 0.020) .947
 FEV1 0.011 (−0.005, 0.027) .186 0.005 (−0.017, 0.026) .679 −0.001 (−0.006, 0.004) .714 0.006 (−0.026, 0.037) .719 0.003 (−0.017, 0.022) .770
 FCV/FEV1 (%) −0.038 (−0.124, 0.049) .396 0.093 (−0.095, 0.281) .332 0.019 (−0.060, 0.098) .642 −0.064 (−0.397, 0.270) .708 0.026 (−0.131, 0.183) .744
 Left hand grip 0.089 (−0.212, 0.391) .562 −0.031 (−0.346, 0.284) .848 0.161 (−0.030, 0.353) .098 −0.096 (−0.875, 0.684) .810 −0.311 (−0.873, 0.252) .279
 Right hand grip 0.093 (−0.255, 0.441) .601 0.124 (−0.222, 0.471) .482 0.114 (−0.117, 0.345) .334 −0.282 (−0.946, 0.382) .405 −0.421 (−1.102, 0.260) .226
 Chronic pain 0.002 (−0.019, 0.023) .860 −0.007 (−0.036, 0.023) .652 0.022 (0.006, 0.039) .009 −0.042 (−0.104, 0.019) .178 0.019 (−0.053, 0.090) .608
 Gum disease −0.001 (−0.030, 0.028) .928 −0.015 (−0.029, −0.001) .040 0.019 (0.002, 0.037) .026 −0.027 (−0.056, 0.003) .074 0.002 (−0.025, 0.028) .901
 Loss of appetite 0.012 (−0.006, −0.079) .183 0.003 (−0.021, 0.028) .784 0.017 (−0.001, 0.036) .058 −0.047 (−0.083, −0.010) .011 0.068 (0.016, 0.120) .010
 Nausea −0.007 (−0.028, 0.015) .547 −0.019 (−0.042, 0.003) .085 0.013 (−0.002, 0.029) .089 −0.025 (−0.077, 0.028) .362 0.034 (−0.010, 0.078) .131
 Weight problems 0.015 (−0.010, 0.040) .245 −0.008 (−0.022, 0.007) .304 0.018 (0.007, 0.030) .002 −0.004 (−0.035, 0.026) .779 0.032 (−0.017, 0.082) .201
 Problems breathing 0.012 (−0.010, 0.034) .284 0.000 (−0.028, 0.027) .982 0.020 (0.005, 0.035) .008 −0.014 (−0.065, 0.037) .593 0.005 (−0.034, 0.044) .808
 Skin problems 0.013 (−0.012, 0.038) .311 0.004 (−0.007, 0.014) .509 0.002 (−0.003, 0.007) .449 −0.006 (−0.025, 0.014) .570 −0.002 (−0.017, 0.013) .798
 Rapid HR 0.020 (0.000, 0.039) .055 0.003 (−0.033, 0.039) .855 0.027 (0.011, 0.044) .001 −0.030 (−0.069, 0.009) .130 0.069 (0.013, 0.112) .002
 Headaches 0.003 (−0.018, 0.024) .792 0.002 (−0.026, 0.029) .911 0.006 (−0.008, 0.021) .394 −0.028 (−0.084, 0.028) .321 0.004 (−0.039, 0.047) .848
 Injuries −0.003 (−0.025, 0.019) .776 −0.004 (−0.018, 0.009) .510 0.001 (−0.007, 0.010) .750 −0.003 (−0.035, 0.030) .859 0.014 (−0.022, 0.049) .446
 Unhealthy diet 0.006 (−0.020, 0.031) .661 −0.001 (−0.029, 0.026) .921 −0.008 (−0.027, 0.011) .417 0.087 (0.005, 0.169) .037 0.008 (−0.041, 0.057) .758
 Healthy diet 0.024 (−0.043, 0.090) .487 0.049 (−0.022, 0.120) .175 −0.014 (−0.065, 0.037) .594 −0.082 (−0.254, 0.089) .347 0.006 (−0.180, 0.192) .951
 Fast food 0.001 (−0.020, 0.022) .914 −0.009 (−0.031, 0.014) .446 0.009 (−0.003, 0.021) .136 0.015 (−0.033, 0.063) .545 −0.002 (−0.037, 0.033) .931
 Exercise engagement −0.846 (−1.237, −0.456) .000 −0.789 (−1.388, −0.190) .010 0.023 (−0.608, 0.654) .942 0.153 (−1.550, 1.857) .860 −0.652 (−1.163, −0.142) .012
Young adult cannabis use
 BMI −0.018 (−0.077, 0.042) .557 −0.024 (−0.085, 0.037) .437 0.085 (0.033, 0.137) .001 −0.109 (−0.189, −0.029) .008 −0.067 (−0.184, 0.050) .261
 Waist circumference 0.051 (−0.130, 0.233) .578 0.017 (−0.162, 0.196) .854 0.240 (0.106, 0.374) .000 −0.246 (−0.455, −0.038) .020 −0.094 (−0.516, 0.328) .663
 Hip circumference −0.019 (−0.128, 0.090) .736 −0.018 (−0.128, 0.092) .747 0.118 (0.019, 0.217) .019 −0.203 (−0.363, −0.044) .012 −0.105 (−0.415, 0.205) .507
 Systolic BP 0.064 (−0.029, 0.114) .246 0.051 (−0.058, 0.159) .361 −0.012 (−0.094, 0.070) .776 0.110 (−0.039, 0.259) .147 0.053 (−0.225, 0.332) .707
 Diastolic BP 0.000 (−0.093, 0.094) .997 −0.031 (−0.128, 0.065) .523 0.084 (0.017, 0.150) .013 −0.012 (−0.138, 0.113) .846 0.107 (−0.088, 0.303) .282
 Resting HR −0.026 (−0.148, 0.096) .678 −0.078 (−0.197, 0.042) .203 0.124 (0.045, 0.203) .002 −0.014 (−0.171, 0.143) .858 0.217 (−0.106, 0.539) .188
 FCV 0.011 (0.004, 0.018) .003 0.007 (−0.002, 0.016) .132 −0.003 (−0.009, 0.003) .337 0.013 (0.003, 0.023) .013 −0.002 (−0.018, 0.013) .769
 FEV1 0.005 (−0.002, 0.013) .148 0.003 (−0.006, 0.011) .513 −0.004 (−0.009, 0.001) .102 0.010 (0.001, 0.019) .029 0.003 (−0.008, 0.015) .573
 FCV/FEV1 (%) −0.068 (−0.152, 0.017) .117 −0.041 (−0.165, 0.054) .395 −0.039 (−0.093, 0.016) .165 0.001 (−0.099, 0.101) .980 0.097 (−0.083, 0.277) .291
 Left hand grip 0.148 (−0.032, 0.328) .107 0.157 (−0.041, 0.356) .121 0.076 (−0.075, 0.226) .325 0.071 (−0.198, 0.340) .606 −0.538 (−0.910, −0.166) .005
 Right hand grip 0.080 (−0.109, 0.270) .406 0.135 (−0.067, 0.337) .190 0.105 (−0.065, 0.275) .225 0.010 (−0.298, 0.317) .951 −0.380 (−0.798, 0.037) .074
 Chronic pain 0.009 (−0.001, 0.020) .086 0.007 (−0.013, 0.026) .504 0.022 (0.008, 0.036) .003 0.005 (−0.019, 0.029) .673 0.002 (−0.045, 0.049) .930
 Gum disease 0.007 (−0.006, 0.021) .294 0.002 (−0.007, 0.012) .645 0.011 (0.000, 0.022) .043 −0.012 (−0.022, −0.001) .029 0.002 (−0.011, 0.014) .774
 Loss of appetite 0.000 (−0.013, 0.013) .961 −0.008 (−0.020, 0.004) .189 0.021 (0.009, 0.032) .001 −0.002 (−0.020, 0.015) .802 0.038 (0.007, 0.064) .017
 Nausea −0.001 (−0.011, 0.009) .882 −0.007 (−0.020, 0.006) .291 0.013 (0.001, 0.024) .029 0.006 (−0.013, 0.025) .526 0.011 (−0.020, 0.041) .486
 Weight problems 0.008 (−0.006, 0.022) .268 −0.001 (−0.008, 0.006) .722 0.011 (0.003, 0.018) .004 0.005 (−0.007, 0.014) .419 0.012 (−0.005, 0.028) .178
 Problems breathing 0.009 (−0.003, 0.022) .125 0.002 (−0.011, 0.015) .760 0.015 (0.005, 0.025) .003 0.010 (−0.009, 0.030) .280 0.028 (−0.005, 0.060) .096
 Skin problems 0.007 (−0.007, 0.021) .311 0.003 (−0.001, 0.008) .138 −0.003 (−0.006, 0.000) .029 0.001 (−0.004, 0.006) .619 −0.002 (−0.008, 0.005) .655
 Rapid HR 0.011 (−0.003, 0.024) .123 0.002 (−0.010, 0.015) .720 0.015 (0.004, 0.025) .006 0.007 (−0.012, 0.026) .473 0.033 (−0.002, 0.068) .067
 Headaches −0.004 (−0.015, 0.007) .481 −0.007 (−0.021, 0.007) .321 0.012 (0.001, 0.023) .036 −0.008 (−0.025, 0.010) .379 −0.017 (−0.049, 0.015) .301
 Injuries 0.007 (−0.003, 0.018) .182 0.001 (−0.009, 0.011) .861 0.008 (0.000, 0.015) .049 0.012 (−0.002, 0.025) .093 −0.001 (−0.021, 0.020) .959
 Unhealthy diet 0.007 (−0.013, 0.026) .491 0.010 (−0.009, 0.030) .302 −0.001 (−0.018, 0.016) .918 −0.003 (−0.030, 0.023) .795 −0.042 (−0.069, −0.014) .003
 Healthy diet −0.011 (−0.054, 0.033) .624 0.009 (−0.036, 0.054) .694 −0.065 (−0.101, −0.030) .000 0.022 (−0.053, 0.097) .561 −0.022 (−0.106, 0.061) .602
 Fast food −0.002 (−0.012, 0.007) .623 −0.005 (−0.016, 0.007) .450 0.018 (0.009, 0.028) .000 −0.026 (−0.040, −0.013) .000 −0.001 (−0.025, 0.024) .944
 Exercise engagement −0.397 (−0.812, 0.018) .061 −0.353 (−0.792, 0.086) .115 0.028 (−0.394, 0.450) .897 −0.010 (−0.552, 0.533) .972 −0.173 (−0.605, 0.260) .434
Adult cannabis use
 BMI 0.009 (−0.041, 0.060) .714 0.010 (−0.042, 0.061) .717 0.018 (−0.035, 0.071) .515 −0.055 (−0.118, 0.008) .089 −0.084 (−0.184, 0.017) .102
 Waist circumference 0.062 (−0.082, 0.206) .397 0.047 (−0.100, 0.193) .533 0.120 (−0.013, 0.252) .077 −0.154 (−0.316, 0.007) .062 −0.330 (−0.603, −0.057) .018
 Hip circumference −0.004 (−0.102, 0.094) .936 0.007 (−0.094, 0.109) .888 −0.010 (−0.109, 0.090) .852 −0.038 (−0.162, 0.085) .543 −0.235 (−0.475, 0.005) .055
 Systolic BP 0.022 (−0.076, 0.119) .661 0.003 (−0.098, 0.104) .954 0.010 (−0.078, 0.097) .830 0.171 (0.055, 0.287) .004 0.260 (0.054, 0.465) .013
 Diastolic BP −0.008 (−0.093, 0.078) .862 −0.035 (−0.121, 0.052) .433 0.078 (0.005, 0.150) .035 0.046 (−0.048, 0141) .336 0.115 (−0.241, 0.470) .528
 Resting HR −0.170 (−0.275, −0.065) .001 −0.227 (−0.327, −0.127) .000 0.185 (0.095, 0.274) .000 −0.073 (−0.178, 0.032) .175 0.340 (−0.080, 0.759) .112
 FCV 0.009 (0.002, 0.015) .008 0.006 (−0.001, 0.013) .087 −0.002 (−0.009, 0.005) .542 0.010 (0.003, 0.017) .008 0.013 (−0.004, 0.030) .139
 FEV1 0.002 (−0.004, 0.007) .613 0.000 (−0.007, 0.006) .930 −0.002 (−0.008, 0.003) .378 0.007 (0.000, 0.013) .045 0.015 (−0.003, 0.033) .105
 FCV/FEV1 (%) −0.107 (−0.220, 0.035) .008 −0.113 (−0.179, −0.018) .015 −0.012 (−0.070, 0.052) .766 −0.020 (−0.101, 0.061) .629 0.015 (−0.213, 0.304) .731
 Left hand grip 0.150 (−0.009, 0.310) .064 0.148 (−0.006, 0.303) .060 −0.001 (−0.163, 0.161) .991 0.094 (−0.091, 0.280) .318 0.388 (−0.438, 1.213) .357
 Right Hand Grip 0.107 (−0.058, 0.217) .203 0.108 (−0.065, 0.282) .222 0.046 (−0.134, 0.227) .614 0.156 (−0.050, 0.362) .139 0.098 (−0.758, 0.954) .822
 Chronic pain 0.011 (0.002, 0.020) .022 0.010 (−0.008, 0.028) .289 0.032 (0.018, 0.046) .000 −0.004 (−0.022, 0.014) .645 0.024 (−0.049, 0.096) .525
 Gum disease 0.009 (−0.003, 0.021) .140 0.003 (−0.007, 0.013) .582 0.016 (0.004, 0.027) .011 −0.008 (−0.016, 0.001) .074 0.016 (−0.030, 0.062) .498
 Loss of appetite 0.018 (0.009, 0.027) .000 0.017 (0.004, 0.030) .010 0.016 (0.002, 0.030) .022 −0.003 (−0.017, 0.011) .677 −0.003 (−0.042, 0.035) .871
 Nausea 0.010 (0.001, 0.018) .033 0.009 (−0.005, 0.023) .203 0.020 (0.007, 0.033) .003 −0.010 (−0.025, 0.005) .209 −0.010 (−0.044, 0.024) .571
 Weight problems 0.012 (0.000, 0.025) .057 0.003 (−0.004, 0.011) .379 0.012 (0.002, 0.021) .015 −0.007 (−0.015, 0.001) .080 0.003 (−0.017, 0.023) .742
 Problems breathing 0.014 (0.003, 0.026) .015 0.008 (−0.005, 0.022) .220 0.023 (0.010, 0.036) .000 −0.002 (−0.013, 0.009) .711 0.030 (−0.025, 0.085) .287
 Skin problems 0.006 (−0.007, 0.019) .358 0.002 (−0.002, 0.006) .342 −0.002 (−0.005, 0.002) .356 0.000 (−0.005, 0.004) .881 0.006 (−0.010, 0.023) .448
 Rapid HR 0.015 (0.004, 0.026) .009 0.010 (−0.004, 0.025) .165 0.021 (0.008, 0.035) .002 0.007 (−0.006, 0.020) .272 0.040 (−0.005, 0.086) .084
 Headaches −0.003 (−0.013, 0.007) .537 −0.009 (−0.022, 0.005) .201 0.016 (0.004, 0.028) .007 0.000 (−0.015, 0.015) .973 0.000 (−0.046, 0.045) .987
 Injuries 0.007 (−0.003, 0.016) .167 0.005 (−0.005, 0.016) .325 0.006 (−0.002, 0.014) .113 0.001 (−0.008, 0.010) .848 0.002 (−0.023, 0.027) .876
 Unhealthy diet 0.009 (−0.014, 0.033) .440 0.010 (−0.015, 0.036) .424 0.005 (−0.011, 0.022) .516 −0.012 (−0.033, 0.009) .275 −0.056 (−0.106, −0.006) .027
 Healthy diet −0.015 (−0.056, 0.025) .454 0.005 (−0.036, 0.045) .822 −0.080 (−0.116, −0.045) .000 0.020 (−0.024, 0.063) .381 −0.012 (−0.156, 0.133) .876
 Fast food 0.005 (−0.005, 0.015) .344 0.002 (−0.009, 0.012) .753 0.017 (0.006, 0.028) .003 −0.012 (−0.024, 0.000) .042 −0.020 (−0.059, 0.019) .313
 Exercise engagement −0.092 (−0.484, 0.300) .645 −0.123 (−0.521, 0.275) .544 0.084 (−0.477, 0.645) .769 −0.019 (−0.536, 0.499) .944 0.812 (−0.777, 2.401) .316
Average lifetime substance use
 BMI −0.018 (−0.111, 0.074) .697 −0.066 (−0.162, 0.030) .179 0.139 (0.055, 0.222) .001 −0.166 (−0.300, −0.032) .015 −0.184 (−0.427, 0.059) .138
 Waist circumference 0.132 (−0.128, 0.391) .320 −0.51 (−0.208, 0.206) .698 0.388 (0.171, 0.604) .000 −0.397 (−0.749, −0.045) .027 −0.449 (−1.246, 0.348) .269
 Hip circumference −0.032 (−0.194, 0.129) .695 −0.111 (−0.286, 0.065) .216 0.188 (0.021, 0.356) .028 −0.206 (−0.474, 0.063) .133 −0.388 (−0.981, 0.204) .199
 Systolic BP 0.058 (−0.100, 0.216) .470 −0.041 (−0.214, 0.131) .638 −0.034 (−0.158, 0.089) .585 0.431 (0.190, 0.672) .000 0.153 (−0.380, 0.687) .573
 Diastolic BP −0.005 (−0.144, 0.134) .946 −0.128 (−0.273, 0.017) .084 0.093 (−0.009, 0.195) .074 0.103 (−0.101, 0.306) .322 0.277 (−0.131, 0.684) .183
 Resting HR −0.135 (−0.331, 0.061) .176 −0.290 (−0.491, −0.089) .005 0.248 (0.126, 0.371) .000 −0.057 (−0.282, 0.168) .621 0.488 (−0.134, 1.109) .124
 FCV 0.010 (−0.007, 0.027) .247 0.007 (−0.011, 0.025) .442 −0.004 (−0.013, 0.005) .411 0.027 (0.010, 0.044) .002 0.002 (−0.029, 0.033) .891
 FEV1 0.002 (−0.013, 0.017) .801 0.003 (−0.012, 0.019) .684 −0.004 (−0.012, 0.003) .254 0.018 (0.003, 0.033) .020 0.013 (−0.012, 0.037) .319
 FCV/FEV1 (%) −0.087 (−0.236, 0.062) .252 −0.012 (−0.158, 0.135) .877 −0.020 (−0.101, 0.061) .626 −0.067 (−0.233, 0.099) .429 0.160 (−0.168, 0.489) .338
 Left hand grip 0.264 (−0.009, .538) .058 0.179 (−0.126, 0.484) .250 0.187 (−0.045, 0.419) .114 0.101 (−0.311, 0.513) .630 −0.732 (−1.625, 0.161) .108
 Right Hand Grip 0.227 (−0.052, 0.507) .111 0.0.082 (−0.240, 0.404) .617 0.239 (−0.026, 0.504) .077 0.166 (−0.263, 0.595) .447 −0.764 (−1.736, 0.207) .123
 Chronic pain 0.027 (−0.002, 0.055) .065 0.005 (−0.025, 0.035) .746 0.035 (0.016, 0.055) .000 −0.006 (−0.042, 0.030) .754 0.014 (−0.089, 0.116) .796
 Gum disease 0.004 (−0.008, 0.015) .515 −0.005 (−0.020, 0.010) .512 0.024 (0.005, 0.042) .014 −0.024 (−0.042, −0.006) .010 0.013 (−0.026, 0.053) .505
 Loss of appetite 0.019 (0.001, 0.037) .039 0.006 (−0.016, 0.027) .616 0.026 (0.007, 0.045) .009 −0.015 (−0.043, 0.013) .289 0.081 (0.003, 0.159) .042
 Nausea 0.007 (−0.013, 0.027) .485 0.000 (−0.022, 0.022) .987 0.019 (0.002, 0.037) .029 −0.014 (−0.044, 0.017) .377 0.024 (−0.038, 0.086) .451
 Weight problems 0.010 (0.000, 0.020) .050 0.001 (−0.011, 0.013) .889 0.018 (0.006, 0.019) .004 −0.009 (−0.025, 0.007) .289 0.025 (−0.015, 0.065) .218
 Problems breathing 0.025 (0.005, 0.046) .015 0.013 (−0.010, 0.035) .267 0.018 (0.003, 0.033) .020 0.002 (−0.025, 0.029) .879 0.058 (−0.014, 0.131) .117
 Skin problems 0.004 (−0.003, 0.011) .312 0.005 (−0.004, 0.013) .279 −0.002 (−0.007, 0.003) .363 0.001 (−0.009, 0.011) .858 0.002 (−0.016, 0.018) .806
 Rapid HR 0.029 (0.024, 0.228) .017 0.013 (−0.013, 0.038) .334 0.023 (0.006, 0.039) .006 0.006 (−0.022, 0.034) .663 0.094 (0.027, 0.161) .006
 Headaches −0.005 (−0.028, 0.018) .659 −0.016 (−0.158, 0.033) .195 0.020 (0.022, 0.205) .015 −0.012 (−0.119, 0.049) .416 −0.027 (−0.125, 0.052) .414
 Injuries 0.009 (−0.006, 0.025) .239 0.005 (−0.011, 0.022) .547 0.008 (−0.003, 0.019) .143 0.005 (−0.013, 0.022) .620 0.004 (−0.041, 0.048) .874
 Unhealthy diet 0.012 (−0.021, 0.045) .478 0.023 (−0.015, 0.061) .243 −0.001 (−0.022, 0.020) .929 −0.005 (−0.047, 0.038) .822 −0.097 (−0.161, −0.034) .003
 Healthy diet −0.017 (−0.085, 0.050) .614 0.021 (−0.056, 0.097) .597 −0.083 (−0.136, −0.030) .002 0.034 (−0.069, 0.138) .513 −0.053 (−0.245, 0.139) .589
 Fast food 0.009 (−0.015, 0.019) .832 −0.002 (−0.020, 0.015) .785 0.022 (0.010, 0.035) .001 −0.037 (−0.060, −0.013) .002 −0.011 (−0.058, 0.036) .636
 Exercise engagement −0.499 (−1.118, 0.121) .115 −0.462 (−1.246, 0.321) .247 0.028 (−0.707, 0.763) .941 −0.110 (−1.122, 0.901) .831 0.055 (−1.300, 1.411) .936
Open in a new tab

Note: All analyses controlled for sex. BMI = Body mass index, BP = blood pressure, HR = heart rate, FCV = forced capacity volume, FEV1 = forced expiratory volume in one second, and exercise engagement = length of exercise in minutes over the past 24 hours. Asterisks indicate significance after correction for multiple testing. Bold indicates significance remained after correction for multiple testing (adjusted p<.05).

Adolescent cannabis frequency.

Adolescent cannabis frequency was only associated with lower adult exercise engagement in the past 24 hours (b=−0.846 minutes, p=.000). Adolescent cannabis frequency was not associated with adult exercise engagement after controlling for tobacco, alcohol, and other drug use frequency.

Young adult cannabis frequency.

Young adult cannabis frequency was not associated with any adult physical health outcomes.

Adult cannabis frequency.

Higher adult cannabis frequency was associated with a lower HR (b=− 0.170, p=.001) and greater frequency of loss of appetite (b=0.018, p=.000). After controlling for tobacco, alcohol, and other drug use frequency, only adult cannabis frequency was associated with a lower resting HR (b=−0.227, p=.000).

Lifetime average cannabis frequency.

Lifetime average cannabis frequency was not associated with adult physical health outcomes.

Tobacco frequency.

Adolescent tobacco frequency was associated with more frequent weight problems and rapid HR. Young adult tobacco frequency was associated with a higher BMI, larger waist circumference, and higher resting HR. In addition, young adult tobacco frequency was also associated with eating fewer fruits and vegetables, eating fast food more frequently, and more frequent loss of appetite. Adult tobacco frequency was associated with a higher resting HR, eating fewer fruits and vegetables, and more chronic pain, problems breathing, and rapid HR. Finally, lifetime average tobacco frequency was associated with a higher BMI, larger waist circumference, higher resting HR, eating fewer fruits and vegetables and eating more fast food.

Alcohol frequency.

Adolescent and adult alcohol frequency was not associated with adult physical health outcomes. Young adult alcohol frequency was associated with eating less fast food. Average lifetime alcohol frequency was associated with a higher systolic BP, lower FCV, and eating less fast food.

Other drug frequency.

Adolescent other drug frequency was associated with more frequent rapid HR. Young adult, adult, and average lifetime other drug use was not associated with adult physical health outcomes.

3.2. Model 2: Within-Family Exposure

Table 4 presents the between-family and within-twin results for the significant phenotypic associations between cannabis frequency and physical health.

Table 4.

Within- and between-family unstandardized regression coefficients for the association between cannabis frequency with physical health

Cannabis Use Substance Use Covariates
Total (95% CI) p-value MZ (95% CI) p-value DZ (95% CI) p-value Between (95% CI) p-value Tobacco Use (95% CI) p-value Alcohol Use (95% CI) p-value Other Drug Use (95% CI) p-value
Adolescent cannabis use
 Exercise engagement −0.249 (–0.933, 0.434) .475 0.094 (−0.237, 0.426) .577 −0.593 (−1.919, 0.734) .381 −1.147 (−1.781, −0.531) .000 - - - - - -
Adult cannabis use
 Resting HR −0.184 (−0.331, −0.037) .014 −0.304 (−0.506, −0.101) .003 −0.064 (−0.277, 0.150) .559 −0.164 (−0.308, −0.020) .025 - - - - - -
 Loss of appetite 0.019 (−0.011, 0.049) .214 0.014 (−0.027, 0.055) .508 0.025 (−0.020, 0.069) .274 0.041 (0.015, 0.066) .002 - - - - - -
Adult cannabis use with covariates
 Resting HR −0.221 (−0.371, −0.071) .004 −0.353 (−0.566, −0.141) .001 −0.084 (−0.294, 0.126) .432 −0.242 (−0.379, −0.105) .001 0.141 (0.053, 0.229) .002 −0.059 (−0.170, 0.052) .298 0.320 (−0.085, 0.724) .122
Open in a new tab

Note: All analyses controlled for sex. HR = heart rate and exercise engagement = length of exercise in minutes over the past 24 hours. † indicates MZ and DZ twins significantly different. Bold indicates p<.05.

Adolescent cannabis frequency.

Only at the between-family level, an increase of one-day/month of cannabis frequency during adolescence is associated with 1.147 fewer minutes of exercise in the past 24 hours during adulthood (p=.000). Effects were not significant within-twin pairs or among only MZ and DZ twins.

Young adult cannabis frequency.

We did not conduct any follow-up analyses because there were no significant phenotypic associations between young adult cannabis frequency and physical health outcomes.

Adult cannabis frequency.

An increase of one-day/month of adult cannabis use was associated with more frequent appetite loss (b=0.041, p=.002) at the between-family level. There were no significant effects between adult cannabis frequency and appetite loss within-twin pairs and among only MZ or DZ twins. An increase of one-day/month of adult cannabis use was associated with lower a resting HR at the total within-twin level (b=−0.184 bpm, p=.014), among only MZ twins (b=−0.304 bpm, p=.003), and at the between-family level (b=−0.164 bpm, p=.025). These effects were not significant among only DZ twins.

Average lifetime cannabis frequency.

We did not conduct any follow-up analyses because there were no significant phenotypic associations between average lifetime cannabis frequency and adult physical health outcomes.

Tobacco frequency.

Adult tobacco frequency was a covariate in the multilevel analyses involving adult cannabis frequency and resting HR. Adult tobacco frequency was associated with a higher resting HR within-twin pairs (b=0.141, p=.002).

Alcohol frequency.

Adult alcohol frequency was not associated with resting HR.

Other drug use frequency.

Adult other drug use frequency was not associated with resting HR.

3.3. Biometric Models

The correlations between the genetic and environmental factors underlying cannabis use and health outcomes are displayed in Supplemental Table 4. In these AE models, the genetic correlations (rG) represent the correlation among familial factors for cannabis use and physical health outcomes. Biometrical models were conducted to follow-up the co-twin control analyses to test whether genetic or environmental risk underlie within-family effects between adolescent cannabis use and length of exercise and adult cannabis use and resting HR and loss of appetite frequency. There was a negative genetic correlation between adolescent cannabis use and adult exercise engagement (rG=−0.68) consistent with the significant between-family effects between these variables. A negative environmental correlation between adult cannabis use and resting HR (rE = −0.29) suggested that unshared environmental factors associated with increased cannabis use and are associated with decreased appetite loss. This effect is consistent with the within-twin pair effects. No other tested genetic or environmental correlations were significant.

We did not find a significant genetic correlation between adult cannabis use with loss of appetite frequency (rG=0.21) and resting HR (rG=0.03), which is inconsistent with the significant between-family effects of these variables. The differences in results (significant between-family effects and non-significant genetic correlations) are likely due to how the co-twin control and biometric analyses were conducted. The co-twin control analyses included covariates (alcohol, tobacco, and other drug use) and the cannabis frequency variable was continuous (0–30 days). However, the biometric models did not include any covariates and the cannabis frequency variable was transformed into an ordinal variable (0, 1–10, >10 days).

4. DISCUSSION

This study estimated the effect of cannabis frequency across adolescence to adulthood on adult physical health outcomes among a sample who primarily use cannabis casually with a co-twin control design. Adolescent cannabis use was associated with less exercise engagement in adulthood at the phenotypic and between-family level. Adult cannabis use was associated with a lower HR and more frequent appetite loss at the phenotypic level and between-family level. At the total within-twin level and among MZ twin pairs, adult cannabis frequency was associated with a lower resting HR. The within-twin pair effects between adult cannabis use and resting HR were supported by a significant negative nonshared environmental correlation. Thus, within-family comparisons provided some support for cannabis use being associated with a lower resting HR, after accounting for familial confounds. Contrary to cannabis use, tobacco use was associated with poorer physical health outcomes, consistent with prior research on the health effects of tobacco use (Shavelle et al., 2008; Sherman, 1991).

Our findings that cannabis use in adulthood was associated with lower HR, but not BP, reveal new information about associations between cannabis and cardiovascular function. Our results are inconsistent with previous research reporting increases in HR acutely (Menkes et al., 1991) and long-term (Schmid et al., 2010). Our results may differ from prior studies because prevalence of obesity in this sample (21%) (which closely resembles the state of Colorado prevalence of obesity (22%); Centers for Disease Control and Prevention, 2020) is much lower than the national prevalence rates, which are around 40% in this age range. This sample is likely more health-conscious than the national average and a lower BMI contributes to a lower resting HR (e.g., Stolarz et al., 2003). We did not find associations between cannabis frequency with BP. Previous research has reported higher systolic BP among those who use cannabis compared to non-using controls (Alshaarawy & Elbaz, 2016) while negative associations have been found when examining dose-dependent associations (Meier et al., 2016; Meier et al., 2019).

Although this study has notable strengths, there are several limitations worth considering. Since the sample was primarily White, non-Hispanic/Latinx, results may not generalize to ethnically/racially diverse individuals. However, similar results have been demonstrated among more ethnic and racially diverse individuals (e.g.,Ngueta et al., 2015; Ross et al., 2020). Prevalence of obesity in this sample (21%) closely resembles the state of Colorado prevalence of obesity (22%; Centers for Disease Control and Prevention, 2020) while the national prevalence rates are around 40% in this age range. Although this sample may be more health-conscious than the national average, similar results have been demonstrated between cannabis use and BMI in studies conducted in other states (Alshaarawy & Anthony, 2015; Danielsson et al., 2016; Gerberich et al., 2003; Hayatbakhsh et al., 2010; Meier et al., 2016; Meier et al., 2019; Ngueta et al., 2015; Penner et al., 2013; Ross et al., 2020). In addition, 2%, 7%, and 9% of adolescent, young adult, and adult participants, respectively, endorsed daily cannabis use, thus results may not generalize to individuals who use cannabis daily. The study did not collect detailed information on potency of cannabis products used or methods of cannabis ingestion. Thus, we were not able to examine whether different potency or methods of cannabis consumption affect health differentially. The participant age range is 28–35; therefore, most participants have not developed more serious complications from cannabis, tobacco, alcohol, or other drug use, although negative effects of substance use on health has been documented in early adulthood (Jha et al., 2013; Meier et al., 2016; Taylor et al., 2002). However, this may be the reason for the small effects observed in the associations between cannabis and tobacco use with physical health.

This study found no evidence of associations between young adult and lifetime average cannabis frequency and worse adult physical health among a normative sample of individuals who use cannabis. We found phenotypic associations between adolescent cannabis frequency with lower exercise engagement in adulthood as well as between adult cannabis frequency and more frequent appetite loss. However, these effects were only present between-families suggesting that familial confounding may contribute to these associations. Specifically, it may be that those who are genetically predisposed to use cannabis are also predisposed to exercise less and lose their appetite more frequently. However, results also suggest that adult cannabis frequency is associated with a lower resting HR, which is consistent with a causal association between cannabis use and resting HR. Results contrast markedly with those for tobacco use, which was consistently associated with worse physical health. In general, these results do not support a causal association between using cannabis once a week (the mean cannabis frequency of the sample in adulthood) and detrimental physical health effects of individuals aged 20–35.

Supplementary Material

1

Highlights.

  • We observed phenotypic associations between cannabis and exercise, heart rate, and appetite loss

  • Families with greater cannabis use exercised less and had more frequent appetite loss

  • Within-twin pair differences in cannabis use were associated with a lower heart rate

  • Results suggest that cannabis use is not associated with detrimental physical health effects

Acknowledgements

This work was supported by the National Institutes of Health grants DA017637, AG046938, AA026635, DA032555, and DA042755. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. The authors have no other conflicts of interest to declare.

Role of Funding Source

This work was supported by the National Institutes of Health grants DA054212, DA017637, AG046938, AA026635, DA032555, and DA042755. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Conflict of Interest Declaration

None.

References

  1. Alshaarawy O, Anthony JC 2015. Cannabis smoking and diabetes mellitus: Results from meta-analysis with eight independent replication samples. Epidemiology. 26, 597. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Alshaarawy O, Elbaz HA 2016. Cannabis use and blood pressure levels: United States National Health and Nutrition Examination Survey, 2005–2012. J Hypertens. 34, 1507. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Bachs L, Mørland H 2001. Acute cardiovascular fatalities following cannabis use. Forensic Sci Int. 124, 200–203. [DOI] [PubMed] [Google Scholar]
  4. Barry D, Petry NM 2009. Associations between body mass index and substance use disorders differ by gender: Results from the National Epidemiologic Survey on Alcohol and Related Conditions. Addict Behav. 3, 51–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Carlin JB, Gurrin LC, Sterne JA, Morley R, Dwyer T 2005. Regression models for twin studies: A critical review. Int J Epidemiol. 34, 1089–1099. [DOI] [PubMed] [Google Scholar]
  6. Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Division of Population Health. 2020. Behavioral Risk Factor Survellience System Prevalence & Trends Data, 2019.
  7. Chantala K, Tabor J 1999. National Longitudinal Study of Adolescent Health: Strategies to perform a design-based analysis using the Add Health data.
  8. Corley RP, Reynolds CA, Wadsworth SJ, Rhea SA, Hewitt JK 2019. The Colorado twin registry: 2019 update. Twin Res Hum Genet. 22, 707–715. [DOI] [PubMed] [Google Scholar]
  9. Danielsson A, Lundin A, Yaregal A, Östenson C, Allebeck P, Agardh E 2016. Cannabis use as risk or protection for type 2 diabetes: A longitudinal study of 18,000 Swedish men and women. J Diabetes Res. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Ellingson JM, Ross JM, Winiger E, Stallings MC, Corley RP, Friedman NP, Hewitt JH, Tapert SF, Brown SA, Wall TL, Hopfer CJ (2021). Familial factors may not explain the effect of moderate‐to‐heavy cannabis use on cognitive functioning in adolescents: a sibling‐comparison study. Addiction. 116, 833–844. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Gerberich SG, Sidney S, Braun BL, Tekawa IS, Tolan KK, Quesenberry CP 2003. Marijuana use and injury events resulting in hospitalization. Ann Epidemiol. 13, 230–237. [DOI] [PubMed] [Google Scholar]
  12. Hamilton CM, Strader LC, Pratt JG, Maiese D, Hendershot T, Kwok RK 2011. The PhenX Toolkit: Get the most from your measures. Am J Epidemiol. 174, 253–260. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Hancox RJ, Poulton R, Ely M, Welch D, Taylor DR, McLachlan CR, Green JM, Moffitt TE, Caspi A, Sears MR 2010. Effects of cannabis on lung function: A population-based cohort study. European Respiratory Journal. 35, 42–47. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Hayatbakhsh MR, O’Callaghan MJ, Mamun AA, Williams GM, Clavarino A, Najman JM 2010. Cannabis use and obesity and young adults. Am J Drug Alcohol Abuse. 36, 350–356. [DOI] [PubMed] [Google Scholar]
  15. Hochberg Y 1988. A sharper Bonferroni procedure for multiple tests of significance. Biometrika. 75, 800–802. [Google Scholar]
  16. National Cancer Institute. 2005. Five-factor screener: National Health Interview Survey (NHIS) diet and nutrition. 100–138. [Google Scholar]
  17. Jackson NJ, Isen JD, Khoddam R, Irons D, Tuvblad C, Iacono WG, McGue M, Raine A, Baker LA 2016. Impact of adolescent marijuana use on intelligence: Results from two longitudinal twin studies. Proc Natl Acad Sci, 113, E500–E508. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Jha P, Ramasundarahettige C, Landsman V, Rostron B, Thun M, Anderson RN, MacAfee T, Peto R 2013. 21st-century hazards of smoking and benefits of cessation in the United States. N Engl J Med. 368, 341–50. [DOI] [PubMed] [Google Scholar]
  19. Jin LZ, Rangan A, Mehlsen J, Andersen LB, Larsen SC, Heitmann BL 2017. Association between use of cannabis in adolescence and weight change into midlife. PloS one. 12, e0168897. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Levendal R, Schumann D, Donath M, Frost C 2012. Cannabis exposure associated with weight reduction and β-cell protection in an obese rat model. Phytomedicine. 19, 575–582. [DOI] [PubMed] [Google Scholar]
  21. Liemburg EJ, Bruins J, van Beveren N, Islam MA, Alizadeh BZ, Bruggeman R, Wiersma D, Cahn W, Kahn R, de Haan L 2016. Cannabis and a lower BMI in psychosis: What is the role of AKT1? Schizophr Res. 176, 95–99. [DOI] [PubMed] [Google Scholar]
  22. Lyons MJ, Bar J, Panizzon M, Toomey R, Eisen S, Xian H, Tsuang M 2004. Neuropsychological consequences of regular marijuana use: a twin study. Psych Med, 34, 1239–1250. [DOI] [PubMed] [Google Scholar]
  23. McGue M, Osler M, Christensen K 2010. Causal inference and observational research: The utility of twins. Perspectives on Psychological Science. 5, 546–556. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Meier MH, Caspi A, Ambler A, Harrington H, Houts R, Keefe RS, McDonald K, Ward A, Poulton R, Moffitt TE 2012. Persistent cannabis users show neuropsychological decline from childhood to midlife. Proc Natl Acad Sci, 109, E2657–E2664. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Meier MH, Caspi A, Cerdá M, Hancox RJ, Harrington H, Houts R, Poulton R, Ramrakha S, Thomson WM, Moffitt TE 2016. Associations between cannabis use and physical health problems in early midlife: A longitudinal comparison of persistent cannabis vs tobacco users. JAMA Psychiatry. 73, 731–740. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Meier MH, Caspi A, Danese A, Fisher HL, Houts R, Arseneault L, Moffitt TE 2017. Associations between adolescent cannabis use and neuropsychological decline: A longitudinal co‐twin control study. Addiction. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Meier MH, Pardini D, Beardslee J, Matthews KA 2019. Associations between cannabis use and cardiometabolic risk factors: A longitudinal study of men. Psychosom Med. 81, 281–288 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Menkes DB, Howard RC, Spears GF, Cairns ER 1991. Salivary THC following cannabis smoking correlates with subjective intoxication and heart rate. Psychopharmacology. 103, 277–279. [DOI] [PubMed] [Google Scholar]
  29. Mittleman MA, Lewis RA, Maclure M, Sherwood JB, Muller JE 2001. Triggering myocardial infarction by marijuana. Circulation. 103, 2805–2809. [DOI] [PubMed] [Google Scholar]
  30. Moore BA, Augustson EM, Moser RP, Budney AJ 2005. Respiratory effects of marijuana and tobacco use in a US sample. J Gen Intern Med. 20, 33–37. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Muniyappa R, Sable S, Ouwerkerk R, Mari A, Gharib AM, Walter M, Courville A, Hall G, Chen KY, Volkow ND 2013. Metabolic effects of chronic cannabis smoking. Diabetes Care. 36, 2415–2422. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Muthén LK, Muthen B 2017. Mplus User’s Guide: Statistical Analysis with Latent Variables.
  33. Neuhaus JM, McCulloch CE 2006. Separating between‐and within‐cluster covariate effects by using conditional and partitioning methods. Journal of the Royal Statistical Society: Series B (Statistical Methodology). 68, 859–872. [Google Scholar]
  34. Nezlek JB Multilevel modeling for psychologists. In: Cooper DL, Long AT, Panter D, Rindskopf, Sher KJ 2012. APA handbook of research methods in psychology, Vol 3 Data analysis and research publications. Washington, DC, US: American Psychological Association. 219–241. [Google Scholar]
  35. Ngueta G, Bélanger RE, Laouan‐Sidi EA, Lucas M 2015. Cannabis use in relation to obesity and insulin resistance in the inuit population. Obesity. 23, 290–295. [DOI] [PubMed] [Google Scholar]
  36. Owen KP, Sutter ME, Albertson TE 2014. Marijuana: Respiratory tract effects. Clin Rev Allergy Immunol. 46, 65–81. [DOI] [PubMed] [Google Scholar]
  37. Penner EA, Buettner H, Mittleman MA 2013. The impact of marijuana use on glucose, insulin, and insulin resistance among US adults. Am J Med. 126, 583–589. [DOI] [PubMed] [Google Scholar]
  38. Pletcher MJ, Vittinghoff E, Kalhan R, Richman J, Safford M, Sidney S, Lin F, & Kertesz S 2012. Association between marijuana exposure and pulmonary function over 20 years. JAMA. 307, 173–181. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Rhea SA, Gross AA, Haberstick BC, Corley RP (2006). Colorado twin registry. Twin Res Hum Genet. 9, 941–949. [DOI] [PubMed] [Google Scholar]
  40. Rhea SA, Gross AA, Haberstick BC, Corley RP (2013). Colorado twin registry: An update. Twin Res Hum Genet. 16, 351–357. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Rodondi N, Pletcher MJ, Liu K, Hulley SB, Sidney S 2006. Marijuana use, diet, body mass index, and cardiovascular risk factors (from the CARDIA study). Am J Cardiol. 98, 478–484. [DOI] [PubMed] [Google Scholar]
  42. Rooke SE, Norberg MM, Copeland J, Swift W 2013. Health outcomes associated with long-term regular cannabis and tobacco smoking. Addict Behav. 38, 2207–2213. [DOI] [PubMed] [Google Scholar]
  43. Ross JM, Ellingson JM, Rhee SH, Hewitt JK, Corley RP, Lessem JM, Friedman NP 2020. Investigating the causal effect of cannabis use on cognitive function with a quasi-experimental co-twin design. Drug Alcohol Depend, 206, 107712. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Ross JM, Graziano P, Pacheco-Colón I, Coxe S, Gonzalez R 2016. Decision-making does not moderate the association between cannabis use and body mass index among adolescent cannabis users. J Int Neuropsychol Soc. 22, 944–949. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Ross JM, Pacheco-Colón I, Hawes SW, Gonzalez R 2020. Bidirectional longitudinal associations between cannabis use and body mass index among adolescents. Cannabis and Cannabinoid Research. 5, 81–88. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Salomonsen-Sautel S, Sakai JT, Thurstone C, Corley RP, Hopfer C 2012. Medical marijuana use among adolescents in substance abuse treatment. J Am Acad Child Adolesc Psychiatry. 51, 694–702. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Schmid K, Schönlebe J, Drexler H, Mueck-Weymann M 2010. The effects of cannabis on heart rate variability and well-being in young men. Pharmacopsychiatry. 43,147–150. [DOI] [PubMed] [Google Scholar]
  48. Scott JC, Slomiak ST, Jones JD, Rosen AF, Moore TM, Gur RC 2018. Association of cannabis with cognitive functioning in adolescents and young adults: A systematic review and meta-analysis. JAMA Psychiatry, 75, 585–595. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Scott JC, Slomiak ST, Jones JD, Rosen AF, Moore TM, Gur RC 2018. Association of cannabis with cognitive functioning in adolescents and young adults: A systematic review and meta-analysis. JAMA Psychiatry, 75, 585–595. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Sethi JM, Rochester CL 2000. Smoking and chronic obstructive pulmonary disease. Clin Chest Med. 21, 67–86. [DOI] [PubMed] [Google Scholar]
  51. Shavelle RM, Paculdo DR, Strauss DJ, & Kush SJ 2008. Smoking habit and mortality: A meta-analysis. J Insur Med. 40, 170–178. [PubMed] [Google Scholar]
  52. Sherman CB 1991. Health effects of cigarette smoking. Clinics in Chest Medicine. 12, 643–658. [PubMed] [Google Scholar]
  53. Simmons M, Tashkin D 1995. The relationship of tobacco and marijuana smoking characteristics. Life Sci, 56, 2185–2191. [DOI] [PubMed] [Google Scholar]
  54. Stolarz K, Staessen JA, Kuznetsova T, Tikhonoff V, Babeanu S, Casiglia E, Fagard RH, Kawecka-Jaszcz K, Nikitin Y, et al. 2003. Host and environmental determinants of heart rate and heart rate variability in four European populations. J Hypertens. 21, 525–535. [DOI] [PubMed] [Google Scholar]
  55. Svedberg P, Gatz M, Lichtenstein P, Sandin S, Pedersen NL 2005. Self-rated health in a longitudinal perspective: A 9-year follow-up twin study. J Gerontol B Psychol Sci Soc Sci. 60, S331–S340. [DOI] [PubMed] [Google Scholar]
  56. Taylor DR, Fergusson DM, Milne BJ, Horwood LJ, Moffitt TE, Sears MR, Poulton R 2002. A longitudinal study of the effects of tobacco and cannabis exposure on lung function in young adults. Addiction. 97, 1055–1061. [DOI] [PubMed] [Google Scholar]
  57. Tetrault JM, Crothers K, Moore BA, Mehra R, Concato J, Fiellin DA 2007. Effects of marijuana smoking on pulmonary function and respiratory complications: A systematic review. Arch Intern Med. 167, 221–228. [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Wadsworth SJ, Corley RP, Munoz E, Trubenstein BP, Knaap E, DeFries JC, Plomin R,Reynolds CA 2019. CATSLife: A study of lifespan behavioral development and cognitive functioning. Twin Res Hum Genet. 22, 695–706. [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Waterreus A, Di Prinzio P, Watts GF, Castle D, Galletly C, Morgan VA 2016. Metabolic syndrome in people with a psychotic illness: Is cannabis protective? Psychol Med. 46, 1651–1662. [DOI] [PubMed] [Google Scholar]
  60. Yankey BN, Strasser S, Okosun IS 2016. A cross-sectional analysis of the association between marijuana and cigarette smoking with metabolic syndrome among adults in the United States. Diabetes Metab Syndr. 10, S89–S95. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS1761784-supplement-1.docx (59.3KB, docx)

RESOURCES