Table 6.

Epidemiologic studies on marijuana use and other cancers

Study location, period, author, referenceCancer siteCharacteristics of casesCharacteristic of controls or cohortExposure assessmentExposure categoriesRR or OR (95% CI)Adjustment for potential confounders and other notes
California, US
  1979–1999,
  Sidney et al.
  (37)
All sites, and
  selected sites
1,421 cancersCohort of
  64,855 Kaiser
  Permanente subscribers who
  received health
  checkups, ages
 15 to 49 years,
  follow-up through
  cancer registry
  and death records
Self-administered
  questionnaires
Ever and current
use
All sites
Tobacco-related
  cancer
Colorectal cancer
Melanoma
Prostate cancer




All sites
Tobacco-related
  cancer1
Colorectal cancer
Melanoma Breast
cancer Cervical
cancer
Men, nontobacco
smokers
0.8 (0.5–1.2)
0.8 (0.2–2.9)

0.7 (0.2–2.1)
0.5 (0.2–1.3)
3.1 (1.0–9.5)
Women,
non-tobacco
smokers
1.1 (0.8–1.3)


0.3 (0.0–2.5)
1.0 (0.4–2.3)
0.8 (0.5–1.3)
1.4 (1.0–2.1)
  • Adjusted for age, race, education, and alcohol use

  • Estimates shown are for nontobacco smokers

  • Definition of ever use was ≥6 times use over lifetime

  • Dose–response relations for duration (years) and frequency (times/week or month) were not observed

Washington, US,
  and Canada,
  1978–1985, Daling
  et al. (38)
Anal148 cases identified
  through the cancer
  registry, <70
  years of age,
  of all histologic
  types, including
in situ and invasive
  lesions. Interviews
  conducted for 71.2%
  of eligible cases
  identified.
166 colon cancer
  cases identified through
  the cancer registry,
  individually matched
  on age, sex,
  year of diagnosis,
  and geographic
  area. Interviews available
  for 67.3% of
  eligible subjects.
Face-to-face
  interviews with
  questionnaire
Ever use0.8 (0.2–4.0)
  • Adjusted for age, residence, cigarette smoking (never, formerly, currently), geographic area

  • Dose–response relations not assessed

Washington, US,
  and Canada, 1979–1990,
  Maden et al.
  (39)
Penile110 cases, identified
  through the cancer
  registry, ≤74 years old,
  including squamous
  carcinoma and
in situ Response rate: 50.2%
355 subjects from
  random digit dialing,
  frequency matched on age,
  reference year,
Response rate: 70.3%
Face-to-face
 interviews with
  questionnaire
Ever use

Frequency Never
≤50 times
>50 times
1.5 (0.7–2.3)


1.0
1.7 (0.8–3.9)
1.0 (0.3–3.6)
  • Adjusted for age, alcohol consumption, cigarette smoking (never, former, current), number of sexual partners

  • Adjusted for age, alcohol consumption, number of sexual partners

California, US, 1989–1992,
  Nelson et al.
  (40)
Non-Hodgkin
  lymphoma
378 cases identified
  through the
  cancer registry,
  18–75 years old,
  residents of
 Los Angeles, English/Spanish
  speaker, HIV seronegative.
  Overall % interviewed
  of NHL cases
  ∼36.7%.
378 subjects individually
  matched on age, sex,
  race/ethnicity,
  neighborhood of residence,
  and interview language
Face-to-face
  interviews
  with questionnaire
Lifetime use (men)
No use
Any use
  1–5 times
  6–900 times
  ≥ 901 times

1.00
0.86 (0.50–1.48)
0.68 (0.34–1.38)
0.93 (0.46–1.88)
1.09 (0.48–2.48)
  • Adjusted for age, sex, race/ethnicity, neighborhood of residence, and interview language

California, US,
  1988–1995,
  Holly et al. (41)
Non-Hodgkin
 lymphoma
1,281 cases identified
  through the Northern
  California cancer registry,
  ages 21 to 74.
  Overall % interviewed
  of NHL cases
  ∼56.7%.
2,095 subjects
  from random digit
  dialing, frequency matched
  on age, sex,
  residence. 78% of
  eligible controls
  completed interviews
Face-to-face
 interviews with
  structured questionnaire
Number of times used
Never
<40
40–999
≥1000
Number of times used
Never
<40
40–999
≥1,000
Women

1.00
0.56 (0.40–0.77)
0.58 (0.35–0.97)
0.71 (0.34–1.5)
Men

1.00
0.64 (0.49–0.84)
0.52 (0.37–0.73)
0.49 (0.31–0.78)
  • Adjusted for age

  • Estimates adjusted for age, education, sexual partners, vaccinations, medications, and other factors were significant for men, and for men and women combined

California, US,
  1977–1999,
  Efird et al.
  (42)
Malignant
 primary glioma
69 cases
  of glioma
Cohort of 105,005
  Kaiser Permanente
  subscribers who
  received health
  checkup, ages ≥25 years,
  follow-up through
  cancer registry
Self-administered
  questionnaires
Ever use

Frequency
  <1/month
  ≥1/month
Unknown
P for trend
1.9 (0.9–4.0)


0.6 (0.1–4.4)
2.8 (1.3–6.2)
1.3 (0.8–2.2) 0.08
  • Adjusted for smoking status (cigarettes, cigars, pipes), sex, race, alcohol, education, coffee intake

US, Chacko
et al. (43)
Transitional cell
carcinoma
of bladder
52 cases age
<60 with transitional
cell carcinoma of
bladder from hospitals.
Response rate
= 88.5%
168 age-matched
controls Response
rate = 69.2%
A self-administered
questionnaire
Marijuana
Ever smoked
Current smoke
Joint-years>40

3.4
1.9
3.5
  • Adjusted for agent orange exposure, radiation exposure, and dye exposure

  • Not adjusted for tobacco smoking

  • CIs were not reported in the publication, though the P for trend for increasing joint-years of marijuana was reported (P = 0.01)

Chicago, IL,
  Baltimore, MD,
  Washington, DC, Pittsburgh,
  PA, and Los Angeles,
  CA, 1984–2003,
  Chao et al.
  (44)
Kaposi sarcomaCohort of 1,335
  white men
  who have sex
  with men who
  were HIV positive
  or seroconverted
  before 2003,
  401 Kaposi
  sarcoma cases
  Follow-up rate: N/A
Interviewer-administered
  questionnaire
Ever marijuana use
Frequency
None
Monthly or less
Weekly or more
1.25 (0.87–1.79)


1.00 1.15 (0.77–1.70)
1.52 (0.99–2.32)
  • Adjusted for age, education, study center, alcohol use, tobacco smoking, number of male sexual partners, lifetime number of sexual partners, receptive anal intercourse and condom use, antiretroviral therapy, CD4 cell count, and sexually transmitted infection

  • Five year lag was applied for exposure

  • Duration of use was not available

Abbreviation: NHL, non-Hodgkin lymphoma.