Marijuana and Sperm Quality

The honest answer is that the two largest studies disagree, and anyone telling you otherwise is picking one. In a cohort of 1,215 healthy young Danish men, using cannabis more than once a week was associated with a 28% lower sperm concentration versus no use (95% CI -48 to -1) in that same cohort. In a separate cohort of 662 men attending a fertility clinic in Boston, men who had ever smoked cannabis had a significantly HIGHER sperm concentration than men who never had — 62.7 million per mL versus 45.4 million per mL, P = 0.0003.

Both papers are real, peer-reviewed and uncorrected. Neither is fringe. What they share is a measurement problem: every study in this field asks men whether they smoke, and almost none measures how much.

The finding that holds up most consistently across the literature is not count at all. It is morphology — the proportion of sperm with a normal shape.

The populations were not the same, and that is most of it.

• Gundersen studied a cohort of 1,215 healthy young men aged 18 to 28 recruited at a compulsory medical examination for military service — a general-population sample, with no fertility problem selecting them in. In that cohort of 1,215, 45% had used cannabis in the preceding 3 months.
• Nassan studied 662 men already attending a fertility centre, so every man in the cohort was there because conception was not happening. The authors state directly that their results may not be generalisable to men in the general population.
• Gundersen's intervals barely exclude no effect. In that cohort the concentration estimate was -28% versus non-users with a 95% CI of -48 to -1, and the count estimate -29% versus non-users with a CI of -46 to -1. An upper bound of -1 means the result is statistically present and fragile at the same time.
• Nassan's authors flag their own risk of a chance finding: the paper made multiple comparisons, and they say chance cannot be excluded. They also raise the possibility that the result reflects the endocannabinoid system's real role in sperm production rather than a spurious signal, and say it needs confirmation.

There is a further explanation nobody has ruled out: men who smoke cannabis may differ from men who do not in ways that have nothing to do with cannabis — age, risk tolerance, other drug use, and in a fertility clinic, the reason they were referred. None of the retrieved studies resolves that. It is an open question, not a settled one in either direction.

In a cohort of 409 men presenting for infertility evaluation at a single institution, 174 of 409 (43%) reported cannabis use. Current users in that cohort had more than double the odds of abnormal strict morphology versus never-users (OR 2.15, 95% CI 1.21-3.79) and of a semen volume below the WHO reference value (OR 2.76, 95% CI 1.19-6.42).

The same cohort produced something that gets left out of most summaries, including the previous version of this page: the odds of below-reference motility were LOWER in current users versus never-users, not higher (OR 0.47, 95% CI 0.25-0.91). The authors describe the effect as potentially detrimental for morphology and volume but possibly protective against abnormal motility. Note that this study used the WHO 2010 fifth-edition criteria, which the 2021 sixth edition has since replaced.

Payne 2019, a systematic review in the Journal of Urology, reaches the same shape of conclusion across the literature: the strongest evidence of cannabis-related change is in semen parameters, including abnormalities in morphology, with hormonal effects inconclusive. Its own caution is that the most striking results come from animal models and have not been replicated in humans.

Sperm carry cannabinoid receptors. The endocannabinoid system — the signalling network that THC binds into — is present in the testis and in the sperm cell itself, and it has a role in normal sperm production and in the changes a sperm cell goes through before it can fertilise an egg. THC is a foreign molecule entering a system that already uses molecules of that family as signals.

That is why the direction of effect is genuinely hard to predict, and why Nassan's authors raise the endocannabinoid system as a possible explanation for a result that points the unexpected way. Payne 2019 summarises the biological evidence as supporting a role for cannabis in reducing count and concentration, inducing morphology abnormalities, reducing motility and viability, and inhibiting the sperm's fertilising capacity — but most of that strength comes from laboratory and animal work, not from men.

• No dose. Not one retrieved study measured how much THC a man consumed. Gundersen used three self-reported frequency bands; Nassan and Hehemann used ever, current and past. Any page telling you that a specific amount is safe, or that more is proportionally worse, is going beyond the data.
• No reversal. No retrieved study tested stopping. Nassan found no significant difference between current and past smokers for concentration (P = 0.60), which is evidence against a simple recovery story as much as for one. The previous version of this page asserted partial or full recovery after stopping; that claim had no source and has been removed.
• No CBD evidence in men. No human study of CBD alone against a semen parameter was located. Claims that CBD protects sperm from THC are not supported by anything retrievable.
• No pregnancy or live-birth outcome. Every study here measured semen parameters or hormones. None followed couples to a pregnancy. A change in a number on a report is not the same as a change in your chance of conceiving.
• No causation. All three human studies are observational and based on self-report, which in a country where cannabis is illegal is a particular problem — men under-report, and under-reporting does not bias results in a predictable direction.
• Nothing about partner effects. The previous version had an FAQ on whether cannabis affects a partner's fertility. No source here addresses that.

Cannabis is a controlled substance in India. The Narcotic Drugs and Psychotropic Substances Act 1985 defines “cannabis (hemp)” at section 2(iii) as charas — the separated resin — and ganja, meaning the flowering or fruiting tops of the plant, expressly excluding the seeds and leaves when they are not accompanied by the tops, together with any mixture or drink prepared from either. Section 27 makes consumption of a narcotic drug punishable; for substances in this category the Act specifies imprisonment which may extend to six months, or a fine which may extend to ten thousand rupees, or both.

That is the law, not a judgement. It matters here for a practical reason: it shapes the data. Almost all of the research above comes from Denmark and the United States, where men could answer a questionnaire about cannabis without legal risk. No Indian dataset on cannabis and semen parameters was found. Any figure you read about Indian men on this subject should be treated with suspicion until the study is named.

It also matters for the conversation with your doctor. NICE's 2026 fertility guideline (NG257, recommendation 1.13.1) tells clinicians to ask people concerned about their fertility about recreational drugs, naming cannabis specifically, and to give advice on the potential impact. A fertility clinician asking is asking clinically.

The defensible position is not that cannabis is proven to harm your fertility. It is that the evidence is mixed, the one consistent signal is on morphology, the exposure has never been measured properly, and stopping has never been tested. If you want to remove a variable while you are trying, removing this one is reasonable on that basis — not because a study has shown it works.

Get an actual measurement before changing your life around a guess. A semen analysis tells you where you stand; Semen Analysis: What It Reveals About Fertility explains what is in the report. If a first result is abnormal, NICE NG257 (1.17.3 and 1.17.4) advises repeating it, ideally about three months later, so a full cycle of sperm formation has completed.

Three months is the real timetable for anything affecting sperm production. When 11 men with normal sperm concentrations were given deuterated water and tracked, newly labelled sperm appeared in the ejaculate after a mean of 64 plus or minus 8 days, with a range of 42 to 76 (Misell 2006, Journal of Urology). A repeat test two weeks after a change measures sample-to-sample variation, not the change.

The better-established lifestyle factors are worth dealing with in the same window. NICE NG257 records an association between smoking and reduced semen quality (1.7.3), states that excessive alcohol is detrimental to semen quality (1.6.2), and advises that a BMI of 30 or over carries an increased risk of reduced fertility (1.9.3). See Does Smoking Lower Sperm Count? and Does Alcohol Reduce Sperm Count?.

And if you are not conceiving, the answer is often not IVF. Assessment of both partners, correcting timing, or a simpler treatment resolves a great many cases. NICE NG257 (1.4.1) records a cumulative figure: over 80% of heterosexual couples in the general population conceive within a year where the woman is under 40 and intercourse is regular and unprotected — a cumulative rate over 90% by the end of the second year.