When the federal government introduced Bill C-16 in December 2025, Justice Minister Sean Fraser called it “one of the most consequential reforms of the Criminal Code in a generation.”¹ Buried within its 200-plus pages is a provision that could fundamentally reshape how sexual assault trials are conducted in Canada; and it is receiving far less attention than it deserves.
Among the proposed amendments is one that would change the threshold for defence access to a complainant’s therapeutic records. Under the current regime, the defence must establish that the records are “likely relevant” to an issue at trial. Bill C-16 would replace that standard with a requirement that the defence establish the records contain evidence that “could raise a reasonable doubt as to the accused’s guilt.” The shift is not merely procedural. It is a quiet but significant reversal of the presumption of innocence.
Where Things Stand
Bill C-16 is currently before the House of Commons Standing Committee on Justice and Human Rights, having passed first reading in December 2025 and second reading on February 2, 2026. At this pace, it could become law by summer 2026.
The timeline is remarkably fast for legislation of this magnitude. The bill was introduced as a response to the Federal Ombudsperson’s report Rethinking Justice for Survivors of Sexual Violence: A Systemic Investigation. Once enacted, these provisions will govern cases for years before appellate courts clarify their constitutional limits.
The time to get the language right is now – at committee – not after the first wrongful conviction it enables.
The Existing Regime and What It Already Does
Sections 278.1 to 278.91 of the Criminal Code codify the records regime and represent Parliament’s considered attempt to balance the privacy interests of complainants against an accused’s constitutionally protected right to make full answer and defence.
Under the current framework, accessing records, including therapeutic records, engages a two-stage process. At the first stage, defence counsel must demonstrate that the records are “likely relevant” to an issue at trial. If that threshold is met, the records are produced to the judge ( not to the defence ) for a second-stage review. At the second stage, the judge examines the records and determines whether, and to what extent, they should be disclosed to the accused, weighing the right to full answer and defence against the complainant’s right to privacy and the reliability of the evidence.
The framework emerged from R. v. O’Connor in 1995, where the Supreme Court of Canada first grappled with defence access to private records in sexual assault proceedings. Parliament codified and refined that framework following R. v. Mills in 1999, which upheld the legislative scheme as a constitutionally sound balancing of competing rights.
The current regime is not a permissive free-for-all.
Section 278.3(4) provides that any one of the following assertions – standing alone – is insufficient to establish likely relevance:
- that the record exists;
- that it relates to treatment or therapy;
- that it relates to the incident itself;
- that it may disclose a prior inconsistent statement;
- that it may bear on credibility or reliability.
The list goes on but the purpose is explicit: “to prevent speculative and unmeritorious requests for production.”² Even where the threshold is met, a judge retains full discretion to limit, redact, or refuse production.
The Paradox at the Heart of Bill C-16
The proposed threshold does not merely raise the bar – it makes the bar functionally unreachable.
It is important to be precise about what is new here and what is not. The structural difficulty of making this application has always existed. Since O’Connor, defence counsel has had to argue for records they have not seen, based on materials they do not possess. There has always been an element of blindness to this process.
But under the current “likely relevance” standard, that blindness was manageable. The test required only a logical connection between what the records might contain and an issue at trial, a connection the defence could infer from the circumstances of the case, the nature of the therapeutic relationship, or inconsistencies already apparent in disclosure. Once that relatively accessible threshold was met, the records went to the trial judge, who could assess whether they warranted production. There was, at minimum, a realistic path to discovery.
Bill C-16 closes that path.
At its core, what Bill C-16 proposes is that an accused must demonstrate their innocence in order to access materials that might help establish it.
That is not a higher bar – it is a different kind of bar entirely. It is a structural impossibility, and it is a next-to-impossible task for a trial judge to adjudicate on a pre-trial motion, before the evidence at trial has even begun.
The Uncomfortable Truth
None of this is an argument against protecting complainants.
Parliament can protect complainants without creating a threshold that will generate years of inconsistent application, invite inevitable constitutional challenges, and – if the courts ultimately find the provision unconstitutional – provide no protection at all.
Balance is a fragile tight rope, but balance is what we must aim for, above all else, especially when freedom is at stake.
The concern with Bill C-16, put plainly, is this: when it comes to therapeutic records, exculpatory evidence must be discoverable. Raising the standard to such a degree that it becomes functionally unworkable does not balance competing rights: it eliminates them.
If one of the rights eliminated is the right to make full answer and defence, the downstream risk is wrongful convictions. That risk undermines public confidence in the accuracy of every verdict the system produces.
The Bottom Line
Bill C-16’s amendments to the records regime are presented as a protection for complainants. In substance, it is parliamentary drafting that looks like compassion and functions like punishment for an accused who has not yet been convicted of anything.
The Standing Committee on Justice and Human Rights has the bill in its hands right now. The time to fix this language is before it becomes law, not after the first wrongful conviction it enables.
