Strengthening Evidence-Based Prosecution of Gender-Based Violence: Challenges in Proving Sexual Violence Cases in Criminal Justice Systems

Abstract

Background: Sexual violence is among the most difficult forms of gender-based violence to prosecute. Criminal cases frequently depend on evidence generated in private settings, delayed reporting, contested consent, fragmented forensic pathways, and credibility assessments that may be influenced by stereotypes. Although reforms have expanded forensic services and specialist responses, high attrition persists across criminal justice systems.

Objectives: This study synthesizes empirical evidence on the factors that strengthen or weaken proof in sexual violence cases, identifies preventable points of evidentiary loss, and develops an evidence-informed prosecution framework.

Methods: A structured evidence synthesis was conducted using empirical studies published from 2000 through 2020 and identified through PubMed/MEDLINE, U.S. Office of Justice Programs/National Institute of Justice/NCJRS repositories, publisher platforms, and backward/forward citation chaining. Eligible studies examined real-world sexual violence cases, criminal justice decision-making, forensic evidence use, or system interventions linked to investigation, referral, charging, trial, or disposition. Twenty-four studies from five national settings were included. Heterogeneity in legal systems, populations, outcomes, and overlapping datasets precluded meta-analysis; findings were narratively synthesized and appraised using design-sensitive domains informed by the Mixed Methods Appraisal Tool.

Results: Six recurrent proof domains emerged: credibility filtering and rape-myth influence (16 studies), corroboration and investigative completeness (15), consent/relationship complexity (13), organizational coordination or specialist response (11), victim engagement/support/withdrawal (10), and forensic/medical evidence quality or use (8). Forensic evidence was most useful when integrated with broader investigation rather than treated as a stand-alone determinant. Specialist SANE/SART models improved evidence collection or case progression in several studies, but did not eliminate attrition. Across settings, inconsistencies, delayed reporting, intoxication, prior relationships, and perceived victim behavior were often interpreted as credibility problems, creating a risk that extralegal heuristics substituted for an evidence-based assessment of the legal elements. Untested-kit initiatives further showed that evidence can be lost through non-use even after collection.

Conclusions: The review supports a shift from a narrow focus on whether evidence exists to whether it is available, utilized, and interpreted fairly. The proposed Evidence Integrity and Prosecution Continuum (EIPC) links these three dimensions to each decision stage, offering a novel, operational framework for bias-resistant, trauma-informed, and legally rigorous prosecution without weakening due-process safeguards.

Publication history: Originally published in print on 15 June 2021. Online abstract record added on 16 September 2026. Full-text PDF will be added when the digital galley is ready.