America tested 100,000 forgotten rape kits. But justice remains elusive.

On summer nights a year apart, two women left bars in downtown Austin, Texas. Before they could get home safely, both were raped.

One knew her rapist. He was an acquaintance, she said, who saw she was intoxicated and offered her a ride on the handlebars of his bicycle before sexually assaulting her as she drifted in and out of consciousness.

The other did not. She told police she was attacked by strangers who pinned down her arms and legs while one of them raped her.

Neither investigation led to an arrest.

Years later in 2021, Austin police received a significant new lead: DNA found on one woman’s jeans matched DNA from a swab of the other’s neck. It suggested that the two women had been attacked by the same man – a serial rapist.

This is the kind of breakthrough officials across the country hoped for when they pledged to test sexual assault evidence that had been neglected for years in law enforcement storage rooms. Since 2015, the U.S. Department of Justice has given out nearly $350 million in grants to 90 local and state agencies through the National Sexual Assault Kit Initiative. Officials promised the money would put rapists behind bars and give victims long-awaited answers.

Instead, a USA TODAY investigation found, cases hit the same roadblocks they did when victims first came forward: kits left untested, haphazard or cursory reviews by police and prosecutors, and a reluctance to inform people about what happened to evidence collected from their own bodies. By the Justice Department's count, the program has led to 100,000 kits being tested and 1,500 convictions so far – nearly half from two agencies, while others have seen meager results.

In Charlotte, North Carolina, a backlog of about 2,300 kits has netted 14 convictions.

In Mobile, Alabama, a backlog of about 1,100 kits has led to convictions of eight men.

In Austin, officials faced a backlog of more than 4,400 kits. They have secured just one conviction.

After the two Austin cases matched through DNA, a detective reviewed the files. Police said they contacted the woman who had reported being raped by a stranger, but she did not want to speak with them, and her investigation was again closed. Officials did not tell her that the man linked to her kit could be a serial rapist.

On the other case, police botched their new review and did nothing.

Krystal Allison, the woman who made that report, did not learn that the acquaintance she accused of rape had been linked to another reported sexual assault until USA TODAY contacted her this summer.

“That seriously sent chills down my spine,” Allison said, her voice shaky. “Nobody has reached out to me at all. I know nothing of it.”

America’s untested rape kits piled up across the country over the course of decades.

Boxes about the size of a hardcover book each held evidence of a reported sexual assault – dried swabs of saliva and semen and blood, strands of hair, debris scraped from under fingernails. Each was collected from a person, most of them women and girls, during an hourslong exam. And each was shelved without being processed for DNA. The evidence crowded storerooms, tangible proof of law enforcement’s failure to support victims and hold rapists accountable.

Testing the kits was supposed to be the first step in righting that wrong. In some places that received federal grants, not even that happened.

At least a dozen grant recipients carved out exceptions to testing, leaving kits unprocessed for a second time. In one California county, officials boasted they had cleared their backlog, but only after deeming more than half of their kits ineligible for testing.

In many cases, officials have done little beyond sending the kits to a lab, reviewing the results and again closing the files. In Maryland, according to a state report, some law enforcement agencies have shown “significant reluctance” to reopen investigations and have even stated outright that they are disregarding DNA matches.

What’s more, some officials all but abandoned the idea of providing victims answers about what happened to their rape kits or apologies for how long testing took. One Kansas police agency has tried to reach just 17 victims from roughly 1,100 sexual assault kits. An official there said there are instances where DNA testing has identified the names of suspects for the first time but the victims have not been told, because officials don’t think their cases can be prosecuted.

No?l Busch-Armendariz, a University of Texas at Austin professor and researcher who has interviewed women with backlogged kits, said it is disheartening that victims – yet again – have been overlooked.

“Survivors of sexual assault mostly feel betrayed by the silence around what happened to them,” she said.

Tell us your story: If you know of a rape case that was neglected, or have experience with the system, tell us here.

Officials from the Department of Justice who oversee the Sexual Assault Kit Initiative declined to be interviewed.

In a written statement, spokeswoman Katherine Brown said the program has been instrumental in helping agencies across the country process untested kits, although she also noted that local laws and regulations can influence what happens next. Angela Williamson, who has led the federal program since its inception, acknowledged in a separate statement that local success will vary based on the level of commitment.

But Williamson said the program has contributed to a “critical cultural shift” around the importance of testing sexual assault evidence. In addition to seeking justice in backlogged cases, the initiative aims to overhaul how the criminal justice system responds to sexual assault. Grantees have also used their funding to improve DNA collection and its use, train police and prosecutors and modernize how survivors and their families are treated.

Williamson said it is too early to gauge the program’s success by the number of rapists who have been convicted because it can take years for cases to move from testing to potential conviction. Agency officials said DNA matches also can pop up long after a kit is tested as more profiles are uploaded to the federal database of DNA samples.

Williamson made a similar argument – that it’s too soon to judge the results – when The Atlantic highlighted the program’s paltry outcomes five years ago.

Yet the program's data tracking is so poor that in some places, the federal government has no idea how many rapists were convicted – and likely never will.

Federal performance data is riddled with errors, misleading figures and blank fields, and officials were overcounting convictions on their website by 20% until after USA TODAY pointed out a discrepancy. At least 10 grant sites told USA TODAY they have not been reliably tracking key metrics such as the number of victims contacted or convictions won. A 2022 audit by the agency’s Office of the Inspector General found grant staff at the Department of Justice failed to identify “problematic and struggling grantees” and that three of five sites reviewed couldn’t provide documents confirming the numbers they reported to the government.

Given the shortcomings of the federal data, USA TODAY contacted grant sites that received funding from fiscal year 2015 to 2022 to confirm key numbers, such as kits sent for testing, cases charged and victims contacted. Reporters then focused on 14 with the most complete data that were funded in the program’s early years, giving them the most time to test, investigate and prosecute.

Two sites – prosecutors’ offices in Cleveland and Detroit – stood out, with more than 600 convictions combined. The remaining 12 reported 123 convictions, fewer than one for every 175 kits sent for testing. If they had been as successful as Cleveland at investigating and prosecuting cases, USA TODAY estimated they together would have secured at least 1,450 more convictions.

Nearly a decade ago, the USA TODAY Network conducted what was then the most detailed nationwide inventory of untested rape kits. Reporters tallied at least 70,000 kits across more than 1,000 agencies, a shocking number that accounted for only a fraction of the nation’s police departments.

Following up on that work, USA TODAY reporters spent the past year and a half investigating the Sexual Assault Kit Initiative, including interviews and correspondence with more than 150 law enforcement officials, advocates, victims and others connected to backlog-clearing projects. Reporters also filed more than 250 public records requests, reviewed police reports from more than 350 cases with backlogged kits and asked 10 experts familiar with investigating and prosecuting sexual assault to review a selection of cases.

Often, at the center of such cases were women who were particularly vulnerable to predators.

They were sleeping on the streets, battling addiction, facing mental health challenges or simply drinking with friends. Some faced law enforcement skepticism from the minute they called for help, such as one Florida woman referred to by a sheriff’s deputy as a “potential victim” 45 times in a 40-line incident report.

Most knew the men they said attacked them. Investigations often reached a stalemate at “he said, she said” – with one party saying the sex was consensual, the other saying it was rape and investigators doing little or no work to determine the truth.

While the Sexual Assault Kit Initiative is meant to challenge the misconceptions about victims that have long plagued the criminal justice system, those biases remain.

A Florida prosecutor rejected a case after a kit was tested by noting the victim’s use of pain medication had impacted her “cognitive abilities.” A Minnesota police department decided to not test a kit because the juvenile victim had been “uncooperative” with their prior investigation, though records show the girl told police she feared she had been raped while passed out at a party and asked: “What if it was my fault?” A Wisconsin detective took no steps to reinvestigate a rape and noted that the suspect had claimed the sex was consensual, even though the same man had been convicted of raping someone else.

Ashley Nix is a sex crimes prosecutor in Tulsa, Oklahoma, whose office has filed charges in numerous cases from the city’s backlog. One victim was a sex worker. Another died in the 12 years since she reported being assaulted, leaving Nix to prosecute the case without her testimony. A third woman – in a case Nix’s predecessor first rejected – was living with an intellectual disability.

“These excuses about messy cases and messy victims … it’s BS,” Nix said. “This is what predators do. They pick targets that people won’t believe.”

Krystal Allison, the woman who told Austin police that she had been raped by an acquaintance, said he knew full well that she was too intoxicated to fight back. She had become separated from a girlfriend on Austin’s bustling Sixth Street when the man stopped and offered her help. From there, her memories are fractured: falling off the handlebars of his bicycle, being in a bed, trying to push him off as he raped her.

In the morning, Allison said, she realized she was at the man’s apartment, and his roommate told her she had been carried up the stairs the night before. She left, walked straight to a hospital, had a sexual assault exam completed and made a police report.

That report would ultimately doom her chance at justice – because the officer wrote that Allison and the man had had “voluntary sex.”

Years later, when the sexual assault kit was tested – and matched to another reported rape – that errant idea became canon, repeated over and over. If anyone had read Allison’s file more closely, they would have found numerous indications that what she reported was not voluntary, but rape.

In an interview with USA TODAY last month, Austin police officials acknowledged they made mistakes on Allison’s case, ones too late to rectify even if they wanted to. The statute of limitations to bring charges closed last year.

Sgt. Patrick Stewart is one of several people at the police agency who reviewed the case and decided Allison did not need to be contacted.

“Unfortunately, we dropped the ball on that one,” he said.

USA TODAY does not identify victims of sexual assault without their permission. Allison and several other women agreed to the usage of their names and photographs.

Hearing from a reporter that the statute of limitations on her case had closed, Allison struggled for words. In the 11 years since she reported being raped, the now 40-year-old mother of two had not given up hope that one day her rapist would be arrested. When the police first closed her case and told her there wasn’t enough evidence to move forward, she recalls asking whether her investigation could be reopened if the man ever raped someone else. She said she was told it could be.

“I truly felt that at some point, he would get caught. You don't get away with something like that forever,” she said. “You don't.”

So far, he has.

Lt. Chris Leleux, who oversees sex crimes investigations at the Austin Police Department, told USA TODAY he is disappointed that so few cases from the city’s backlog have resulted in convictions. When officials pledge to test old rape kits, though, he said they often naively present it as a panacea.

“These cases,” he said, “they’re landing back into a flawed system.”

Beyond unsolved rape cases, a broader swath of crimes

When police and prosecutors fail to hold rapists accountable, more people get hurt. They are raped and kidnapped. Stalked and harassed. Beaten and killed.

Rapists are often repeat offenders, research shows. And because so few sexual assaults are reported to police – roughly a third, studies estimate – the stakes for solving them are remarkably high.

USA TODAY reviewed the criminal histories of more than 250 men named as suspects in rape cases where kits were not originally tested.

Their backgrounds – before and after those kits were collected – are full of not only additional sexual offenses, but a litany of violent crimes including kidnapping, armed robbery, assault and child abuse. About 1 in 10 have been charged with domestic violence.

A Florida man was convicted of domestic battery at least six times, including attacking a woman after she refused to have sex with him. A Wisconsin man slashed a woman’s tires and grabbed her by the neck, and then beat up a friend who tried to help her. A Minnesota man was required to register as a sex offender after groping a young girl in the Barbie aisle at Target.

All three men were also named as rape suspects. All three backlogged rape kits were counted as part of the Sexual Assault Kit Initiative. But none of the men was arrested in those cases.

Neither was Mark Burruss.

In 2013, a Wisconsin woman named Amber told police she had agreed to have sex with two men she met online but that the consensual encounter turned violent and the two men took turns raping her. A month later, she decided she did not want to pursue the case, and her rape kit was shelved.

Three years passed before it was tested as part of Wisconsin's grant program. DNA found in the kit matched to Burruss, who had by then been accused of two sexual assaults against minors, court records filed by prosecutors show – the oldest dating to 1995.

When interviewed by a detective, he said the sex with Amber was “rough” but had been entirely consensual. Amber maintained that it wasn’t. She told detectives she wanted to pursue the case this time around and hadn’t in 2013 only because her boyfriend discouraged her.

But a prosecutor in 2018 declined to file charges. The office cited “credibility issues” with Amber because she initially had not been forthcoming about meeting the men online and had previously said she did not want to go forward with the case.

Two weeks later, Burruss groped a 13-year-old girl’s breasts and vagina over her clothing. The teen messaged a relative minutes after the assault, pleading with him to come get her. “I don’t feel safe,” she wrote.

Burruss was convicted in that case and served two years in prison. He is currently awaiting trial on charges he raped an 11-year-old girl in 2016 – after Amber first came forward but before her kit was tested. He did not respond to a request for comment made through his attorney.

Amber, who asked to use only her first name to protect her privacy, said that when she first reported her rape, a police officer told her that because she had met the men online and agreed to have sex with them, she “basically gave them permission to do whatever they wanted.”

She did not know that Burruss had been convicted in another sexual assault after prosecutors failed to take her case in 2018, but she was not surprised. It was obvious, she said, that “he had done this before – and it wouldn’t be his last time.”

Kit backlog shocks nation, spurs promise of reform

In 2009, one of Wayne County Prosecutor Kym Worthy’s assistants discovered 11,000 untested rape kits in a Detroit police evidence warehouse. Worthy was outraged.

Officials in a handful of other U.S. cities had already found similar caches, and more communities would soon follow.

These untested kits – estimated by one study to total between 300,000 and 400,000 – have come to be known collectively as “the rape kit backlog.” But that description is misleading. The word backlog implies they were in line to be tested. In reality, they sat forgotten. Some grew mold. Some were damaged in floods. Some lacked labels linking them to police reports, so they had to be manually catalogued one by one.

Public outrage built as media reports told the stories of unsolved rapes with untested kits. In 2010, the Joyful Heart Foundation, a nonprofit founded by “Law and Order: Special Victims Unit” actress Mariska Hargitay, listed elimination of the backlog as its top priority. The same year, members of Congress convened one of many hearings, joining survivors and advocates who demanded to know: How did this happen?

Research has since shown that decisions on testing were often left to the discretion of individual detectives. And too often, those detectives thought victims were lying. Their decisions were also impacted by changes in DNA technology and limited lab resources, leading many police to exclude entire categories, such as cases where they already knew the suspect's name.

As Detroit and other cities began to test their backlogged kits, it soon became clear how misguided that mindset had been as the DNA in kits began matching to other cases.

Men who raped strangers had also raped people they knew, and links among those cases identified serial rapists who had eluded police. In other scenarios, suspects who had claimed the woman consented to sex were now facing multiple accusations – turning “he said, she said” into “he said, she said, she said, she said.”

In 2015, Joe Biden, then vice president, announced the National Sexual Assault Kit Initiative.

“When we solve these cases, we get rapists off the streets,” he said. “For most survivors, seeing their rapists brought to justice, and knowing that they will not return, brings peace of mind and a sense of closure. The grants we’re announcing today to reduce the national rape kit backlog will bring that sense of closure and safety to victims while improving community safety.”

Since then, the federal government has not only given grants to 90 state and local agencies but also hired a private nonprofit to provide training and assistance. Through that program, experts from across the criminal justice and advocacy systems are available to consult with local teams and have created training materials to guide their work.

But their recommendations are ultimately just suggestions – which some teams have chosen to ignore.

Rebecca Campbell, a professor at Michigan State University, is part of the grant’s training program and has worked with dozens of sites. Campbell said it has been “eye-opening and difficult and challenging” to see the uneven results. In some places, she said, members of local teams have remained resistant even to the idea that the kits should be tested.

“There’s sort of a sense of, ‘Why in the world do we have this grant? Why are we doing this project?’” Campbell said. “And there’s active hostility to the project, the grant, because they don’t believe that there’s a problem.”

At the other extreme, she said, in some places officials have fully bought into the mission. For them, the challenge can be burnout and lack of resources.

David Thomas, a law enforcement consultant and former program manager at the International Association of Chiefs of Police, has also worked with the grant’s training program. He said agencies that underperform need to be held to a higher standard. That starts with local police officials demanding thorough reviews of every case. And it extends, he said, to officials at the Department of Justice being willing to withhold grant funds.

“We talk about ‘We need to hold perpetrators accountable,’” he said. “Agencies that aren’t living up to their promise … we need to hold them accountable.”

The program has seen one of its greatest successes in Wayne County. The kits Worthy’s assistant found in 2009 have all been tested, resulting in more than 5,400 investigations and 257 convictions so far, according to the prosecutor’s office.

Worthy, who was raped as a law student, said that within minutes of hearing about the kits, she knew she wanted to process them, resolve every case and make sure another backlog never developed. She wanted to change the culture that had once made it acceptable to stash rape kits in a dilapidated warehouse, like junk.

Other city leaders were daunted by the cost, she said. Four years later, Detroit would file for the largest municipal bankruptcy in U.S. history.

Over time, Worthy scrounged together money to start addressing the backlog, even seeking private donations as small as $5. Nonprofit groups helped with fundraising, and the Detroit Crime Commission negotiated lower testing rates with private labs. Her office received a grant from the National Institute of Justice, as well as several grants from the Sexual Assault Kit Initiative totaling $9.5 million. The Michigan State Legislature allocated funds, and the county also increased Worthy’s budget to help support the program.

Worthy said it has been a “long and difficult entangled road getting to where we are now.” She is frustrated by any agency that considers testing its kits alone a measure of success.

“I can’t say you did nothing. But you didn’t do anything that’s going to lead anybody to justice if you didn’t do anything to seek to get these perpetrators, these rapists, off the street,” she said. “Because these rapists don’t stop.”

Campbell, who has worked as a researcher with the Detroit team, said they set a standard every site should be striving to meet.

“If you can do it under what I would argue are some of the hardest circumstances in the U.S.,” she said, “it is possible.”

Despite DNA advances, law enforcement efforts fall short

In 2022, a state prosecutor in Jacksonville, Florida, reviewed a rape case from 2010 and decided, despite new DNA evidence, that there was no “reasonable probability of conviction.” The woman had taken a pain pill before the incident, impacting her memory. There were no witnesses to corroborate her account. And she had willingly gone into the suspect’s hotel room the morning after she claimed he raped her.

It would be a tough sell to a jury.

Unless you consider the same facts through a different lens.

The suspect saw the woman take the medication, so he could have known that she would be groggy when he snuck into her room later that night. Most sexual assaults have no witnesses, and this woman had at least mentioned the rape to a friend who still recalled the conversation years later.

And yes, she had gone into the man’s hotel room the day after the incident. While there, she covertly got his name off a prescription pill bottle – identifying her attacker for authorities.

If anything, the case developed new strengths with time, according to experts who reviewed the file for USA TODAY. When a detective reinterviewed the woman, she said the assault haunted her and she had been celibate ever since – evidence of trauma that could be compelling to jurors. When a detective interviewed the suspect, he denied ever having sex with the woman even though his DNA was found in her sexual assault kit – raising questions about his credibility.

“What more do you want?” asked Elizabeth Donegan, a retired senior sergeant with the Austin Police Department and expert in investigating sexual assault. “To me this is an easy case to move forward. Super easy. You have a (prosecutor) that doesn’t know what they’re doing, doesn’t understand the crime of sexual assault.”

A close examination of cases from the Jacksonville Sheriff’s Office shows the long odds for a backlogged kit to result in an arrest, let alone a conviction. Across the United States, it often requires near perfect conditions: first, a police agency with the resources, training and enthusiasm needed to reinvigorate old cases and, second, a prosecutor who has the skills and willingness to take a suspect to trial. One missing link can cause a case to careen off course.

Some cases do not move forward because of reasons not easily overcome by police or prosecutors. Once tested, roughly half of backlogged rape kits have a useable DNA profile, and only half of those result in a DNA hit. A match could be to another case with an unknown assailant, providing police a lead but no name. The statute of limitations to bring charges might be closed. Or a suspect might be deceased.

Frequently, officials say, people contacted following testing that happened years after the incident no longer want to pursue the case. Dropping the investigation at that point can cut both ways: It can honor a victim’s wishes or it can signal that police didn’t do enough to earn her trust.

But USA TODAY also found that police have given cases only a cursory review after DNA testing, instead relying on conclusions from the prior investigation. Lisae C. Jordan, executive director and counsel at the Maryland Coalition Against Sexual Assault, said some agencies view the state’s backlog project as “second guessing” their decision not to test the kits to begin with.

Prosecutors, too, have resisted giving investigations a fresh look. Katie Whisman, a former employee at the Kansas Bureau of Investigation who oversaw the state’s backlog effort, recalled one local prosecutor telling police that they were wasting their time reopening investigations. The office had rejected the cases once and would not review them again.

When asked why so few cases move forward, law enforcement officials are often vague. In Charlotte, North Carolina, records show authorities identified suspects in several investigations where women had reported being raped by armed attackers they didn’t know – including one who said she was kidnapped at gunpoint and held for five days. The cases were closed without arrests. Michael Stolp of the Mecklenburg County District Attorney's Office told USA TODAY that after review by police and a prosecutor, “consensus … was that an arrest and prosecution was not appropriate based on all the available evidence.” He didn’t elaborate on what evidence they relied on and did not respond when repeatedly asked if the victims had been told of the testing results.

In Jacksonville, the prosecutor’s and sheriff’s offices have received nearly $10 million in federal grants since fiscal 2015.

Investigations there have dragged on for years, sometimes with long stretches of no activity, a review of more than 100 files showed. The woman who reported securing her attacker’s name from a pill bottle in his hotel room first learned in 2017 that her rape kit had been tested. Records show she did not hear from a detective again for nearly two years. By then, a new investigator had been assigned. After that detective sent the case to state prosecutors in 2019, it took another 2? years for the office to decline to file charges.

“I didn’t quite realize there was another way for the system to traumatize victims,” said Kelsey McKay, a former prosecutor who founded a nonprofit focused on improving the way criminal justice professionals interact with sexual assault survivors. “If they have survived until now, now there’s some hope involved. And as you see with all these cases, now even with DNA, we don’t believe you.”

McKay, who reviewed several cases at USA TODAY’s request, said that because most victims know their rapists and cases hinge on consent, police and prosecutors often focus on the victim’s credibility when they should be focused on the suspect’s.

She acknowledged that even the most skilled prosecutors face a steep challenge in convincing 12 jurors who might lack a nuanced understanding of sexual violence. But she said prosecutors need to be more willing to take on that challenge, including explaining to jurors that predators target vulnerable women because they know those women are less likely to be believed.

Failing to prosecute, she said, just proves those perpetrators right.

McKay and the five other experts who reviewed the Jacksonville hotel case for USA TODAY all thought the prosecutor’s office could have filed charges.

“If not this, what kinds of cases do we take with vulnerable victims?” she asked.

In Jacksonville, prosecutors have filed charges in cases from the backlog that involve women who were homeless, intoxicated or using drugs. But almost without exception, those cases included another element: The victim said she did not know her attacker. Of 29 suspects the office has charged, 80% were accused of raping a stranger, in many cases women they kidnapped at knife- or gunpoint or with whom they had only briefly interacted before the assault.

Multiple studies have concluded that in almost all rape cases, victims know their attackers.

The Jacksonville prosecutor’s office has not filed a single case where the suspect and victim had been in an intimate relationship. In the reports USA TODAY reviewed, detectives at the sheriff’s office often made little effort to investigate such cases after the kits were tested, closing many with the same language: “A review of the case showed that this was the same suspect named in the original investigation. Case remains Exceptionally Cleared.”

Jacksonville Sheriff’s Office Sgt. Robert Freitas, who is part of the agency’s cold case unit and worked on investigations related to backlogged kits, told USA TODAY that as officials sifted through hundreds of DNA matches many cases remained difficult to prove because it was still a suspect’s word against a victim’s.

“We’re not going to make an arrest just to make an arrest,” he said. “It’s unfortunate sometimes. You’ll see a match, but the state’s attorney has a higher burden of proof.”

David Chapman, a spokesman for the State Attorney’s Office Fourth Circuit in Florida, told USA TODAY that the office hasn’t charged any backlog cases stemming from domestic violence because those suspects were already known to law enforcement during the first investigation. He described the federal initiative’s “main purpose” as prosecuting previously unknown assailants – even though participants are offered training on how to prosecute rapes involving partners and known parties.

He added that when screening sexual assault cases from the backlog, his office weighs the fact that defendants can claim their due process was violated because of how long it took to bring charges. The office also considers other factors defense attorneys might highlight at trial. For example, he said, if a victim was a sex worker or using illegal drugs, the defense will say they had a motivation to lie.

Chapman provided a summary explaining how the office weighed several cases reviewed by USA TODAY. The first factor he noted: criminal histories of the suspect, as well as the victim.

“They are classifying an entire group of women as unrapable,” said Campbell, the professor who works with the grant’s training program. “That is absolutely unacceptable.”

In another case from Jacksonville’s backlog, an 18-year-old woman in 2012 told the sheriff’s office that she had been raped by a man she knew only as “Charlie Fresh.”

That woman, Elizabeth Levy, told USA TODAY that she hasn’t forgotten how the deputy who took the report responded: He laughed as she struggled to recount the details while under the influence of crack cocaine.

Levy had gone to a motel near the interstate to buy drugs. She said Charlie offered to let her stay in his room – then gave her cocaine, refused to let her leave, threatened her and raped her. Soon after making a report to the sheriff’s office, she stopped responding to deputies’ calls.

Her sexual assault kit was tested in 2017 and matched to Charlie Pressley, a Florida man who had recently been convicted on felony drug charges and who, law enforcement records show, had twice been accused of sexual assault. In one of those cases, he was charged and pleaded to a lesser offense.

Even though Levy knew she could be attacked at a criminal trial – cast as untrustworthy and unsympathetic because she had gone to the motel to use cocaine – she agreed to help with the investigation. Shown a photo lineup, she picked out Pressley immediately. She told USA TODAY that she had only stopped pursuing the case because she was afraid of Pressley, and that had since changed.

“I just wanted some justice,” Levy recalled in an interview earlier this year. “Honestly, seeing that mugshot of his face, it just did something to my soul.”

The investigation that followed stretched 2? years. Months passed with no activity. Detectives reinterviewed witnesses from the motel and looked for the other women who had accused Pressley of sexual assault. They also interviewed Pressley, who said he had never had sex with Levy.

Over that time, officials lost contact with Levy, who was incarcerated when they first reached her but was released during the investigation. Though their protocols called for it, the sheriff’s office never connected Levy with a victim advocate who could have helped her stay engaged. Levy said she recalls one Facebook message from a sheriff’s deputy after her release, but by then she was again struggling with addiction. She also had lost faith in the detectives, who told her they would keep in touch and then went dark.

She spoke with a detective only once after the photo lineup, sheriff’s office records show. He asked if she had agreed to have sex with Pressley in exchange for drugs.

She maintained that it was rape.

Three years after testing Levy’s sexual assault kit, the Jacksonville Sheriff’s Office in June 2020 again suspended the case and blamed Levy, writing “victim failed to cooperate” in the file. The prosecutor’s office told USA TODAY that there had been numerous inconsistencies in Levy’s statements and that her drug use while under court order not to use drugs “would call into question her reliability and credibility as to her version of events.”

Pressley, currently in prison on an unrelated drug conviction, did not respond to a request for comment. He is set to be released this fall.

Levy was released from prison last week after serving 16 months for a parole violation on a drug charge. She recently graduated from an empowerment program for female inmates and is moving into a transitional house in Jacksonville.

She still struggles with blaming herself, both for the assault and for the fact that Pressley was never arrested. But she knows what happened wasn’t her fault.

“I’m not going to lie and say I’m a saint, because I’m not. There are lot of things that I did that caused this situation to happen,” she said. “But I didn’t ask for it.”

‘Nobody wants to apologize’

Officials from the Department of Justice have stressed that securing convictions is not the Sexual Assault Kit Initiative’s only goal, or even the main one. Williamson, who oversees the program, told USA TODAY the primary focus is resolving cases to provide “answers for victims or surviving family members.”

Yet most people whose sexual assault kits have been tested don’t even know they were tested, much less the results.

Police in Wichita, Kansas, attempted to notify 17 victims from roughly 1,100 kits and were able to reach 11 of them. Two suspects have been convicted, and a warrant was issued for a third.

Capt. Jason Stephens, who oversaw sex crimes investigations for the Wichita Police Department when Kansas in 2015 received a federal grant to address its backlog, said officials were concerned about upsetting victims by bringing up their rape if police had no plans to reopen an investigation. This means there are cases where police know who a suspect is, but the victim does not.

That is exactly what Mary Stolz, executive director of the Wichita Area Sexual Assault Center, feared when she heard officials attempted to reach just 17 people. She considered a scenario where a woman moves next door to her rapist, or goes on a date with him, or he becomes her child’s teacher – all while the police know that man’s name and have not told her.

“That is such an egregious violation of human rights,” Stolz said.

Victim notification is arguably the most sensitive step officials face when clearing a stockpile of old kits. For one person, receiving a call about their sexual assault could serve only to open a wound. For another, it could help close a wound that never healed, providing not just information but access to resources such as counseling.

Faced with the dilemma, most grant sites have leaned on the side of limited contact. Of 42 notification policies obtained by USA TODAY, only eight recommend contacting all or close to all victims. At the other extreme, at least six policies directly link whether a victim is notified to whether authorities think their case could be investigated or prosecuted.

USA TODAY examined victim notification rates for 14 agencies with early grants and the most complete notification statistics, comparing the number of victims contacted to kits sent for testing. The ratio does not take into account that some victims may have had more than one kit in the backlog.

Taken together, four of those 14 grantees with the highest rates notified about one person for every two kits sent for testing. The others reached just one victim for every 15 kits, and a few of those contacted barely anyone. In both Wisconsin and Orange County, California, officials reached one person for every 43 kits sent for testing.

Campbell said some grant teams have been almost paralyzed by the fear of traumatizing survivors by bringing up their rapes and “practically convinced themselves that it would be unethical to contact victims.” In one meeting, she recalls saying with exasperation: “You just need to start.”

Then there are other places, she said, where concern for victims has become an excuse to do nothing.

“Somebody on that team has to come face to face with the survivor and whatever feeling that survivor has – anger, relief, fury, whatever it may be,” she said. “And they need to be able to have something to say. Hopefully it's an apology. And in a lot of jurisdictions, nobody wants to apologize.”

Fewer than half of the policies reviewed by USA TODAY mention the need for an apology. The Duluth Police Department’s in Minnesota is one of them.

Mary Faulkner, a victim advocate and Duluth’s grant site coordinator, said that when considering how to approach notification, the local team knew people across the small city of 90,000 residents had always wondered what happened to their cases. Perhaps some had even heard on the news or from a friend that the city was testing old kits and wondered if they would get a call.

“We were very concerned about the sense that people would feel like they were forgotten twice,” Faulkner said.

Starting from an assumption that every victim would be contacted, officials only deviated if there was a specific cause for caution. The team refined its process as they saw what worked. Every victim they skipped, they documented why.

From 444 kits sent for testing, Duluth attempted to reach 274 people. They successfully contacted 187 of them. Another 16 victims had died since making their reports.

Prosecutors have filed charges in 21 cases and convicted seven people so far.

Faulkner made many of the calls herself. Some did not go well, and those conversations stay with her. More often, people thanked her but said they had moved on with their lives. Some were curious about what was being done to make things better for victims now. And some said they had always wondered if their kit hadn’t been tested because of something they did wrong.

“My hope is that it was helpful to have someone apologize and say that it wasn’t your fault,” Faulkner said.

Jordan Martin has never received that kind of call.

One evening in 2018, the Wichita woman decided to spend the night at a hotel, thinking it would be safer than driving home from a bar. An acquaintance offered her a ride, she said, and then followed her to her room and raped her.

She called the police in the morning. The detective’s questions were short and sharp, as if he was at the end of a shift and eager to leave, she said. A few weeks later, Martin got a voicemail saying her case had been closed. The suspect said they had had consensual sex. The police said nothing about whether her kit had been processed.

“It just felt cold, that voicemail,” she said.

Numb and angry, she turned to alcohol before realizing she needed help and reaching out to the Wichita Area Sexual Assault Center. In time she learned to cope better with her emotions and found purpose in helping others do the same. Today, Martin is the center’s director of advocacy.

Wichita police officials told USA TODAY that the department has instituted reforms to better respond to sexual assault, including increasing training and creating a comfortable room where victims are interviewed. The department announced a policy of testing all rape kits in late 2018, years ahead of the rest of the state.

Martin said that since her own rape, she has seen the local police improve how they treat victims.

Yet it still bothers her that she has no idea what happened to her kit.

“I did what the detectives – I did everything that they wanted me to do by talking with them, by going to the hospital before taking a shower to try to get this person off of me,” she said. “And for me just to always be like, ‘Well, damn, did they even care?’”

Claims of success obscure shortcomings

When Iowa wrapped up its six-year grant program in 2021, the state’s Senate Democratic Caucus posted a celebratory message on social media: “After years of work, Iowa’s rape kit backlog has finally been eliminated!”

Except Iowa did not eliminate its backlog.

It tested just two of every five kits that had accumulated in law enforcement storage rooms, the oldest dating to 1992. And the state still does not test all kits today.

Kerri True-Funk, a member of the state grant team who worked for the Iowa Coalition Against Sexual Assault, said that looking back, she wonders what could have been accomplished if Iowa law enforcement officials had used the initiative as an impetus for change.

“It definitely could have been a pivot point for the state,” she said. “It could have done, maybe, not that 180-degree turn – it could have made a 90-degree turn. It could have moved the state forward.”

Iowa’s Office of the Attorney General received its federal grant in 2015 and the following year surveyed law enforcement agencies across the state about their backlogs. In total, 4,264 kits were tallied. State officials, in news coverage and state reports, said that there would not be enough money to test them all.

Iowa and other grant recipients have set their own criteria in deciding which kits to prioritize. The federal grant program recommends that sites test all kits from reported crimes. But at least a dozen sites have eschewed that suggestion and carved out exceptions.

Kits were excluded for a variety of reasons, including if a suspect’s DNA was already on file from a prior case or if someone already had been convicted of the crime, eliminating the possibility that testing could clear someone wrongfully convicted.

Communities also have not tested kits from cases police concluded were “unfounded” – meaning they believed the victim made a false report or baseless accusation. Research shows police often overuse and misapply the label in sexual assaults.

Minnesota’s Bureau of Criminal Apprehension last year announced that its testing was “complete,” and then explained that just two-thirds of kits were sent for testing. In California, the Orange County District Attorney’s Office in 2022 said officials had “successfully cleared a 30-year countywide backlog of untested sexual assault kits” when it had processed less than half of the county’s kits. In Mobile, Alabama, at the end of negotiations among the grant team, about a quarter of backlogged kits were excluded from testing.

In Iowa, officials based their testing strategy on the reasons law enforcement officials said the kits were not tested to begin with, according to state reports. Cases that were prioritized for testing included those where the police had doubted the truthfulness of the accusation, prosecutors had not requested the kit be tested or the suspect claimed the sexual contact was consensual.

But the state decided not to test another 1,300 kits – nearly a third of the backlog – for which police said the victim either did not want to file charges or had shown a “lack of cooperation” after reporting the assault.

True-Funk said that at the time she was conflicted: She wanted to honor victims’ wishes but also raised concerns about testing based on how those wishes were summarized by police. People tend to disengage if they feel mistreated by law enforcement, she knew. They also change their minds.

The backlog is littered with such examples.

In one, a woman in Detroit told investigators that after her boyfriend had beaten and raped her, she stopped talking to police because she was young, afraid of retaliation and not emotionally prepared. Contacted after her kit was tested, that had changed.

“She now believes she has the strength to do it,” a detective wrote in the case file. Her rapist was convicted and sentenced to up to 18 years in prison.

In Iowa, her kit likely would not have been tested.

True-Funk said her concerns were drowned out at committee meetings, which regularly had about a dozen attendees representing law enforcement agencies and prosecutors’ offices but just a few advocates and one self-described sexual assault survivor. When the group later discussed how kits should be tested in the future, True-Funk said she was “shouted down.”

In 2020, Iowa launched an online tracking system where victims can see if their kits have been tested. Yet police departments still have discretion over which kits to test.

And the state lab discourages them from sending a kit if they already know the suspect’s name.

“Where the issue is consent and not identity, please consider not submitting DNA evidence for processing unless needed,” the state lab’s submission guidelines read. Several police departments told USA TODAY they follow the lab's guidance.

A spokeswoman at the Iowa Department of Public Safety, which runs the state crime lab, did not answer questions about the policy and declined an interview request.

Many of the state’s backlogged kits not tested under the grant program are regularly ticking past the statute of limitations, which for adult victims is 10 years. USA TODAY found that at least 500 kits have passed that mark in the years since Iowa completed its inventory, and that number will swell to 625 by the end of next year. But the statute of limitations offers another possibility. In at least some cases, once a DNA identification is made, charges can be pursued for three more years. So if Iowa were to test those kits, it might result in new charges.

The kits that were tested resulted in about 300 DNA matches, leads sent to the local agencies where the cases originated. State officials said they couldn’t require local agencies to keep them updated on what happened next but were aware of 53 victims who were contacted after testing and four cases charged, three ending in convictions. Earlier this year, police in Davenport, Iowa, charged two additional men in connection with a backlogged kit tested under the program, according to local news reports.

Officials at the Iowa Attorney General’s Office declined repeated interview requests. But in a statement, communications director Alyssa Brouillet maintained that the state cleared its backlog under the grant program. The argument, however, is semantics: Brouillet said that every kit “given to our office” was tested – “where there is evidence to be gained from testing … and the victim chooses to proceed criminally.”

She also said that local police agencies notify all victims before the statute of limitations on their case expires. But USA TODAY found that doesn’t always happen. Police in Iowa City and West Des Moines said they don’t typically make such notifications.

After receiving $3 million from the Department of Justice, Iowa had great leeway in deciding how to spend the money.

The program does limit sites to spending only half the money on testing, though even then, they can request an exception. Federal officials want funds left over for the critical steps that come next – supporting survivors, investigating cases and convicting predators.

In Iowa, like most states in America, all those needs are urgent, and there is never enough money. Victim advocates often burn out on the low-paying job in a few years. Rural hospitals lack sexual assault nurse examiners or several facilities may have to share one. Iowa defunded its dedicated sexual assault hotline years ago.

And yet, when Iowa finished its grant program, it gave up nearly a third of its funding, becoming one of dozens of grantees that together have left nearly $9.5 million unspent.

The $880,000 Iowa forfeited would have been enough to cover a weeklong training on investigating and prosecuting sexual assault for every county in the state.

Or teach more than 500 nurses the basics of administering a sexual assault exam.

Or test a thousand more sexual assault kits collected – then neglected – long ago.

Contributing: Dian Zhang and Savannah Kuchar, USA TODAY; Lee Rood, Des Moines Register

Tricia L. Nadolny, Nick Penzenstadler and Gina Barton are investigative reporters for USA TODAY. Jayme Fraser is a data reporter on the USA TODAY investigations team.

Nadolny can be reached at [email protected]. Penzenstadler can be reached at [email protected] or on Signal at 720-507-5273. Barton can be reached at [email protected] or 262-757-8640. Fraser can be reached at [email protected] or on Signal and WhatsApp at 541-362-1393.

This article originally appeared on USA TODAY: America tested 100,000 forgotten rape kits. Justice remains elusive.