At 2 a.m. on May 7, 1998, a Philadelphia 911 operator got a call from someone reporting that he could hear screams and choking sounds coming from the apartment across the hall. When police arrived 12 minutes later, they knocked on the door of Shannon Schieber, a 23-year-old doctoral candidate, but when no one answered, they left.

Ms. Schieber was murdered that night: raped and strangled by a serial offender. Although not his first victim – it would later be revealed she was his fifth – she was the first he killed.

In the aftermath of the Schieber case, it became clear that, due to a decades-old culture of disregarding sexual assault complaints, the Philadelphia police department hadn’t figured out they were dealing with a serial predator on the loose.

The story shook the city. A subsequent investigative series in the Philadelphia Inquirer revealed that the sex crimes unit had been downgrading thousands of rape allegations since the 1980s to the non-criminal, “investigation of a person” category, including allegations from two of the serial rapist’s previous victims.

The newspaper also revealed that the police department had classified 18 per cent of all rape complaints as “unfounded” – a term that means police have determined the allegation was baseless and that no crime occurred. It was the highest average among the 10 largest cities in the United States at the time, the Inquirer reported.

In response to the uproar, Commissioner John Timoney ordered a review of every unfounded and non-criminal sexual assault allegation going back five years. This audit, which revealed more than 2,000 sexual offences had been misclassified and not properly investigated, became the foundation for one of the most lauded police oversight programs in the world – a program that Canadian law enforcement is under increasing pressure to adopt. This week, as Baltimore County announced the findings from a third-party review – similar to Philadelphia’s – of police handling of sexual-assault cases, Canadian police services began to reckon with how to address similar systemic issues within their forces in the wake of The Globe’s investigation.

The Philadelphia Model, as it’s become known, has been running for 17 years. Once a year, a group of advocates and representatives from the Women’s Law Project, a legal centre in the city devoted to women’s and girls’ rights, are invited to go through the police service’s sexual assault files alongside high-ranking officers in search of deficiencies and biases. In the time it’s been implemented, the rape unfounded rate in the city has plummeted to just 4 per cent, contrasted with the American national average of about 7 per cent.

By comparison, The Globe and Mail’s investigation revealed that, on average, Canadian police services are dismissing 19 per cent of all sexual assault allegations as unfounded – just above what the Philadelphia police service’s rate was in the late 1990s. Reactions from across the country have been swift. So far, at least 29 police services have announced reviews, including the two largest forces in the country, the RCMP and OPP. But it’s not yet established how exactly police intend to proceed with their reviews.

Philadelphia’s model appears, on the surface, one that could be imported north of the border easily enough. The model has garnered support from advocacy groups, former federal justice minister Allan Rock, and prominent Canadian legal scholars who specialize in sexual-assault law. A little over a year ago, Ontario’s committee on sexual assault and violence recommended exploring adopting the model or something similar. Yet, to date, police services have been resistant.

Part of the reason is a tradition of insularity here. Whereas in the U.S., unfounded statistics, collected by the FBI, are made freely available to the public, Canadian unfounded statistics can only be gathered through individual freedom of information requests to each of the country’s roughly 180 forces. The Globe had to file FOI requests to every police service in the country to obtain data from more than 870 police jurisdictions.

The results point to a problem: the last year Statistics Canada released unfounded rates, the national average for sexual assault was 16 per cent and other violent crimes was 7 per cent, suggesting that, over the last 15 years, the percentage of unfounded cases has actually increased.

Sunny Marriner, the person at the helm of the campaign to bring the Philadelphia Model to Canada, says she is heartened by the fact that police services across the country are announcing plans to review case files following the Globe’s Unfounded investigation, but she cautions that if the system is actually going to change, police services shouldn’t be reviewing themselves.

“I’m really supportive of any community doing case reviews, but to be worth doing, to be successful, it needs to be the right kind of review,” she said. “You need the expertise of those who work on the frontline. Sexual assault centres. Advocates who deal with survivors. Only 5 per cent of women report to police. We still talk to the other 95 per cent,” said Ms. Marriner, the executive director of the Ottawa Rape Crisis Centre.

In February, 2016, the Ottawa Police Service became the first Canadian force to formally reject the model, citing privacy concerns following two years of negotiations.

Carol Tracy, the executive director of the Women’s Law Project in Philadelphia, who has been part of the oversight committee since its inception, said any civilian involved in the review signs a confidentiality agreement.

(While American privacy laws are less strict than in Canada, this was also the protocol on this side of the border about 14 years ago, when researchers Linda Light and Gisela Ruebsaat were granted access to police investigative files in British Columbia for a study of unfounded cases.)

“The Philly Model, it’s not that complicated and it’s nothing to be afraid of,” Ms. Tracy told the Globe.

Ms. Tracy says it takes about three days to go through all the files. A captain and a shift lieutenant sit in on the review. If anyone from Ms. Tracy’s team notices a problem, they flag the case. If the officers are not able to explain the problem, the commanding officers will send the case back.

She is clear that these reviews can be educational for both sides. “That’s one of the things – the learning process from us,” she said.

“I have a lot more understanding of how hard it is to do this. I think part of the reason that the captain likes this and I think by now the unit sees it as a positive thing, is that sometimes our eyes just see things that they don’t any more… we’re not there to do gotchas.”

She says that sometimes the problems are clerical – a rape kit wasn’t included in the file, for example, which is easy to fix. In the case where a suspect wasn’t interviewed, it may be that the issue was the lawyer refused to co-operate. Other times, the problem may reflect more on police behaviour during an investigation. In the case where an officer may be aggressively interrogating a complainant, the captain may choose to take corrective action but the committee isn’t involved.

The Globe attempted to obtain a sample of unfounded sexual assault files from the country’s largest 25 police services through freedom of information, by asking that the department redact all identifying information, including names, dates, and addresses. Even then, all but five services flatly rejected the request. Those that did send information included only a synopsis of the case and even those were heavily redacted, except for the Windsor Police Service, which also happens to have one of the country’s lowest unfounded rates, at 3 per cent.

Those who study sexual assault and the law say the secrecy here means police services have little incentive to improve.

Only when an officer is charged under the Police Act – a rare event in instances when the misconduct allegation pertains to mishandling an investigation – and the issue goes to a formal disciplinary hearing are details of the case publicly available.

This was the situation an Ottawa woman found herself in after filing a complaint with Ontario’s Office of the Independent Police Review Director.

In July 27, 2014, B.D. contacted the Ottawa Police Service to report that her ex-partner had sexually assaulted her a year earlier. She was warned that because the case was technically historical, it may be a while before an investigator could get to it.

On September 3, the constable phoned to say he would be starting the investigation soon. Then two months went by without any news. When B.D. phoned him back in early November, she was told the case had been closed. The detective insisted he had tried to contact her, but B.D. says she saw no evidence of this.

Galit Rodan/The Globe and Mail

B.D. convinced the officer to reopen the case but, the Ottawa woman alleges, over the ensuing three months, the officer lied to her, failed to keep her informed, and ultimately closed the investigation – again – without telling her, six months after she initially reported.

The then 34-year-old filed a complaint with the OIPRD, a civilian oversight body that investigates allegations of police misconduct.

The agency passed the case back to Ottawa’s professional standards unit, which is a typical process. By analyzing the police service’s internal telephone and cellphone logs, the professional standards investigators concluded that the constable had not tried to contact B.D., despite his claims.

Furthermore, the investigators determined that the notes the constable took during his investigation were “wholly inadequate,” a report on the probe states.

The officer was to face Police Act charges of insubordination, neglect of duty and deceit, but B.D. says she was told about a week before the hearing that he was retiring.

There would be no public record of the case at all had B.D. not released the professional standards report to local media.

“I’m left really upset that the hearing never happened. It wasn’t even about him being charged, I just really wanted to hear why he did what did,” B.D. said. “My complaint is more that the police didn’t do their job than the fact they didn’t lay charges [against my ex-partner].”

After B.D. went public, Ottawa’s chief of police told Metro news: “There is a level of frustration on my part that I cannot bring that member to account for his or her actions … we found that there was some misconduct. For me not to able to conclude that and leave that pending is frustrating for me.”

Without consistent and transparent disclosure of internal procedures, smaller investigative missteps are unlikely to meet the threshold of misconduct – a classification for which the bar is set quite high – and can’t be tracked as easily. And when a misdemeanour is identified, the disciplinary action can range from a reprimand, to mandatory training, to suspension without pay or dismissal.

David Butt, a Toronto lawyer who regularly acts for complainants in sexual assault cases, but who also frequently represents police officers during disciplinary hearings, said it would be unusual for an officer to be charged under the Police Act for mishandling an investigation. Mr. Butt said things like an officer failing to interview witnesses, or neglecting to keep a complainant updated, are more likely to be dealt with informally.

“There are best practices and you kind of fell down on the job – and then there’s misconduct,” he said. “Everything can’t be punitive. There has to be room for mistakes. Mistakes made that are teachable moments.”

Ian McPhail, Civilian Review and Complaints Commission for the RCMP, said that transparency has become an even bigger problem for Canada’s national police force after the passage of Bill C-42, the Enhancing Royal Canadian Mounted Police Accountability Act. The new legislation encourages informal resolution, which means none of it is entered on the public record.

“If you take a look at the internal code of conduct as to how complaints are dealt with in the new system, it’s actually much less open and transparent than it was just a few years ago,” he said.

Ms. Tracy said that for Philadelphia, shining light on the process helped everyone.

In the 17 years since the model was implemented, she said she has seen a “vast improvement” in the quality of the investigations. As officers came to realize that each case was going to be scrutinized, they took greater care to avoid mistakes. And on the other side, the officer’s superiors worked harder to keep an eye on things.

“I think it’s improved supervision,” she said. “When we first started looking at this problem and dealing with experts, one academic said to me, ‘There’s always talk about the need for training, they need training!’ He said, ‘Don’t let training be a scapegoat for bad supervision.’”

Robyn Doolittle is a reporter with The Globe and Mail’s investigative team.

Have you reported a sexual assault to the police? If you would be willing to share your experience with The Globe and Mail, please email robyndoolittle@globeandmail.com

If you notice an error, please send an email to datafeedback@globeandmail.com.