Moments before Palo Alto police officers fatally shot William Raff on Christmas Day 2015 in front of a Forest Avenue group home, they made a last-ditch request for a "less lethal" weapon to be delivered to the scene.

A video taken from a patrol camera shows Raff running out of the building and into the middle of the street with a table knife in his hands, and two officers ordering him to drop the knife. A third officer, watching Raff hop from side to side in the middle of the street, screams, "Go get the Sage!" to a colleague, who relayed the request over the radio.

The Sage, a launcher that fires "less lethal" ammunition designed to disable threatening individuals, never had a chance to be delivered. Moments after Raff ran out, he charged directly at the officers while yelling and ignoring their orders to drop the knife. Officers shot him four times in what was the department's first fatal officer-involved shooting since 2002.

Now, nearly two years later, the Sage is becoming a more common tool for police officers. According to a new report from Independent Police Auditor Michael Gennaco, who reviewed the shooting and the department's investigation and subsequent actions, the 2015 incident has prompted the Palo Alto Police Department to make the Sage launcher more readily available for officers.

The department's shift toward less-lethal munition comes despite the audit's determination that a Sage probably wouldn't have made a difference in the Raff incident, which unfolded within seconds. The weapon, according to the audit, "could not have been retrieved from a vehicle and deployed within the mere seconds that transpired before the subject charged the officers."

Both the department and the Santa Clara County District Attorney's Office already concluded that the December 2015 shooting was justified. In a report that was released publicly in May 2016, Deputy District Attorney Charles Gillingham concluded that "the totality of the evidence leads only to the conclusion that William Raff was intent on dying at the hands of police officers on Dec. 25, 2015."

The audit, which was performed by Gennaco and Stephen Connolly, did not contest this finding, yet in reviewing the case, he also considered areas in which he felt the department's administrative review of the incident fell short of best practices. As part of the audit, he found that none of the three responding officers had any less-lethal munitions in their patrol cars and that only one was qualified to deploy less-lethal munitions. This information, according to the audit, prompted the department to make a commitment to place a Sage launcher in every patrol car.

It also prompted all police officers to be "trained and qualified on the Sage" -- an initiative that has been completed. The council also approved the department's request for $15,000 to buy less-lethal ammunition. The budget document notes that as part of the department's "increased emphasis on de-escalation techniques," it has "increased the availability of Sage launchers from one per shift to one per car so that this tool can be rapidly deployed in dangerous situations."

Police Lt. James Reifschneider noted that at the time of the Raff incident, the lion's share of department personnel had received Sage training, which consists of a one-day initial course and then refresher courses throughout the year as part of regular firearms training.

Reifschneider said the Sage is typically used in pre-planned situations involving multiple officers, such as when the SWAT team executes a high-risk search warrant, or when the department receives a call about a person who is armed with something other than a firearm. The optimal way to use it is to have one officer deploy the Sage, allowing the other officer to then take the person struck into custody.

"It's a great tool, and we want all the officers to have it available if the situation presents itself," Reifschneider said. "But it's only applicable to a finite number of situations."

As a result, the Sage hasn't gotten too much use in the department. Reifschneider said the department has had a few different versions of the Sage over the past decade (the current one uses 37 mm rounds) and that have been fired on only a handful of occasions for calls for service.

Now, all patrol vehicles are equipped with Sage launchers, Reifschneider said. About half of them are equipped with mounts for the launchers. The older vehicles, which don't have mounts and which are in the process of being replaced, now have Sage launchers in their trunks.

The decision to make Sage launchers more available to officers isn't the only change that the department has implemented since the December 2015 shooting, which prompted then-Chief Dennis Burns to launch a systemic review of the department's tactics and training, according to Gennaco's Los Angeles firm, OIR Group.

Among the audit's key findings was the department's recognition that it needs to re-emphasize specialized training in crisis intervention. The report notes that the Palo Alto Police Department was the first agency in the county to commit to training all officers in Crisis Intervention Training and the first agency to voluntarily participate in training administered by the County's Office of Mental Health. The audit calls the department's decision to increase training "exemplary."

"I do think there was a holistic approach taken after the fact in looking at what we could do better going forward and want we can do in terms of getting our officers as many options as possible," Reifschneider said.

The systemic review also identified an interest in providing more scenario-based training to officers. The department obtained a firearm training system with the capability to simulate 500 scenarios, and instructors have "integrated their training to focus on tactical and force options such as de-escalation, repositioning, less lethal munitions and seeking backups."

The department also administered a four-hour de-escalation training to officers and decided to provide intervention, communication and tactics training.

While lauding the department's effort to review its policies, Gennaco also quibbled with some of its actions in investigating the incident. He took issue, for example, with the fact that the officers involved in the incident weren't asked to undergo second interviews as part of the administrative review (they were interviewed shortly after the incident); and the fact that one of the officers drove himself away from the scene, despite investigative protocols that require officers involved in deadly force incidents to be driven (the issue is somewhat muddled by the fact that it was the officer who fired the ineffective Taser at Raff who drove himself).

Gennaco also found that in at least one interview with an on-scene officer, the interviewer used "leading" questions ("Were you scared for your partner's safety?") that led to one-word answers ("Yes"). He noted that a question that "effectively includes its own answer, and only requires agreement from the subject, has the potential to undermine the appearance of objectivity and the legitimacy of response."

"This is particularly sensitive in the arena of officers investigating their colleagues over the use of deadly force, where the potential for public skepticism about investigative integrity is already high."