Public Safety Committee Special Called Meeting Transcript – 09/21/2020 Title: City of Austin Channel: 6 - COAUS Recorded On: 9/21/2020 6:00:00 AM Original Air Date: 9/21/2020 Transcript Generated by SnapStream ================================== [2:06:02 PM] >> Flannigan: There is a quorum so let's go ahead and begin. It is 2:05 P.M. We are on the virtual dais. I am chair councilmember Jimmy Flannigan calling this meeting to order. As we normally do, I always like to do a technology check-in at the beginning. I will mute you if you forget. Please don't take it personally. We have various staff in and out of the meeting in the waiting room as we go through our items. We have minutes to approve from three prior public safety meetings do. I have a motion to approve those minutes? Vice chair Casar moves, mayor pro tem seconds. If there is no objection, then those minutes are approved. Fantastic. [2:07:06 PM] Let's move on to our second item on the agenda, getting an update on the reimagining public safety. Take it away. >> Thank you, councilmember Flannigan, members of the public safety committee. I'm the deputy city manager pleased to present you with a quick update on what we've been doing the past few weeks working on reimagining public safetyment I'll start first with the task force, in fact. It is progressing, at the request of the task force we added the Texas harm reduction and we have met four times already since mid-august. Most recently we had a conversation and finalized a document on governance which included the task force request to now take some time and shift our meetings to be more public facing. And so we will start to have at our next meeting our task force meetings will be [2:08:07 PM] broadcast. We are working with ctm and with our coms folks to make sure that doesn't conflict with any standing council meeting or boards or commission meeting. And I'm excited to say that we will start out our next meeting with where the task force wanted to ground us all, which is in a history of race and policing. And so several folks on the task force, including Brian oaks, our director of office of equity, will be leading us in a really robust conversation about that as this group continues to move forward and shape out its priorities. We have not yet named co-chairs. I've had co-chair nominations and we hope to finalize that soon. Meanwhile, internally staff has continued to work on some of the things that are on its plate as we move forward. As you all know, the website phase one launched and it [2:09:07 PM] really serves as an initial portal. We will be updating that website. We will be changing that website. It will be iterative as we move forward, but wanted to get a place to launch as we started in that vein. I believe that Spanish materials for listening sessions as we move forward are being finalized this week and our web translation into Spanish should be as part of our next phase which starts now in October. So we should be getting that out as fast as possible. We had our first listening session this last Saturday, and thank you councilmember Flannigan for hosting that. I hope it was fruitful to you and to your constituents, but staff thought it was a really good discussion, inviting folks in a scenario based session to rethink when are the times they would want to [2:10:07 PM] call police officers and what are the other resources that they would like to have in a future facing world to address some of the different scenarios. We have d7, d5 and district 8 set up as well and we will continue to get confirmation on the rest as we move forward. Part of our communications plan also included having equity grants, many equity grants for reimagining. That request for grant application is going to run through the office of equity, and it is ready so we are just waiting final assignment and purchasing approval. We hope to use those to host topical listening sessions with organizations, not the video but these could be very specific about what does it look like to see -- certainly the history on race and policing but also [2:11:07 PM] how does law enforcement show up in parks or for youth or in different areas. So we are going to be thinking through and getting those grant submissions from community as we move forward. We also have the survey up and running in speak up Austin, and we'll be rolling out those invite to community partners to host listening sessions with fully translated materials for those as well. So our strategy is certainly continues to be robust. We'll continue to be built out and as others have suggestions or thoughts, we will certainly try to incorporate those. We have invited the task force to also lean into that conversation and ask anyone interested on the task force if they have suggestions for communications either for the city as a whole, city staff or for the task force activities to let us know. Hrd has been and other folks, obviously, but hrd [2:12:08 PM] has been meeting with those internal groups where we are looking at the administrative portion of what are those things that could be to use the term that came up in budget, that could be decoupled or could be thought of as transferred or done by civilian staff. Some work that is already done by civilian staff within the department. So we have met with finance and building services, permitting, transportation, the call center folks, so that included 911, ems, 311 and fire, forensics, transportation, code, fleet and pard. So the next step of all those conversation is hrd is providing the cmo reimagining leadership team with an insight into the issues of note, the time line, et cetera that we need to do moving forward. I believe they are trying to get that on our calendars either end of week or next week, so we're looking to learn from some of those conversations. [2:13:08 PM] We're also working internally to speak directly and do focus groups with APD non-sworn positions, so hrd has held administrative sessions. They are communications scheduled for October 2nd and our scheduling additional focus deep dive meetings with support services strategy, strategic support, community partnerships, victim services, internal affairs, special investigations and special events. Some of those we have already talked to at a very high level and now it's really about the nitty-gritty of who does what and what are the pieces we can parcel out. So all of that work is moving forward. I wanted to quickly provide an update on some of the work that pertains to the resolutions that have passed. And I should note too and should have noted at the outset we have our monthly update that is coming through assistant city [2:14:09 PM] manager Arellano is working on that and we will provide all of this to you in writing as part of that update as well. I wanted to say that all know not a resolution, an a pd initiative with the equity office, the assessment that Dr. Joyce James has been doing to look at this strategies and the culture of APD, that was expected to come forward at the end of August. That was delayed slightly in part because covid just makes it harder for us to do some of that work, but the core report is coming out by Dr. James the week of October 12th, and there will be some supplemental material, so we'll be able to share that with this council body in a matter of weeks. There's also the resolution 66 is moving forward in terms of the multi-pronged investigation of the extent to which forms of racism, bigotry and discrimination are present in the protocol, [2:15:12 PM] practices and behaviors of APD. The final report and presentation and its resolution form was expected in December 6th, 2021 or no later than. I'll say that work is moving forward. Fair are mu scodin who will be calling in and monitoring this meeting, she has been named the project manager or point on this resolution once a vendor is selected and informs me they are finalizing the terms of the vendor and hope to bring that to council shortly. We also have some work happening on the comprehensive review of metrics benchmarks and baseline measures relating to sd23 and zero based policy goals, resolution 50. Those recommendations are intended to come back at October 31st, 2020. Opm informs me that they had a large scale meeting with APD, opo, equity, [2:16:18 PM] innovations and working on creating data sets. We're aiming to have those measures in mid-october, and the director also said they are aiming to have their 2019 racial profiling report done in October and that is also part of that resolution 50. So happy to stand for questions, but that's just a quick high-level update of some of the work we have been doing and staff. We are mindful of the sense of urgency as we move forward. Know that we will be coming back and exploring some of the financial impacts and the administrative needs as we start to look through how do we shuffle things and still provide services and provide them -- and provide space for the task force to also reimagine some of those areas where some additional thought and some insight from community is needed. Happy to stand for questions. >> Flannigan: Thank you. [2:17:19 PM] I see mayor Adler is with us, also councilmembers Ellis, kitchen and tovo are in the meeting. We also have chief Manley and assistant director gay anbury oaks in the back stage if there are questions for those folks and I can bring them into the meeting if necessary. Any questions of our deputy city manager? I will say, you know, thanks to the staff for taking the first attempt at a workshop, public safety workshop for district 6. It was a really an invaluable experience for my constituents. It was much more of a workshop than a listening session, I think, because it was really about really taking some examples and kind of use cases to use a technology term and really going to play out how should we be [inaudible] Around these different scenarios. I thought it interesting to [2:18:20 PM] hear from community members that generally speaking we're in support of doing things in a different way or exploring new solutions and wanting to see that, but all of us acknowledging that there was a lot of work to do to figure out what the solutions were going to be. And I think we all know that because that's what we set out in the budget was this work that you just went through the long laundry list of very important and exhaustive in all the terms work to get folks engaged. What I find to be true is we're going to keep doing this work through the public safety committee, and I think these updates are valuable both for council and the public, but this is going to take as long as it's going to take and we're going to be sure what we do is right things. The mayor said this many times with the safety of the public is always going to come first. So we're not making despite some of the headlines, we're not making rash decisions, we're just identifying areas where decisions are to be made. I think it's a lot of good [2:19:21 PM] work moving forward and thanks to all the citizens and the community members participating in the task force. That is not easy work to get that many voices into a room. It's a lot more than 11 and we struggle when it's just 11 of us. Thank you. >> Thank you. >> Flannigan: Chris anything I need to add? If there's no questions on the update, let's go ahead and move into our third item, discussion of 911 data. I'm going to bring some new folks into the meeting, if you bear with me for a second. Okay. I think we have the folks we want to have in the meeting at thisoint. [2:20:24 PM] Kerry, would you like me to bring up your presentation? >> Yes, please. >> Flannigan: Okay. Everybody can see it just fine? Kerry, take it away. >> Okay. Thank you, councilmember Flannigan and councilmembers, mayor, Kerry o'connor, chief innovation officer for the city. For today's presentation on the call service data, the quick agenda we're going to do a recap and underscore the purpose of this analysis building off of last presentation. We'll offer some key insights and show you some of the dashboard, as well as [2:21:25 PM] some key opportunities that are revealed, and at that point we'll open it up for conversation and there's other dashboards we can review of interest. The tail end of this presentation is a long appendix of references and definition and some of those extra dashboards. So we can make reference to those as necessary throughout the presentation. Next slide. So the goal as we've been going through this presentation, and you can move to the next slide, are really to get as clear a picture of reality as possiblement as we dug into the data, we realized there's some process issues we need to understand about how this data happens in Austin at our level. And as much as possible, we're working to create a transparent and jointly held understanding of how the calls for service data works. And then we've set up dashboards to assist the [2:22:25 PM] Austin community answer their lines of inquiry as we co-creatively reimagine public safety. With this understanding and these dashboards, we're trying to do a bit of a self-service model on some of these data analysis, and then further lines of inquiry where we can bring that data back to the committee, the task force and the community. Next slide. I'm going to turn it over now to Jonathan, the chief data officer for APD. He was the one who instructed me on sort of the variability and reliability issues with the cultural service data so I'm going to leave it to him to do a quick recap on things we need to understand. >> Thanks, everyone. So working with Kerry has provided, you know, think about one of the key goals as I transition to this [2:23:26 PM] position which is about APD's data being transparent. We want to give you a sense of the complexity of the data because we have to have a very strong understanding of what the data are, how they come to be. And so we can really begin to understand how they can answer certain questions but not other questions. We also hope a greater understanding of the way [indiscernible] Exists can hopefully further the discussion of solutions while at the same time helping us figure out better ways to collect data and to be able to answer the kinds of questions that we can. So when we think about the data, a couple of important things to remember on the cad data is primarily about call data, what the public is calling in to inquire about, to ask for assistance and whatnot. The data changed throughout the proes is. When a call initially comes in to 911, an individual is [2:24:28 PM] going to share as much information as they can, the call taker is going to discuss with them and as you can see they are going to [indiscernible]. Sometimes that may be the extent of the information you have in the system about that incident. Other times the incident would -- the process moves forward and you can have dispatch. Dispatch indicates in need for service and we're going to respond. At that time this kind of incident type is going to relate to this initial problem code. A lot of times looking at the APD data, that initial problem code may be the extent of what's available. And so yet it's important to understand that there's a lot of uncertainty as to whether that's actually what's being responded to or not. So as officers arrive on scene in response to what's happening, they are going to be collecting information about what's happening on scene, and as a result they might update that problem code. That's going to give you a little better sense of this [2:25:30 PM] happening as well the kind of dispatch that it should have been updated in the system and we're going to mark the time they got there. But you can see we're already indicating in this process that what we understand the event to be is actually changing throughout time. This can continue to go through in the system where after the incident is over and the entire data collection has happened and the officer can make an interpretation as to what actually occurred, that disposition code can be updated again. So the reliability issue that comes into play when we discuss this is because this changes over time, we to understand at which point the disposition was coded in the system to give us a sense of what the value of that information is. So this relates to questions we would like to be able to answer when we -- we've been engaged in discussions about the types of calls for service that can be transferred to other organizations or maybe handled through other [2:26:30 PM] processes. And this understanding helps us illustrate the need for additional analysis to understand, but this information as it comes in may not paint a complete picture so we can begin to think about the kind of error rates involved in this process and making decisions at that time. So this gives a general sense of what happens in the process and how that relates to the data that are actually in the system. So the key points to understand from this starting with the first one is that a call for service itself relates to a huge amount of situational ambiguity. The average citizen calling in to ask for assistance is generally not educated about the definitions of various types of trends that we use as a police department. They are going to tell us what's happening and that's going to be interpreted, so there's a huge amount of error there. You know, that is -- runs into a second problem where this data system is built to try to manage an activity. [2:27:31 PM] It's not actually built to support research or analysis, it's not actually built to run a sort of process on to move in the direction of efficiency, so there are many things within the data that the way they are set up they are functional and they serve the operational need, but they don't actually serve the analytic function that could come thereafter to figure out what best practices should be. We see very often people looking for calls for service data, that's kind of a hyper focus on calls that have a large volume. What is it primarily that we are responding to as a department. But as you go through the different codes and look at them in great detail, there are a huge number of sort of small volume problem categories that are important. They can be important in the sense they relate to certain public safety problems, they can be very important in terms of the amount of time it takes officers to respond to those calls of service, and they can also be important in that they may reflect certain kinds of [2:28:31 PM] underlying issues that we might want to consider in the context of other possibilities for response. There are a number of categorizations, it's a relatively large number of categories that we look at, so we have to think about even these certain categories are small, there might be also very important to understand is that these data reflect what the citizens are telling, what they are telling the police department and their understanding of issues happening throughout the city. It is not really a proxy for current. You will notice in the notes that we have a number of different issues that don't exactly line up, right? So this can relate to that issue of whether or not information has been updated in the system. But also, you know, reflecting the fact that as the information comes in it is not -- it is not remarkably reliable as to what [2:29:33 PM] the event actually is. So to move beyond that we have to move into moving backwards from the date we have in our record management system, that system unlike the cad which tells us what the public is asking for what the what the public is groig the police department but primarily what the department is doing in response to this. And so there are two systems, a little bit different, meaning that the data in these systems actually answer very different set of questions so as we try to work towards transparency and sharing more data, it is very important that we generate an understanding that some kind of questions really have to come from one set of data, whereas other kind of questions have to come from another set . We start to understand the type of events where an officer wrote a report, there might be a citation issues or an arrest issue, clearly those are very important for understanding some aspects. It is also important to understand that we can use [2:30:34 PM] datasets to reflect certain challenges as we move forward with considering a new rms system for the department. Those are some of the issues we are looking at and so the final issue that comes in here that cause as great deal of confusion is that as they transitioned to a different reporting standard, it counts entire I are different and simply on that any single incident can be a variety of different -- crimes, right somewheres under the old system we are simply counting the top crime and that's all it is, in a Neider system we are trying to be more specific and capturing everything that happened in the event, our different systems, our different reports for reporting standards use one or the other classification and this can cause a great deal of confusion, it might issue a report but not reflect what is in a different dataset or might [2:31:34 PM] be a bit different than maybe what we are doing analyzing these issues. So all of that is very important to understanding in addressing those move in the direction to move forward nibrs. >> Moving forward in a more detailed way. So I will turn it back over to cherry to put this in context of the reimagining effort. >> , Thank you, Jonathan. >> Just notice here -- in the summary. We can move to the next slide. So this data for reimagining we want to be cognitive that the, of the situational ambiguity and talking about what officers think it is, and how the situation is framed, how things escalate or deescalate before or after arriving on the scene and how investigation on scene [2:32:34 PM] may change the nature of the problem category. We want to be cognitive of the data reliability, thinking about how this data is created and how the systems, categories have been changed over 16 years, for example I am told about 86 anybody call -- that of course in our data we have hundreds and that's because some of these categories are -- in the system or sometimes they have to add an extra D in order to administratively handle what appears on their screen. And finally we want to inform a future oriented exploration of what could be, rather than taking a strict -- validity. So by that I mean, when I look at this data I am thinking in terms of patterns, flows, journeys, orders of mag at this attitude, possible intervention points, potential high pop these and/or measuring the success of pilots rather than looking at [2:33:35 PM] specific percentage points. Because of the data variability and reliability there may be things that are generationally true or relatively true for which we don't have a specific point with very lower record rates if that makes sense. And so let's get to how we can then apply this for the next slide. In thinking sort of of the journey of the incident, where there is an initiation of a call, there is the dispatcher that calls, the call is seen, the incident gets closed out initially but then perhaps for some calls there is further inquiry or investigation and then you have the very long panel of the various criminal justice system processes. When we think about reimagining the kinds of ideas that have been before us have been, how do we prevent certain incidents? How do we transform how we [2:34:37 PM] happened certain incidents? How do we divert things out of a particular system? Or how do we make just decouple things completely from how the system is currently run? And then when we think about the supporting that that can inform this reimagining I kind of have tried to align here, we look at how calls are initiated or their volume or their times or their trends, to think about how we can decouple or divert. An example of prevention invention on the ems side is the health paramedics trying to prevent nonemergency use of ambulances, so we know we can use those processes in teams and strategies in different ways -- this is an example of diversion and most recently mental health cause is an example of diversion. So this is sort of to say how can we use this at that data in a reimagining process. This is a further suggestion. Next slide. [2:35:40 PM] Key insights. This is a visualization that looks at that call transformation process, where we start with ambiguity and then over time some of the incidents are updated and then we go to a disposition process and you can sort of see then how it transitions from the computer-aided dispatch to the crime report. So this is representation of 2019 calls for service and the 2019 crime report data. And from here you see 517,000 calls, you can see 65 percent of those are dispatched, 35 percent of those are officer initiated. Out of all of those types of calls, 35 percent of them get updated with a call problem a update and of those 517,000 calls 24 percent get a written report. Out of those 517,000 calls, [2:36:42 PM] 21 percent of them can be traced over to crime report data. We haven't traced that ourselves but proportionately this is where the representation is. And then four percent of those 517,000 calls are cleared by arrest. Other possible end point include a citation, warping, crash report, field observation. So this gives you a accepts of magnitude and flow of the dat through the system. Next slide. So orienting to this data our top insights are 41 percent of police officer time is spent on priority 3 calls. I will go through the priorities in just a minute but prior at this 3 is where property is not at life or property is not at risk and an immediate response is not required and this accounts for 51 percent of total calls. Since five percent of all calls for service are dispatched, where 35 percent are officer [2:37:44 PM] initiated. And I think this is an important clarification because if we are talking that about data we are talking about calls for service, not just 911 calls. The 911 calls are being represented by those calls that are dispatched. When we look at the top initial call Robbie volume for the dispatched calls, we are looking at the disturbance, alarm, burglary, burglar, suspicious person, doc, co violation which is any ordinance violations that requires a law enforcement response. And this is sort of thinking about those initial call problems as the information is coming in through the community members and how 0 that is, how that is get getting categorized. This data cannot tell you the experience of those calls. Either from the point of view of the minute member or the call taker and so that is something that we pay wish to dive deeper into in deeper sessions. [2:38:45 PM] Call categories for officer initiated calls are traffic stops, on sight incidents, checking in area, special assignment and directed patrol. Next slide. >> This is an example, this is a dashboard that is really generally looking at three and a half years of calls for service. And this is where you can see popping out in the upper left hand part of the screen that a large volume of calls are in priority 3. So priority definitions of course starting at zero that's that is where something is in progress and everyone is on the scene. It is the most urgent call for service. Priority one, perhaps something just occurred and people may be on the scene, it is still very urgent, but just a little bit from -- priority 2 is in progress, it just occurred, there is minimal or no immediate [2:39:45 PM] threat, provider at this 3 is life or property is not at risk and an immediate response is not required. Priority 4 is where the police are not required but documentation is .. You can see in the numbers in the top left hand thing that the numbers of calls for service in three and a half years priority 4 are very low. So we omitted them from the dashboard visualizations. Over in the bottom right-hand corner of the screen you can see a time spent analysis which we have broken out in two parts. The am response time is traditionally how we think of our performance measures for 911 calls for service. So we separated that out. That is the initial call time are from the time the primary officer arrives on the scene. There is the bottom average time on scene is all of the police officers primarily in backup, how long they spent on the scene and this is according to the different sort of fryer at this [2:40:47 PM] definitions. This analysis is an average and as we dug through the data, averages don't tell you a lot about how much time officers are spending on various things. Because there is a wide range of distribution is very great. If you click on the next slide we have offered a dashboard that really digs into how officers spend their time. We looked at the range of police officers assigned by call, sometimes there is a single officer, but other times there are multiple backup officers and so for this count of calls you can see you know 700, 42,000, had one officer respond, 607,000 had two officers, three officers, four officers and so forth. You can also dig into that data by priority call. You look at the average response time by range of officers, so if we think about a single officer, [2:41:50 PM] two, three, four, so we are thinking about bands of some incidents had five or eight officers, others had eight plus officers. This can give you a accepts of the affirm response time. And then down at the bottom this is the distribution and right here is a screen shot of the dashboard, but the link is in the bottom corner so if you were to link over to it which you don't have to do right now, but anybody can take this lichg and to look at this dashboard and see sort of where is the immediate kran time on scene for the different goings dents where we may have more than one officer. Up at the top right hand corner, you can look at it by type of initial call problem, so really we have given you the ability to dig into this that a number of different ways and this data is responsive so if you wanted to click into traffic stop, you wanted to cliff into new report, other pieces of the -- board would change. [2:42:51 PM] Next slide. >> And in insight we mentioned that around 65 percent of calls were dispatched versus 35 percent officer initiated. This is the dashboard where cow can dig into three and a half years worth of calls for service data and see the different patterns. And we mentioned that there were different initial call problems, whether dispatched or officer initiated. Again this dashboard is responsive if you were to click on dispatched calls the entire dashboard would turn blue and you would see the responsive information, including the affirm time spent. Next slide. Average time spent. .. >> We wanted to get a sense of that call problem transformation and here is are we start to read into where are those areas of further inquiry, where this data can potentially be useful for a cocreatively safety rose. With enwe look at disturbance other, which was one of the top [2:43:53 PM] initial call problem categories for dispatch calls, 49 percent of those calls were updated. And this cloud gives you a sense of volume and magnitude as well as the long tame of types of calls that they were updated to. So perhaps the caller, you know, initially said, there is something going on. And either that call escalated, deescalated for true nature of that incident became known. As we were working through this data, it was noted for instance the other day there was a call problem that was a disturbance other, someone said that their boyfriend was making a scene in a restaurant and later that scene exited and went to his car and turned into a suicide. So we cannot know how these [2:44:53 PM] calls are transformed, but retrospectively we can look at the data and we can see what were the patterns in our community and for the three and a half year dataset, family disturbance, criminal trespass, assault with injury, were some of the larger call transformations that happened. Next slide. >> 29 percent were suspicious call programs call categories were updated to these kind of call problems, suspicious vehicle -- complainant, criminal trespass. Again if we wanted to dig further in, we would grab parameters of the data and we would do a deeper dive to understand the true nature of some of these categories. And how they might lend themselves to a reimagining [2:45:53 PM] process. >> Next slide:55 percent of urgent calls were updated to these categories. Edp intervention is -- disturbed person, deceased person, assist ems. Were some of the call problem categories. Next slide. >> This is the updated dashboad with mental health calls. You may recall from the last presentation where we only had one month of data, now we have six months of data where all of the other dashboards are three and a half years, the mental health diversion really startled in earnest at the beginning of this calendar year. Remember here are relative, the numbers here are relatively small in the grand scheme, ten percent of all calls were flagged over the past six months as having a mental health component. [2:46:53 PM] But I would also just honor this program is just getting started and as the program goes on, we will be able to add more initial call problem categories to be screened for potential mental health flags. And this is is an example, councilmember Flannigan where I would like to click on the link to go to the dashboard. So here, if we go to the counted calls by initial problem category, one of the key areas of interest is when we talk about diversion of calls, if you scroll into that little top quadrant where you have the call problems and it starts with check welfare urgent if you put your cursor in there and scroll down a little bit you will come down to one pretty shortly and it will say ccd counselor [2:48:03 PM] . >> Maybe up just a little bit more, I think it is in the 300 raping. There you go. Ccd counselor is a Trudy version, true, diversion category .. 328 officer initiated. >> -- 328 mental health calls, went to a counselor, not an officer. And this is an example where we really have to ups administratively how this data works in order to pull out the measurement of a true die version. But as you go through this dashboard .. You can see the type of initial call problem categories that are being addressed as well as the time seen, the range of officers, and you can see a relatively stable call volume month over month. >> We can go back to the slide [2:49:04 PM] presentation for a moment and then I think we are about ready to open up for questions and conversation. >> Out of this initial deep dive into three and a half years of worth of data, we see some key areas of opportunity. We can deep dive into specific initial and updated call problems. Nealtively we can go back in the other direction as Jonathan had mentioned and we can look at the rmf data and work backwards and say how did this call start, how it began and how it started, anyway it started with the call and working with the system or working in the end and see how it started that deep dive can look differently by different types of inquiries, if we were to do a deep dive inquiry to traffic stops that would look very different than a deep dive into the service other calls. >> By deep diving into the data we can inform the pilot of [2:50:04 PM] alternate responses, not unlike a lot of the key work that has already been done with the mental health slide. And then we need to be able to strengthen our data collection and analysis, but we will note that there are chance in undertaking that which underlies a real change management rose that will ultimately, Jonathan is here as the chief data officer. Next slide. Here, there are other -- words where you can just doif into initial call problem category or false alarm indicators. And in our next steps we sort of indicated what we said we would do on August 6th, where are we today in terms of what we have achieved, and where would we still like to go with our analysis, we need to do more of cleaning the location base data, once we do that we can look at [2:51:06 PM] gentrification map and sort of a service correlation that maybe happens over time. We want to do journey maps, the and we still want to have a conversation about not just how to represent this data in a dashboard on tableau but how do we have open data, you know, in different ways that they represent this data in an open data format and we can have that conversation here in Austin as well. So with that I will turn it over for questions and the any commentary on this data and the next step. Thank you, Carrie, really fascinating information in this deeper dive and glad to see that first month of work, first month am really expanded into something really substantive. If there are questions for Carrie that you want me to reference to keep the slides up, [2:52:07 PM] pain we can do those first. Or if there are any other questions we can better see each other. >> Anybody have questions for staff? >> Councilmember Casar. >> Tovo: I do. I don't think I,. Ca. >> Casar:, I can do that, first of all, .. This is really helpful and more comprehensive. Comprehensivewhile being deep but also self explanatory in many ways .. I didn't word that as nicely as your presentation but just a really nice job thank you. And I am glad to see that evolving so there is a lot there and I think there is a lot that maps out that we can work on in the future. There are two areas tt you identified in your next steps that are areas we have been talking about for for a long time though, because they stick out around burglar alarms and [2:53:08 PM] mental health calls. It seems like you are going to do continued work in those two areas. I know there are a lot of areas you have to work with the reimagining task force moving forward but on those two it seems like you already started to dig in, what are we learning from this continued analysis that we should know about or what ideas are you starting to have maybe alongside talking with to other people about those two you know big pieces of calls where we might be thinking about doing things differently? >> I think that for the burglar alarm, it might be useful to bring in -- for the mental health program have that diversion, how that diversion is going, I believe Ken Murphy can speak to some of that on going work. >> Casar: And, Carrie, more than -- absolutely I want them [2:54:09 PM] to speak to that, I guess, from the data, what are we learning that is new? Or what are thing that are new that might help us in, informing those conversations with integral care and the police department and all of the people implementing this, but just from the -- now you have looked at this so much more carefully, what is sticking out to you on this website? >> On the mental health calls what is picking up to me, we had a conversation with bj from the -- institute who with was the initial report that you all received in 2019, and the categorization is a bit of a challenge. When you look at some of the categorization of mental health calls you might see known or unknown mental health calls, well to me, no or unknown are two different categories, but they have been lumped into the same category, which makes understanding like if I know that there is absolutely no mental health flag, then I would [2:55:10 PM] say no. But if I am not sure, I might say unknown. That is an example of where the categorization can be a bit of a challenge and it sort of speaks to how some of our data is treated administratively, maybe not necessarily with analysis in mind. Knowing how to hunt for the counselor to say this is 100 percent tried and Trudy version .. I wouldn't have necessarily looked in that area in order to find that data signal. So thinking about randomly do a diversion program, how is it we are category recognizing it so we can measure it so we can understand our progress I think is going to be really important going forward, and that is one of the take-aways that I have here. In terms of burglar alarms, one of the things we are waiting to do is clean up the location base [2:56:13 PM] data so that we make sure that we are being accurate when we present this to you. But we are looking at perhaps doing at least an initial cursory count, similar to how we did the time spent analysis, where we are like here is the volume, you know, repeat, longitude and latitude is happening. Where perhaps we can see a curve where if we have this information where after three false alarms you have a fee to pay we can hopefully look into the data and say well what is the distribution of calls? Are we seeing a large distribution at 3:00 or at four calls or are we seeing it drop off, you know, and we can start to then go in and assess how that fee program is working in order to prevent false alarms, but without really having [2:57:14 PM] cleaned to the location sort of data I would be worried we weren't giving you accurate information so that is the next step on that. >> Casar: On the false alarms we talked before about, I am not sure if you were there or not, about seeing residential versus commercial alarms because those are different systems. >> Yes. >> >> Casar: And then figuring out what percentage is false in either case, just because they are kind of different things S that something you have been able to actually look at that or is that. >> We have to pull data from a different system to look at the commercial properties. I think part of the -- comparing it to permitting so we are make sure we are comparing the addresses and it is a commercial property. >> Casar: Are those 100,000 or 90,000 or so you showed those are commercial and residential combined? >> That is correct. [2:58:14 PM] That is all calls. >> Casar: Great. It would be useful to know if there is more false in one category than a another. >> Yes. >> 80 percent of them are residential, you know, that would be really useful. >> Okay. Thank you. >> My apologies for not being able to answer that to you. This time around, if we were to have a data cleaning problem, but I promise you we are not going to forget it. >> Casar: That's okay. I forgot if I mentioned it to you or someone else. And of course I am happy to have APD speak to that, my questions are really about the data so, of course, chief or anyone else wants to go over that but I feel I have had my questions answered here. >> Mayor Adler: Councilmember alter. >> Alter:, thank you I am trying to find the slide number now. I think it is slide 14 I want to say, where you have the circle with the officer initiated calls [2:59:16 PM] of 35 percent. Let me bring it back up. >> Mayor Adler: This one. >> Yes. So I wasn't -- the count of ca calls by prior at this, by that color, so it looks like almost all of our officer initiated ones are priority 3 calls. And that is just really interesting and obviously, you know, there may be some way to papping how they are spending their time if we need to get to the priority one, 0, 1, and 2 and I understand this pay be standard within the police protocols, but that that doesn't really speak to me. I am just wondering what we could dive down into there to [3:00:18 PM] understand better the officer anybody indicated calls and the choices that are being made. In particular, it looks like a traffic stop for, are officer initiated and considered priority 3 and that's what I heard at all time from my constituents they want more of. So I am just a little -- I would love to know what you have to say about that in this data:sure. And I would like to perhaps turn it over to Jonathan or lieutenant Murphy, but from my perspective looking at this data, this is how the calls get started. And the nature of how calls get started by the community members or by officers you can also see the patterns where traffic stops, on sight incidents and special assignments, those have higher volume of officer [3:01:19 PM] initiation, whereas what the community is calling about is, you know, disturbance, other, alarm, burglar, check welfare, suspicious person, code violations, that require a law enforcement response, and I think that when you look at the dashboard it is really, what is the nature and characteristic of each type of call? Like if you want to get into the process, then certainly we can look for chief of staff can tell you how do these things kick off and how do assignments get made? That's not where we started how this dashboard works. It is more a retrospective on what happened, what are the volumes? How were they - - how were the calls concluded? , Darby sector, for example, and when you click on the link to, this you can actually -- and perhaps, yes, it is worth doing that. You see how things change, so if I were to click on officer [3:02:21 PM] initiated, the whole dashboard changes, just to show you exactly those trend lines -- and if I don't way tonight go back to normal, just click on officer initiated again and the dashboard resets. If you were to click on say the George sector, down below, you will see categories of calls, downtown area, you are going to see a really different sort of fingerprint, if you will. Of the types of calls that are -- that are happening there. If you click on George again it will reset and reset to normal. If you want to know sort of what are the kinds of calls that are getting a report written, you could click on report written and this ten dashboard will change again, and so I think it is really important to think of this dashboard as a retrospective of what happened rather than something pro descriptive of what is [3:03:22 PM] happening. .. In order to thing about those volumes or those patterns proscriptive tell me about the 41,000 calls that had a report written about them I would like to know about that subset, what is going on there. That's the kind of line of inquiry we can take from here and start to dig into with Jonathan's help and lieutenant Murphy's help to provide any insights that are helpful to think about, you know, how we wish to do things differently. >> >> Alter: So when there is a no report written -- this pay be more for APD. What does that mean? Like more than half of the calls -- it looks like more than half of the calls are officer initiated, have no report. >> And I will call attention to [3:04:25 PM] all of these very same questions. There are a lot -- there is a lot of lingo here. On -- 28 and 29, 27, 27 was all of the definitions of the different dispositions. 28 says, you know, what triggers a required for a report to be written and 29 says what are the different ways an incident might not require a report to be rin/written. And so we can go in and learn a little bit more about that. So one of the things that we try to, do because I know is a question that we all have, is -- it says report written mean that there was a criminal offense? And our data didn't link up very well. I think we mentioned we had -- computer-aided dispatch, and we [3:05:26 PM] are trying to take the master incident number and link the two together, say what does this tell us? And the numbers weren't quite limping up for us so we had to do some transformation on that, but we found wane a reason to believe that not all reports written equals a criminal offense and some of the report not written was represented in the crime data, so we know there is some variable there and we haven't finished answer what that looks like, but that is sort of part of the next step of saying, we start to understand how these calls are transforming. We don't have the full understanding of how it traces to an outcome, but that is where Jonathan was suggesting we can start backwards with the rms and understand how things start. But I think there is a reasonable proxy that if a report is written that there was [3:06:28 PM] something suspected about a criminal offense but also see that crashes involving bicyclists are simply -- get a report written about them. So raps Jonathan or the lieutenant might want to say more about that. >> I would just say, generally, and, you know, you started with a very important point, that looking at traffic enforcement and I think Carrie has tried to make the point that, you know, the dispatch data really shows you what is coming in, what is coming in via the 911 system and the officer initiated is sort of an entirely different set of data. But you might want to link this to a higher level discussion about the various roles that the police department plays. And so we might think about this. I might kind of generally categorize it as the response, [3:07:28 PM] and so those would be the dispatch calls, fanned other area, the officer initiated we have a number of different things like traffic and any other part, generically would be something like patrol, right? So something would be counted on patrol. And as you think about reenvisioning public safety, I think the data helps to the extent that they help you understand the magnitude of each of these activities, right? And so, you know, as you talk about traffic enforcement being important to your constituents, the calls being a lower priority as far as response goes, that is because we really should think about response, control and traffic enforcement assort of three different very, three different, very different funks so it could be helpful to [3:08:29 PM] organize the kind of thinking along those lines. The officer initiated doesn't in and of itself give you that kind of breakdown but you get some sense of that at looking at the officer initiated, a subset of that by deliberating on officer initiated calls, and then you can see on the left-hand side of the dashboard the number of traffic stops, you see T rest of the categories that are there and that kind of helps you the breakdown but it is important to recognize as Carrie said, you know, this is kind of a picture of, it is kind of a snapshot that gives you some sense of how things are starting and yet at the same time it is also important to think about, you know, conceptually as well the functions of policing, you know, and that might help with the discussion about what they should be. [3:09:30 PM] >> Can you hear me now okay. Carrie. Maybe it should have gone to ems, to fire, I don't know if we can just send it to mental health. When the call was initiated by police, it is not going to help me understand unless we do a deeper dive into this particular issue to understand if there are certain things that could be rerouted elsewhere; is that correct? >> There is not telling me anything about ems or 911 or if something got routed to the wrong bucket beyond what is in the police bucket; is that correct? >> Yes. And it kind of speaks to, you know, it does. This is sort of what is happening in the -- and that, which is a very important discussion about what are the calls coming in and perhaps which one could be rerouted a [3:10:31 PM] handed differently, you know, it points to something that I at this Kerry tried to illustrate earlier which is the various categories of information that you see among the initial information set, they are not fully exhaustive and really don't capture every possibility. They are not mutually exclusive. So as you can see with the -- it is something seeping very different but ultimately it turned out to be the same thing. And so, you know, we can look at the initial incident type information among these calls that are coming to the department and we can to the extent possible with the data systems, look at what those ultimately turned out to be and what the actual outcome of that was. Was there a citation? Was there an arrest? And give us some more context for what happened. And that is something that -- but it is important to note this [3:11:32 PM] is not getting a picture of -- outside of calls that are 384 calls that are getting diversion of the program, this isn't giving a sense of what just happened had these been routed elsewhere. And it can be difficult to know, one of the things we hoped to do on the list to follow up on is to share with you a real presentation on the day in the life of an officer to give you a sense what does it mean when there is not a report written this what does it mean there was no citation and no arrest is because very often that doesn't necessarily mean that the -- that the response wasn't ultimately a part of the solution to the incident that merited the call in the first place. Council member John Murphy here. If I can just explain a little more about the routing so when a person called 911 the first is [3:12:33 PM] do you need police, fire or ems and soon mental health services. So if the caller said I need ems, it is, we a automatically conference in the ems dispatcher. And/or the fire dispatcher if they say fire, if they say we don't know then we continue with a line of questioning to try to figure out what actually, what services they actually need. And if it goes on and say we are dispatching officers, but then things change, then we can still conference in ems. So all of the calls you are seeing referenced in the dashboards they are all police calls. They do not include fire or ems because they have a different dataset. We are in siloed datasets so this is all police. Counts at does that, councilmember alter does that clear up your question? Yes. Okay .. I want to jump into the [3:13:34 PM] category or priority 3 calls. And especially the officer initiated calls, because when we talk about officer initiated there is no response the because the officer sees something and they initiated initiate the call like a traffic stop or a subject stop. And so there is no response. So when we talk about officer initiated it is what officers are seeing when they are out there on patrol and they are calling in and skagg call to be made for it now. A vast majority of those priority 3 calls when we talk about special assignment, because there was a large number of special assignment until recently a year ago, special assignment meant secondary employment, when officers were working a secondary job as security somewhere, either for capital metro or they were working a UT game or they were working one of our festivals in town, because every other officer has to log on when they are working a secondary employment job, it fell under [3:14:35 PM] these special assignment categories. A year ago we separated them out and now we have a secondary employment and a special assignment meaning now the special assignment is only if they are asked to do something special on patrol, instead of just regular patrol and responding to calls, they need to focus on one area, maybe a hot spot where a crime, crime has been occurring. And they are under a special assignment so that is what we consider special assignment. Now, also, there is a high volume of those priority, priority 3 calls that are directed patrols which again focuses the on hotspots where our analysts say hey we are having a spike in burglary of vox in this area or having a spike of burglary of nonresident in this area officers will put themselves out on a directed patrol in the area when they are not responding to calls to check, to check those areas and do a high visibility patrol. Does that help clean up some of [3:15:36 PM] the priority three? >> Calls and why there are so many? >> . So much good information in this dashboard I just want to keep clicking around. The officer initiated stuff is so interesting because it really does get into a level of detail on the job itself that is very hard to represent in the data. That might be an interesting conversation, chief can think about, how to talk through what that really means so that we can better understand. I would hate for like this no report thing being reduced down to there wasn't a fine when we are hearing that is not what that means. And we certainly as a council said sometimes you don't have to criminalize people, but the response was still needed, so, you know, there are a lot of [3:16:37 PM] detail and nuance in there. The one that I wanted to ask about was, the officer initiated, this category of on sight incident. Do we know what that means? Sorry if it was in the presentation and I missed it. Councilmember member, Kenneth Murphy here again. When an officer sees something and they call it in over the radio and asks the dispatch tore create a call for service, most of the time dispatch her put them as an on sight incident because they don't know what it is yet. Fllt the officer clarifies and figures out what is going on on the scene so they may see two people arguing and say put me on an on sight at this location. And until he gets out and actually investigates what is going on, it may be a family disturbance and then the call will change to a family disturbance. And maybe your report was written, maybe your report was [3:17:37 PM] not written, however, you know, a policy that is -- they should Wray report. You know, so -- and ten, going back to reports, no report doesn't always mean that something wasn't documented because while we are not capturing in the no reports are citations, we are not capturing crash reports and no reports, and we are not capturing field observations. So which is also included in our rms systems. >> Does that make sense when you dig into the and George has by far the most on sight incidents you might expect a more density of people, asking for engagement. >> Yes, sir. >> Every other part of time that frankly a lot more spread out and people are more distanced each other, it may two way town. >> Correct. You are absolutely right and that's why we see, like you said that's why you see a cohigh count of on sight instances because we have a higher density [3:18:38 PM] of police officers in a smaller area and historically before covid a lot of things happened downtown on Thursday, Friday, and Saturday nights so there were a lot of on sight incidents by the officers controlling on bike and horse and foot roll. Yes, sir. You are correct. >> It is interesting you see the officer initiated take a big spike, right around 11:00 o'clock. >> Yes, sir. That's usually downtown, when we see the spike in calls and, you know, you have fights, you have little verbal disturbances going on between 11:00 and 2:30 A.M. As when we see spike in the downtown area of on sight incidents precovid, right? >> Yes. It is almost interesting the spoke happens more at 11:00 and less at 1:00. Just think about how it is an interesting little data point to see on the chart. >> Yes, sir. >> Other questions? >> You may have answered this [3:19:39 PM] already chairman but is there a link to that sight and is the presentation -- I didn't see it in backup. >> I don't think, Kerry, we put it in backup yet but -- >> I just sent it to Katie powers earlier today so it should be arriving there and the presentation has the lichtion, so every slide we see a -- board there is a link there, I think to that dashboard. And then some of the top problem categories are also found in the appendix, I think the top 30 or 40 call problem categories. And provided the definition back there, so a lot of that is in there. >> And city manager, it might be valuable to have these links on the reimagining public safety pages that the staff has created. >> We plan to do that and send it to the task force directly. >> Excellent. There is so much detail in this -- >> Go ahead. >> Go ahead. [3:20:41 PM] >> One final thing I might note too is, when it comes to really digging into some of these calls, we do have some really innovative officers that have created community observation or critical observation in the community, a thing called -- there was a presentation given I think to the public safety commission back in the, before covid, not that long ago but -- where they would take tease calls for service and they would analyze them by the trigger that was present, trying to look for a root cause to try to call upon a more sustainable response. That initiative not out to two different police sectors, it doesn't have a lot of data associated with it, but as a lot of folks have explained to me, you know, if a gun is present, if a mental ill seasons present, if alcohol is present like there are ditch triggers that are happening in a particular scene and they could keep track of [3:21:43 PM] those triggers and try to get underneath some of the trends that they were seeing, we hope to dig a little bit deeper into that initiative to see what we can learn from their persons in trying to do that, to even better understand some of these trend lines that we are seeing in the calls for service data. >> Thank you, Kerry, I will be glad to get this up on the city side too. Such density of data that a little bit of community crowd sourcing can be useful too. Councilmember Cass. >> Casar: I want to thank you all, before we close because we went so deep into things, it seems like a lot of the focus and conversation we have had recently also is validated by some of -- because of course we to get important phone calls about assaults or armed robberies, but I want to just reiterate this because a lot of folks watching may not have the main screen up, basically what [3:22:43 PM] front of the dashboard seems to show is, that two-thirds of the calls or of the police, majority of police time goes to going to responding to 911 calls of which the big 3, top three incidents are disturbances, which you said, post likely later get coded as family, as family incidents, appeal violence, other family interpersonal disturbance, second, burglar alarms and then third welfare calls that more often than not are then later coded as having to do with emotional distress or mental health. So those three areas of burglar alarms, mental health and family violence issues seem to be the main things driving the things that police are spending most of their time on when they are getting dispatched to calls and then on the officer initiated side, which is a smaller portion [3:23:46 PM] of their time, the number one thing is traffic stops, which were all, we are all talking Y talking about and thinking about and second to that is those incident of camping things as you go on patrol, which seems to most heavily be happening downtown. So that seems to me to be like the overarching stuff, but I think a lot of what this council is looking at on the right track of looking at the big drivers of police time. >> Councilmember tovo, I see your hand up, your video comes in and out. >> It is frozen with my hand up. >> Tovo: I can't control it for some reason. But thank you for -- >> So I am really looking at this for the first time. I think it was paid out but hot -- I think it was for system reason not on mailing list -- and I haven't had a chance to really dig into this data. I want to thank you all for it. [3:24:46 PM] This is extremely useful analysis and I also would like a moment to really thank our dispatchers, I know you have had a -- it is pretty extraordinary, the level of focus, the level of focus and experience and just ability to manage a very stressful situation that all of our 911 and 311, are balancing, of course the work is just extraordinary so thank you, thank you to all of our police who do that. It is really important to us. Kerry, I had a couple of questions for you and I think I have one for our police department. As we look at were you able to make any -- you know, as I look at the number of reports and really trying to understand that situation where reports -- to those of you who have emphasize this doesn't necessarily mean it [3:25:48 PM] wasn't recorded in some way, because they were in separate categories where -- a crash report might have been written but not a full report, are there other times where especially downtown where officers pick going from one situation to another situation to another situation that could also account for their being a lack of a full report written? >> And I if that's sat question for the police department, but -- >> I think it is a question for the police -- >> >> Tovo: I would be interested if there is any geographic -- the extent to which geographic that you can see a yes graphic trend with regard to -- >> To my understanding, I think it is to perhaps, I will go to lieutenant Murphy to make sure I say this correctly, officers, if they have to fill out a report they have to fill out a report. If they can do it in the moment, [3:26:48 PM] they will do it in the moment. But maybe they have to go from incident to incident to incident and maybe they don't get to the report until a later time. Incidentally, this can also relate to a variation in the times spent on the scene. You know, if they weren't able to clock out and then they clocked out after they finished the report, that can introduce some variability, but my understanding if a report has to be filed, a report will be filed, if it can't be filed in the exact moment it will be filed afterwards. Did I do that right, lieutenant Murphy? >> Yes, ma'am. You are correct. According to policy, if a report needs to be written it must be written before the end of the offer offer cease, officer's shift so they may jump from call to call or, you know, a primary into a backup situation and is especially downtown, however the reports are written before the end of their shift. [3:27:50 PM] >> Tovo: As Kerry said it could introduce some variability in terms of how much that gets reflected, that scenario gets reflected in terms how much time on the scene is registered, correct? >> Yes, ma'am, that is correct. >> And then I think you or one of your colleagues would be the right person to help us understand or help me understand what assist complainant, what are the range of things that are within that? I notice that as a larger or it repeats itself. >> If you could five us some examples of the range of things from across the severity sector. >> System, a -- complainants are normally very low on the severity scale, if you will. They can range from anything from assisting a person to a motorist an officer seeking motorist changing a tire. And an officer being waved down [3:28:51 PM] to provide directions. Or information. So there is a wide range, and right off the top of my head I can't think of anymore than those. Those examples. >> However, it is something where a report does not need to be written and it doesn't necessarily fall into any -- any of our other cat title codes, if you will. >> Tovo: And so, and actually just a couple of days ago I had the experience of riding behind an officer or riding behind a scene where an officer was doing just that, changing a tire, changing a tire. But really, noting that on page 16, it says 29 percent of suspicious person incidents were updated to reflect -- complainant. So that is one of the reasons I asked my question, it seems to me that if people are calling about a suspicious person or [3:29:52 PM] being dispatched, officers are being dispatched because of a call about a suspicious person, what does that mean when they are reclassified or classified as -- so maybe bears for info about it, maybe not necessarily right now. >> Yes, ma'am. It would be really difficult to tell without doifg into doing a deep dive into each call. There was a suspicious person and then later change to an -- complainant, however right off the top of my head, I can think of an example, maybe someone seeds in their neighborhood a vehicle driving around or somebody standing in their neighborhood, they haven't seen before. And they call -- and we title it a suspicious person because we don't know what is going on yet and the caller actually says, this person is suspicious and well why is it suspicious? We haven't seen them before in the neighborhood and they are just driving around or walking around and then officers get on the scene and they are watching for someone or they are lost, you know, and so then they give [3:30:54 PM] the person direction and change it to an assist complainant from suspicious person, does that make sense? >> Tovo: Yes, thank you for talking me through it. >> In 171 of the things especially as we try to make sense so only of the, to some of the broader trends I want to ask a couple of detailed question questions about this. Edp, edp intervention or emergency detention. To what extent, to what extent do some of this data reflect overlap? >> I am just struck by the fact that some are updated to be emotionally disturbed person. Versus emotionally disturbed person intervention versus emergency detention. Which I understand is in a classification of its own but -- >> Are we -- is it possible some of this data is also over language, overlapping? In other words, a similar [3:31:55 PM] incident might be classified one as an intervention versus -- >> Yes, ma'am. You are correct, the detention is one, stand-alone, because that's when the officers take someone into custody for their own protection. >> Tovo: Which there are very few professionals who are under state law able to do. >> Yes, ma'am. >> Tovo: One of those two categories. >> Yes, ma'am. Absolutely. However, just the edp title code and the Ed. Intervention, it could be one and the same. Depending, again, you know, so it is where you didn't take into someone into custody yet they wrote a report so when one officer may title he adp intervention one may title it just emotionally disturbed person. So it can be one and the same, and that kind of segues into just I want to let everyone know, the councilmembers know the snapshot you saw for our ecd [3:32:56 PM] councilor on the screen and the numbers you saw, remember that is just a senate shop in, snapshot in time of the first six months of the ram. Since then, just today, we have had over 800 diversions, so it is a far cry from the 300 we saw in scrup, because June is when we really started to escalate the program, and we included the title codes of check welfare and check welfare urgent as possible diversions. Instead of sending officers, and so we have seen a significant increase. And in August, we actually hit our benchmark, which was 200 diversions for the month, we actually went over to 225 diversions, and just a segue into what is possible, when I looked at when a counselor was unavailable, either on a call, out for lunch or they weren't on duty, there was another football 157 calls we possibly could have [3:33:56 PM] transferred to the counselor, which would give us for the month a total of 382 calls and 82 percent diversion rates mean we would have approximately 318 Trudy versions. So we are heading in the right direction and I think we will see a significant increase in these diversions once we are able to have a counselor on 24/7. >> Tovo: Thank you for that additional information. That is really very helpful. Did you say 300 diversions. >> To date, to date, to date, yes, ma'am, since the start of the program in December of last career. >> We are over 800, 800 -- thank you and thanks for the information about the potential overlap as we are trying to get a sense of the picture of the calls that could potential be -- >> Thinking about that is helpful. I think I have just two, one quick one and one possibly, tell [3:34:57 PM] us, sir, teleserving is when someone calls 311, and it is politician 7 trick, they place a customer service request in a queue and then we have 911 operators who are cross trained as nonemergency operators and so they do dual robs. And these -- these are nonemergency operatrs based on the priority, we have five Breyer at this levels in our customer service request queue, we will call citizens back and assist them and either by providing information, providing guidance, or taking a report or writing a supplement to an existing report. So that is what we call teleserve, historically. >> Tovo: Thank you very much. >> Yes, ma'am. >> Tovo: I'm sorry if I overlooked that in the if it is not there that might be something that would be useful for the public. , The public to know. [3:35:59 PM] And the definitions. I want to just -- my last question talking about the category of things that follow situations that might fall within family disturbance. We have had more a variety of community conversations around around what gets classified, what might be classified as family disturbance and I note that -- other calls were updated to reflect family disturbance. And I don't know that today is the day to have this information, but I would like to understand what kind of data we have to really, to help the policymakers, to help the council, to help us all, the community at large understand what that we have to reflect when an officer might need to be present versus situations that [3:37:02 PM] don't involve violence. So I don't know if you could speak to that issue today, but if so, I would invite you to and also to share with us how we would even begin to understand that, as to to what those family disturbance -- how likely it is that family disturbances -- a would fall into a category where violence might be -- might be present or might be at high ri risk. >> Yes, ma'am, I can speak generally on the appeal disturbances. Family disturbance, when a family disturbance report is written, it is when there has been no violence and when there has been no assault taking place, there is no injury or anything. Involved. It was when folks are just arguing, whether it is direct family members or we classify [3:38:03 PM] just roommates who saying you have two people living together and sharing an apartment, we also classify that under family disturbance. Or if there are ex-spouses and they are having a verbal argument it is also classified and should be written up under family is disturbance. >> So anything we see classified as -- family disturbance did not involve, did not involve violence at all? >> Yes, ma'am. Okay. >> Yes, ma'am. If a felon would have been a part of it it would have been an assault with injury or an aggravated assault or assault strangulation with the tag family violence. >> Tovo: Okay. Thank you. >> Yes, ma'am and I would just like to thank you also for recognizing our dispatchers and call takers. They do a tremendous khalidi job [3:39:05 PM] and they are very good at it and we are lucky to have such good operators and dispatchers here working for city of Austin and also in case you all don't know, just a tag, so you can, if go uh want to see what our operators 911 operators specifically are doing, they are a part of a filming project with ABC called emergency call and it airs Monday, September 28th at 9:00 P.M. On ABC and we are actually first up in the series premiere. So it is going to be a great series. >> Well, thank you so much for sharing that. >> Tovo: Would you mind saying that again. September 28, 9:00 P.M. And the show's name is? >> Emergency call. >> Tovo: Emergency call? >> Yes, ma'am. It is on ABC. So we were the pilot project for the series here in Austin a year and a half ago and 10 ABC expanded it and representative in this series, would be five [3:40:06 PM] different call centers, but we are the largest call center, emergency call center among the five and again we art to show off and so I have seen the trailer on the first 15 minutes and I think everyone will enjoy it and really gives you an in-depth look at what is the 911 operators do, because it is solely focuses on the 911 operators. >> Tovo: Thank you very much. >> Yes, ma'am. Thank you. >> Mayor Adler: I don't know why that made me laugh so much. It is like bringing on a fest for the latest promise. That was pretty god lieutenant Murphy. So we will watch for our amazing 911 operators and their television debut. How exciting. >> Yes, sir and hosted by Luke Wilson, to tag it even further. >> Mayor Adler: We will have Luke on the show later. See how this goes. Councilmember harper-madison joined us too. Any other questions for our staff on the fin 11 call [3:41:09 PM] analysis? All right. Really great work, Kerry, lieutenant Murphy, chief gay, all of the staff. This is really critical information as we dig into areas of best opportunity to relieve pressure on our staff, our civilian and our sworn to really good stuff and looking forward to continue to work on this and expanding alternatives and to hear the diversion numbers going up and, you know, one of things that we invested in care advertise nice to see a presentation from a provider turn into actual data later, so kudos to everybody for that. If there are no other questions, then I think we are ready to move into our final item in executive session. So I am going to read the script and then I will see everybody on that side. Thanks, everybody for being on the public side of the heating today. This meeting will go into closed session to take one item. The commit will consider item [3:42:11 PM] 4 -- issues concerning chapter 143, public safety contractual obligations, hearing no objection to going into executive session, we will now go into executive session. See you all there. >> [5:19:15 PM] <<Flannigan: We are out of Closed Session. In Closed Session we considered legal matters related to Item: 4 It is 5:19 p.m. and this meeting of the Public Safety Committee is adjourned.