Alameda County Grand Jury • 2022-2023

Oakland 911 Still Lacking a Timely Response

Published: June 09, 2023 118 pages Consolidated Report
View Original PDF

Note: Missing finding numbers detected: F2, F3, F4, F5

Findings and Recommendations 2 findings

F1 Page 27
Completing their own assigned work required by their position 2. Supplementing the work of dispatchers when there are not enough dispatchers on the floor 3. Supplementing the work of upper management, when the upper managers are unavailable Managers are currently working their own full day shifts, while also on call 24/7 to deal with issues. The ECC has currently authorized 8 managerial positions. A recent study recommended 15 supervisors/managers for Oakland. The Oakland mayor’s proposed budget recommends not changing the level, staying at 8 supervisors/managers. Non-Emergency Work The employees at the ECC also have non-emergency duties, as described above, including after-hours Internal Affairs calls, inappropriate 311 calls that should have gone to Public Works, calls for public records (which can take many hours of research to respond to), returning calls for those who hung up before connecting, and other calls – a total of 60% of the calls – that do not directly result in OPD or OFD dispatches to emergencies. Moving some of this workload to other departments would free up more time for ECC staff to answer 911 calls in a timely manner. Training is always an ongoing part of work in the ECC. When the proposed CAD software is finally deployed, it will require additional training time, which is time away from answering the incoming 911 calls. Police Officer Dispatching and Response Time The federal Uniform Crime Reporting (UCR) program is handled by the FBI, with input from every U.S. city. The UCR tracks the average response time for Priority One calls in Oakland— imminent injury, prevention of violent crime, and incidents involving weapons. The data has varied over the past 13 years and appears to be unrelated to staffing levels of 600 to 800 sworn officers in the OPD. The average response time for years 2017 through 2020 was 12 minutes. In 2021, the annual average jumped to 19 minutes, and the number of sworn 27 officers was 690. There is no published data available for 2022. The 2017 Oakland Audit stated that about 12% of police dispatches were Priority One. Priority Two calls in Oakland—urgent, but not immediate emergencies, such as in-progress misdemeanors, disputes, and just-occurred felonies—had an average response time for 2017 through 2020 of about two and a half hours. In 2021 the average response time was close to four hours. In 2017, Priority Two calls accounted for 37% of the dispatches. Priority Three calls accounted for 32% of dispatches in 2017. The FBI does not track Priority Four calls – for reporting abandoned cars and other low priority issues. Priority 4 calls accounted for 19% of dispatches in 2017. When citizens complain about the long response time, it may be that they are unaware of the different priorities. Just as in an emergency room at a hospital, people are served in the order of seriousness, not in the order of arrival. Several witnesses asserted that OPD is understaffed relative to the resident population the department serves. However, FBI statistics tell a different story. In 2019, the FBI published information on police staffing relative to population for all cities. The Grand Jury examined 15 cities in California with a population of 250,000 residents or more. Oakland employed 1.7 Law Enforcement Officers (LEO) per 1,000 residents. In contrast, San Francisco employed 2.6 LEOs per 1,000 residents and Fremont employed 0.8 LEOs per 1,000 residents. Clearly, OPD staffing ratios are in the middle of the regional spread for large California cities. Thus, the FBI statistics do not reflect the OPD management's perception of understaffing. There may be other issues that would warrant a higher officer to resident ratio, that might affect response time to 911 calls, but that is outside of the scope of this investigation. Needed Technology Upgrades The Grand Jury also learned that OPD’s 20-year-old software and hardware are not keeping up with evolving telecommunications technology. Components of the technology include caller identification and routing, call intake, computer aided dispatch (CAD), and records management systems (RMS) required by the FBI. Phone Answering System and Data Intake One improvement successfully implemented at the ECC regards the phone system. The old system, as noted in the 2019–2020 Grand Jury report, experienced frequent failures and dropped calls, and had a data entry system tied to an out-of-date records management system. That phone system has recently been replaced with a new system by AT&T. The new phone and data entry system, VESTA, has an expanded computer interface page that is more ergonomic and has decreased the operators’ repetitive stress injuries. It will also work with the proposed records management system when it is installed. Now that the operators have 28 been fully trained on the new VESTA software, there are no apparent decreases in productivity. The Computer Assisted Dispatch (CAD) System The existing CAD system in the ECC is provided by Motorola. The CAD system for the mobile data terminals in the patrol cars is not Motorola, resulting in some data loss as data goes from one CAD system to another. The proposed CAD system will be part of the Motorola PremierOne system and will integrate the ECC operator input (VESTA), mobile data terminals in patrol cars, and the records management system. The proposed system will have new software, new servers, new network switches, and be partly cloud-based. The proposed system will require more data entry at the time of the call. Currently, operators can complete an entry in two or three clicks, but the proposed system will require nine or ten clicks. This may decrease call volume handling slightly. On the other hand, the proposed CAD system will do a better job of locating the nearest officer to the incident, for quicker response and management of personnel. Next Generation 911 Call Identification and Routing The 911 call identification and routing system is evolving, along with evolving phone technology. When homes and offices had land line phones tied to unique addresses, the reverse phone look up system worked well. If the ECC knew the phone number, the ECC knew precisely where the call was coming from and could easily dispatch police and fire to the scene. Now, at least 80% of emergency calls come in via cell phones, sometimes in moving cars, which are much harder to locate precisely. The technology to locate cell calls via cell-tower triangulation is evolving and improving all the time. There is also an increase in calls that come in from private telephone exchanges in large office complexes and from Voice Over Internet Protocol (VoIP) services like Skype and Google Voice that are not tied to a location. There are newer technologies for locating callers available, called Enhanced 911 (E911) and more recently, Next Generation 911 (NG911) that are slowly rolling out nationwide. NG911, when fully implemented, will be able to handle incident related video and images from 911 callers, texts, traffic webcam integration, reverse 911, and OnStar systems from cars. All emergency communications centers are mandated to eventually upgrade to NG911. Any technology upgrades made now will have to have compatibility with NG911 in mind. Record Management System In 2000, the FBI issued a new crime data Record Management System (RMS), the National Incident-Based Reporting System (NIBRS), to replace the 1930s-era Uniform Crime Reporting (UCR) system. The NIBRS system became federally mandated in 2021, thus Oakland must upgrade its technology, train its people, and allocate resources to implement 29 the NIBRS data collection. OPD’s current CAD and RMS software, both supplied by Motorola, are 20 years old and are based on the old, out-of-date UCR data collection system. Upgrading to current FBI reporting standards will require installing NIBRS-based software and hardware into the ECC, patrol cars, and OPD desk stations, and be integrated with the proposed CAD system. Implementation of NIBRS will also require at least 3 months to train the police force. The existing Motorola software and hardware at the Emergency Communications Center is so out of date, it is no longer supportable. The Oakland Police Department is forced to buy replacement parts on eBay because Motorola doesn’t supply them anymore. Motorola has only one technician who still knows how to work on the 20-year-old software. Oakland and one other city are the only The existing Motorola software and customers left in the United States using hardware at the Emergency this software. Communications Center is so out of date, it is no longer supportable. The Oakland Police Department is forced to buy Part of the reluctance to change to replacement parts on eBay because NIBRS is that it collects far more data. Motorola doesn’t supply them anymore. In the old UCR system, a hierarchy was Motorola has only one technician who used and only the most serious crime in still knows how to work on the 20-year- a multi-crime incident was reported for old software. Oakland and one other city national statistics. For example, a are the only customers left in the United robbery-homicide would be listed as a States using this software. homicide, and the robbery would not be added to the statistics, which results in incomplete crime statistics. In NIBRS, both would be reported. Currently, it can take 60–90 days to consolidate records so that they are available for public review. With the proposed system, officers in the field will record the data real-time, for almost instantaneous viewing on an open data platform online, for internal and public viewing, and for response to constituent data requests. Observers will note an apparent increase in crime, but this is a result of more accurate, complete crime records. The NIBRS rollout and expanded availability of data will require effort on the part of the public information officer to explain how to interpret the data. Delayed Implementation New technology takes time to implement and OPD has not devoted the resources needed to finish the implementation of the proposed CAD and RMS. OPD has consistently stated from the 2019–2020 Grand Jury investigation to the present day, that the proposed CAD and RMS would be installed, trained for, and implemented “6 months from now”. The proposed technology was purchased in 2018, and in the intervening five years, not enough time, money, and staffing has been devoted to the implementation and training. If installation of 30 the proposed hardware and software is perpetually delayed to “6 months from now”, the city risks a catastrophic failure of the system. If that happens, they will be back to paper and pencil, as they were during the recent ransomware attack. Most of the staff and management involved in implementing the proposed system have left the city or are on leave, including key people in IT and OPD. There is no longer an Executive Steering Committee in the City Administrator’s office to give the work priority. The Grand Jury was told that getting the management staffing back up to functional levels to complete this job would be “like trying to turn a supertanker”. Specialized work is needed to integrate the proposed CAD/RMS system with 20 to 30 other systems, such as the Criminal Information Management System (CRIMS), the Consolidated Arrest Reports (ECAR), Forensic Logic, Alameda County databases, Contra Costa County databases, and other mandated data sharing. There is also reporting required by the OPD Consent Decree, unique to Oakland, including the Use of Force (Vision Technologies software). There is not currently and there has never been a dedicated, full-time team of dispatchers, patrol officers, records managers, and fire personnel assigned to adapt the proposed CAD and RMS systems to Oakland’s specific needs. Oakland Police Department’s IT team requested three to five full time employees (FTEs) from the Police Department and three to five FTEs from the Fire Department to develop the required configurations, but these positions were never supplied. San Francisco, on the other hand, has 20 people working full time on their CAD/RMS upgrade. Oakland has two employees mostly dedicated to this work. When the NIBRS system is finally ready to implement, efficiency in the short term will drop as it will take 3 months of installation and training for the associated personnel and their 500 plus patrol cars, 150 fire units, and numerous desktops. Several members of OPD/ECC Management said there is reluctance in the rank-and-file officers to change to the proposed system, and reluctance to spend more officer time on training and away from dispatching to the field to do "paperwork”. Upper management in OPD and the City are verbally supportive but have not provided all the needed resources. The OFD, which will share the proposed CAD system with the OPD, is eager to have the proposed system, but is not providing subject experts to implement the proposed system. Operators, dispatchers, and managers at the ECC are generally supportive of the proposed upgrades. So far, over $8M has been spent by Motorola’s team to install, configure, integrate, migrate, and coordinate with these other systems, but the work is not yet complete and won’t be for “another 6 months". Future Work Motorola recommends making major upgrades to their products every five years. The proposed CAD and RMS systems were purchased in 2018, so the products are already five years old. CAD and RMS systems are usually updated every few years to be properly 31 maintained and responsive to the latest technologies and legal mandates. At the other end of the spectrum, some witnesses expressed the opinion that OPD may not consider upgrading again until forced to by lack of technical support (just as they are now forced to upgrade their 20-year-old system). Periodic minor upgrades can be made to the updated system, but only if OPD and OFD agree to the upgrades and deem them worthy of shutting down the system for a few hours every so often and training the rank-and-file in the new additions. Procurement takes time—sometimes many months—so OPD needs to put them in their forecast now. Disaster Preparedness Most of the witnesses that the Grand Jury questioned were unaware of disaster preparedness plans for the ECC. The Grand Jury was encouraged to learn about the cloud- based functionality portion of the proposed CAD system. Such a system allows for flexible movement of physical ECC operations to the Oakland Emergency Operations Center (EOC), in the event of a natural disaster or social unrest. Since the ECC building is at heightened risk for sea level rise, flooding, tsunamis, and earthquake-generated liquefaction, cloud- based software is good disaster preparedness. But if AT&T, which currently routes 911 calls to the ECC, is not prepared to handle an increase in 911 calls during a major event, then, the Grand Jury was told, “it doesn’t really matter”. In a worst-case scenario, ECC personnel would have to rely on standard police radios for relaying emergency information and dispatching personnel to incidents. There is interoperability between various agencies in Alameda County for radio communications, but not for CAD. Oakland’s ECC has a different existing and proposed CAD system than the Alameda County Sheriff’s Department, as well as neighboring cities and counties. CONCLUSION The 2022–2023 Grand Jury has found the same personnel issues first reported by the 2019– 2020 Grand Jury. Hiring has been slightly streamlined, but there are still far too many vacancies. Human Resources Management should work with the union representation to assess other means to increase staffing, such as competitive salaries and benefits, or the use of non-sworn instead of sworn officers for the supervisors and managers. The staffing level does not meet the recommendations of the 2017 City of Oakland Audit nor the independent audit in 2019 by Federal Engineering, Inc. The recommendations made by the 2019–2020 Grand Jury in still stand: Hire more managers, operators, and dispatchers; hire them more quickly; distribute the work shifts to match the workload, and minimize the mandatory overtime that is causing stress, absenteeism and attrition. Until this is done, there simply aren't enough bodies employed at the ECC to answer phone calls in a timely manner. The other major concern is the stalled technology upgrade. The situation appears to be getting more dire, as key people leave, the remaining team members are unable to implement the 32 work, the patrol officers are resisting, and the 20-year-old system is waiting to fail catastrophically. FINDINGS Staffing and Workload Finding #23-1: The Emergency Communications Center lacks adequate staffing to promptly answer all 911 calls and, working with the City of Oakland Human Resources Management Department, has been unable to fill open positions in a timely manner. Finding #23-2: The ECC staff spends part of its time doing work that is not 911 related that could be handled by others. Finding #23-3: Emergency Communications Center staffing levels within a 24-hour period do not align with peak call volumes; union and seniority rules take precedent over wise use of resources. Technology Finding #23-4: The ECC is in dire need of technology upgrades and is not preparing for future upgrades. Finding #23-5: The team for installing the proposed Computer Aided Dispatch (CAD) and Records Management System (RMS) is severely understaffed and underprioritized. Finding #23-6: The "troubled” phone system reported by the 2019–2020 Grand Jury has been replaced and is functioning well, decreasing the number of dropped and abandoned calls. Finding #23-7: The ECC is not well prepared for a major disaster and lacks a disaster recovery plan.
No recommendations for this finding
F6 Page 50
What is the intent for the housing units (initially intended for remodeling in the grant application)? 7. How is the new unit more cost effective given the projected increase in projected costs ($19M) and the apparent elimination of housing units? 8. What is the breakdown for where additional County matching funds ($19M) come from? 9. What is the annual cost increase associated with the enhanced mental health services and staffing levels? (Staffing plan also requested) 10. Will the new facility be run by the Alameda County Sheriff’s Department or Behavioral Health Care Services? 11. How will the MHPSU project impact the operation and management of Santa Rita Jail? 12. Is there anything that prevents the County from changing the use of the new unit in the future (repurposing some of the new building into non-mental health)? 13. Were advisory boards or community interest groups consulted/informed of the (May 9, 2023, BoS Resolution) hearing? 14. Will the Mental Health Advisory Board and Care First Taskforce be informed or involved in the design modification process? The JLBC proposed the State Public Works Board re-schedule their review of the scope changes until their next scheduled meeting in August 2023 or until such time as the JLBC has to review the responses from the County. As explained above, a delay of three months results in additional costs approaching $700,000. Until then, the MHPSU is once again on hold. CONCLUSION The dire need for more mental health services for incarcerated people has been known for years. The State system was found to be underserving mentally ill prisoners in 2011. Babu v. Ahern brought more focus to the growing mental health issues that persons in custody were experiencing at Santa Rita Jail in 2018. The resulting Consent Decree encompasses many needed changes and, equally importantly, spurred the County into action on the construction of a mental health facility at SRJ. It is to the detriment of incarcerated persons, 50 and to County itself, that the $54M funding received in 2015 did not prompt immediate action. The impacts caused by the delayed construction of the Mental Health Program Service Unit have had serious consequences for inmates struggling with mental health issues in Santa Rita Jail as evidenced in the suit Babu v Ahern and the resulting Consent Decree mandating radical changes in mental health care treatment at SRJ. Since January 2023, SRJ reported five suicides. With a best-case scenario targeted completion date of 2028, any further preventable delays would be disastrous. The questions posed by the Joint Legislative Budget Committee (JLBC) mirror those the Grand Jury has grappled with through this investigation, particularly those about the care, stabilization and housing of persons with severe to serious mental illness who will not be accommodated in the MHPSU. Perhaps the JLBC could be prevailed upon for funds to build the initial structure with two housing units remodeled for severe patients and a medical building constructed attaching the two units; as well as the revised MBH to serve as a treatment center for less severe patients, education and counseling facilities, and office space for clinical staff? Regardless, the JLBC questions need to be addressed as soon as possible to avoid a delay in excess of three months. The County “matching funds” increase of $19M for the scope change to the Mental Health Program Service Unit remains a concern. As the construction moves further and further down the road, considerations about inflation, increased costs for labor, building materials as the construction may bring the County’s ability to cover continued increases in costs to halt the project altogether. The GSA continues to coordinate with the Alameda County Sheriff’s Office, Behavioral Health Care Services and BSCC to ensure the needs of these very different stakeholders are met in the context of how the new MHPSU will serve each agency. The Grand Jury is cautiously optimistic that there has been some improvement in communication and collaboration between all parties involved in the MHPSU project. The GSA and the Alameda County Sheriff’s Office are under new leadership and the focus of both departments point toward an atmosphere of cooperation. In order to ensure effective oversight, the Board of Supervisors should receive quarterly reports from the GSA on the status of the MHPSU project with evidence of progress made. The benefit of quarterly reports allows the Board of Supervisors to review, prevent delays, avoid problems and potentially reduce lawsuits, like Babu vs. Ahern. Adherence to the project schedule and budget mean more time and money to complete and support other County projects. A constant theme from witnesses speaks to staff shortages in all County departments concerned in the MHPSU project. It is important that the Board of Supervisors give direction to Human Resource Services to develop a new hiring model to correct this problem affecting GSA, the County Administrators' office, Alameda County Sheriff’s Office, Behavioral Health Care Services and the County as a whole. At the very least, conducting the promised audit of GSA could point to areas that need immediate attention, along with a progressive plan to staff the department fully. Without adequate staffing, management will not be able to organize and direct workflow and procedures to assist all employees in meeting their goals with alacrity. The Mental Health Program Service Unit project is different from other county building projects because it is partially funded through a Lease Revenue Bond Agreement. The review and approval process from multiple State Agencies is onerous and lacks concentrated coordination. This has contributed to the delays in ways that the County, sadly, has no control over. However, strict attention to project coordination and avoiding costly delays is even more critical. As questions swirl regarding how well the redesigned Mental Health Program Service Unit will serve the mentally ill population at Santa Rita Jail, the focus needs to shift to those individuals in desperate need of care. Ideally, an interim model would be set up to deliver more targeted mental health care to persons in custody now rather than later. FINDINGS Finding #23-8: The substandard level of care for mental health remains unchanged at Santa Rita Jail, and is likely to remain substandard until a new Mental Health Program Service Unit is built. Finding #23-9: The redesigned scope of the Mental Health Program Service Unit no longer includes the renovation of housing units or medical facilities included in the budget of the original proposal; nor does it provide beds, facilities to stabilize patients in crisis or accommodate long-term living units for persons with severe mental illness while they are incarcerated, despite an increase of $19M in County matching funds. Finding #23-10: The Board of Supervisors has failed to provide public updates on progress reports from the General Services Agency on the status of the Mental Health Program Service Unit project. Finding #23-11: Staffing shortages exacerbated by lengthy County job posting requirements have contributed to delays in the construction of Mental Health Program Service Unit due to unfilled job requisitions. Finding #23-12: There is no evidence that the Board of Supervisors conducted an audit of the GSA or updated its response to Grand Jury Recommendation 20-27, as promised in its original response to the 2019–2020 Grand Jury Report, “Alameda County’s General Services Agency Too Often Fails at Capital Project Management.” Finding #23-13: County matching funds to complete the Mental Health Program Services Unit project have increased by approximately $19M to date and will continue to rise with additional delays, inflation, cost of labor and increase in building material costs. Finding #23-14: Departments not familiar with construction of Capital projects and the approval processes that entails, need clear updates on the Mental Health Program Service Unit “in laymen’s terms”.
No recommendations for this finding

Conclusions 24

Agency Responses 1

Government agencies' official responses to this report's findings and recommendations. Click on a response to see the structured breakdown.

No Responses Found 1

Government entities assigned to respond to this report. No response documents have been linked in our database.

Alameda County Board of Supervisors Elected County Office