Body-worn cameras are widely used by state and local law enforcement agencies in the United States. They are worn principally by officers in the performance of duties that require open and direct contact with the public. Despite their widespread and growing adoption, the current evidence regarding the effectiveness of body-worn cameras is mixed. Some studies suggest that body-worn cameras may offer benefits while others show either no impact or possible negative effects. The mixed results of these studies strongly imply that additional research is needed. In particular, more studies employing randomized control trials [1] are needed.
Use of Body-Worn Cameras
In November 2018, the Bureau of Justice Statistics (BJS) published a report on the use of body-worn cameras by law enforcement agencies in the United States in 2016.[2] This report showed that:
- 47% of general-purpose law enforcement agencies had acquired body-worn cameras; for large police departments, that number is 80%.
- Among agencies that had acquired body-worn cameras, 60% of local police departments and 49% of sheriffs' offices had fully deployed their body-worn cameras.
- Overall, in agencies that had acquired body-worn cameras there were 29 body-worn cameras in service per 100 full time officers (expected to increase to 50/100 by late 2017 based on the 2016 data).
- About 86% of general-purpose law enforcement agencies that had acquired body-worn cameras had a formal body-worn camera policy.
- Agencies not using body-worn cameras stated cost (hardware acquisition, video storage, system maintenance) to be the primary disincentive.
Are Body-Worn Cameras Effective?
According to the 2018 BJS report, the main reasons (about 80% each) that local police and sheriffs’ offices had acquired body-worn cameras were to improve officer safety, increase evidence quality, reduce civilian complaints, and reduce agency liability.
Research does not necessarily support the effectiveness of body-worn cameras in achieving those desired outcomes. A comprehensive review of 70 studies of body-worn cameras use found that the larger body of research on body-worn cameras showed no consistent or no statistically significant effects.[3] This meta-analysis was rated by CrimeSolutions and resulted in a No Effects rating for the impact of body-worn cameras on use of force, assaults on officers, officer-initiated calls for service, arrests, traffic stops and tickets, and field interviews (i.e., stop and frisk). See Practice Profile: Body-Worn Cameras’ Effects on Police Officer Behavior. This profile is based on a meta-analysis published in 2020.
These mixed findings are further reflected in findings from evaluations of ten body-worn camera programs that have met the stringent criteria for inclusion in NIJ’s CrimeSolutions, see Table 1.[4] Across these evaluations, researchers looked at a range of outcomes, including use of force, citizen complaints, arrests, and assaults on officers. Four of the body-worn camera programs evaluated were found to have no, limited, or even negative effects.
Based on these reviews and the existing research on the impact of body-worn cameras use, it is clear that further research is essential to determine the value of body-worn cameras use and potentially the more effective ways body-worn cameras could be deployed. Given the growing use of body-worn cameras, it would be best to build in rigorous evaluations as law enforcement agencies expand their use of this technology.
| Program | Rating | Description | Study Year |
|---|---|---|---|
| Police Body-Worn Cameras (Boston, Massachusetts) | Effective | This involves police officers wearing cameras on their uniforms to improve the civility of their interactions with citizens. The program is rated Effective. There were statistically significant reductions in citizen complaints against police and police use-of-force reports for officers who wore cameras, compared with those who did not, and statistically significant reductions in complaints against control officers in the treatment districts, compared with officers in the untreated districts. | 2020 |
| Effects of Body-Worn Cameras on Reducing Rates of Citizen Fatalities | Promising | Researcher looked at law enforcement’s use of cameras to record interactions with civilians to reduce citizen fatalities using a subsample of U.S. police departments drawn from the Bureau of Justice Statistics’ Law Enforcement Management and Administrative Statistics Body-Worn Camera Supplement. Based n that study, CrimeSolutions rated the use of body-worn cameras to reduce citizen fatality rates as Promising. Agencies that acquired cameras had statistically significant decreases in fatal police–citizen encounters after three years, compared with agencies that did not acquire cameras. There were no statistically significant differences in fatal encounters between a reduced set of agencies with cameras and matched agencies without cameras. | 2021 |
| Police Body-Worn Cameras (Birmingham South, UK) | Promising | In Birmingham, UK, evaluators found that deploying body-worn cameras resulted in a statistically significant reduction in citizen injury, but no statistically significant reduction in officer use of force or injury. | 2015 |
| Police Body-Worn Cameras (Rialto, Calif.) | Promising | In Rialto, CA, evaluators found a statistically significant reduction in police use-of-force but no significant difference in citizen complaints. | 2015 |
| Body-Worn Cameras (Las Vegas Metropolitan Police Department | Promising | In Las Vegas, Nevada, and evaluation of the Metropolitan Police Department’s use of body-worn cameras revealed that the use of body-worn cameras resulted in a statistically significant reduction in both complaints and use of force. | 2018 |
| Police Body-Worn Cameras (Phoenix, Arizona) | Promising | In Phoenix, AZ, evaluators found body-worn camera use resulted in statistically significant decreases in citizen complaints, and there were mixed results regarding camera use on arrest rates. There were no statistically significant differences in citizen resistance. There was a statistically significant increase in use of force, and less proactive, officer-initiated contact. | 2021, 2017 |
| Police Body-Worn Cameras for Intimate-Partner Violence Cases (Phoenix, Ariz.) | Promising | Evaluators looked at a program that equips police with on-officer cameras to record contacts with civilians during intimate-partner violence incidents. Camera use was statistically significantly more likely to result in arrests, charges filed, cases furthered, and both guilty pleas and verdicts. There was no statistically significant difference in sentence length. However, there was a statistically significantly greater reduction in case processing time in cases not involving a camera. | 2016 |
| Police Body-Worn Cameras (Washington, D.C.) | No Effects | In Washington, DC, evaluators found no statistically significant differences in police use-of-force, nor the number of citizen complaints. | 2017 |
| Police Body-Worn Cameras in Eight Police Departments in the United Kingdom and United States | No Effects | A multi-site evaluation of eight departments in the US and the UK found, overall, found no statistically significant differences in police use of force, number of citizen complaints, or number of arrests for disorderly conduct for police officers who wore body-worn cameras, compared with officers who did not wear them. | 2016 |
| New York City Police Department Body-Worn Cameras’ Effects on Civility and Lawfulness of Police–Citizen Encounters and Policing Activities | No Effects | In New York City, camera use had no statistically significant effects on number of arrests, arrests with force, and summonses issued after 1 year. Officers wearing cameras had statistically significant reductions in complaints filed against them and made more stop reports. Camera use resulted in a statistically significantly reduced likelihood of lawful stops and frisks but fewer subjects searched. | 2021 |
| Milwaukee (Wisconsin) Police Department Body-Worn Cameras | No Effects | In Milwaukee, camera use had no statistically significant effects on officers’ total number of proactive activities, specifically on the number of traffic stops or business checks, nor on arrests, citizen complaints, and use-of-force incidents. Officers with cameras conducted statistically significantly fewer subject stops, and statistically significantly more park and walks. | 2020, 2018 |
Primer on Body-Worn Cameras for Law Enforcement
Developed by the NIJ-funded NLECTC Sensor, Surveillance and Biometric Technologies Center of Excellence, A Primer on Body-Worn Cameras for Law Enforcement provides an introduction to body-worn camera systems. The report discusses the functions and features of body-worn camera systems and highlights issues and factors that law enforcement organizations should consider before and during implementation.
Read an abstract and access the full primer.
Technical Guidance on Body-Worn Camera Technologies
Agencies should consider how body worn cameras will meet their mission needs and requirements prior to procurement and use of the technology. To provide general guidance to law enforcement practitioners, NIJ, NIST and the FBI developed a table listing operating characteristics and associated functionality descriptions based on existing technical resources about criminal justice use of video.[5],[6] The operating characteristics and associated functionality descriptions in the table can help agencies determine what they need as they consider the commercial products available.
| Operating Characteristic | Functionality Description | |
|---|---|---|
| Recording Format | Video and audio to record and export in a standard, open, non-proprietary format, including both Codec and Container, such that it can be replayed in freely available software (e.g., VLC player) without processing or conversion. Standard open formats should be used for interoperability. Examples include MP4 and MKV. Data formats that can only be viewed within manufacturer-specific replay software are not recommended. | |
| Video Resolution | VGA, HD 720P, and 1080 HD are predominant standard resolutions. The higher the resolution, the more storage is needed. Estimates below were created assuming H.264 compression with medium to high motion at 30 frames per second (fps) derived using a heuristic formula widely used in industry. Actual storage utilized is dependent on scene complexity and the motion of the video captured. Consider what sort of analysis may be conducted on the video before selecting a resolution. For example, if forensic video analysis is anticipated, a resolution less than VGA is not recommended.
| |
| Video Encoding/ Compression | Use of the lowest possible amount of compression in order to maximize the amount of information available to law enforcement. Consider what sort of analysis may be conducted on the video before selecting video encoding or compression. Examples include MPEG-4, H.264, and H.265. H.264 is an improvement over MPEG-4 compression. H.265 is a new standard which further reduces storage needs while maintaining viewing quality. | |
| Frame Rate | 30 frames per second (fps) is a standard video frame rate. Higher recording speeds capture more motion detail but require increased storage. Frame rates lower than 25 fps suffer from increased motion blur. | |
| Horizontal Field of View | Adequate to capture a majority of activity at a reasonable distance. This would likely require at least a 90 degree field of view. Wide angle lenses capture more of a scene, but distort the view and lose detail towards the edges of the frame. Software tools may be required to properly analyze or view the video from extremely wide angle video. | |
| Camera Focus | Device should be able to focus on all objects from about 1 foot away to infinity. Continuous autofocus or fixed focus should be employed for usability. Manual settings should be avoided as they can distract the user. Motion jitter and blur can be significant when the camera is moving. Automatic image stabilization can reduce this effect. | |
| Audio Quality | The system is capable of clearly capturing conversational speech at a distance of 3 feet without wind or excessive background noise. | |
| Separate Audio Resolution and Encoding/ Compression | If the device will be used in a mode to record audio only, compressed audio requires less storage than video (4–60 MB per hour per microphone depending on desired quality). If high speech quality is needed, a sampling rate of at least 22 kHz with at least 24-bit capture is suggested per microphone. Higher values might be necessary to capture increased fidelity at a distance. Standard open encoding with speech quality resolution suggested. Examples include MP3 and WMA. | |
| Recording Triggering | Cameras could record continuously or be user-triggered or event-triggered. Cameras take time to start recording video after being powered on and after recording is initiated. This recording latency period should be minimal. | |
| Night-time/Low Light Functionality | Quality of video footage recorded in low light or night conditions should be useable. Visible flash and infrared illumination can increase the quality of video taken at night but will affect battery life. Low-light filtering, infrared, near infrared, and other low-light compensation technologies or mechanical filters can increase the quality of video taken in low light and severe weather conditions but can affect scene and motion detail. | |
| Synchronization and Metadata | The device is capable of recording audio simultaneously and time synchronized with video. Consider the additional information that should be collected with the recorded material. Automatically generated data about the wearer, location, date, and time can be collected and packaged in the video format. Device clock must be synchronized with an external universal clock, either GPS or another source, when the unit is plugged in for absolute time of day to ensure accuracy. | |
| Tamper Resistance | The device prohibits recordings from being edited or deleted and should not overwrite existing data before they have been transferred. Systems that can export a hash value of files being transferred may provide an enhanced capability to demonstrate tamper resistance. Standard encryption such as AES can be employed to protect data and improve the management of lost devices and memory cards. | |
| Data Transfer | Recommend standard USB2/USB3 compliant connection (mini/micro) for charging and/or data transfer. USB3 is preferred as speeds are considerably faster. The connections should be standard on both the device and on any docking station. Data connections that use a proprietary form factor are not recommended. | |
| Data Export | Device exports all recorded footage to data archiving or data management system in its original file format and without loss of quality or associated metadata. Device should record an audit log which should include information such as device serial number and device events; e.g., on/off, charging, start/stop recording, remaining storage capacity. | |
| Onboard Storage | Storage can be integrated into the device or provided on removable industry standard memory cards. Removable media has utility in terms of versatility and expansion but comes with security risks. Consider whether enough storage should be available to record a full shift by the officer wearing the device, such as 8–12 hours of non-volatile onboard storage. Loss of power must not cause data to be lost or corrupted. | |
| Battery Life | Consider whether the battery should provide enough power to record a full shift by the officer wearing the device, such as an 8–12 hour battery life. Devices that do not run on rechargeable batteries are not recommended. | |
| Durability | Device should withstand considerable and repetitive pressure, vibration, and mechanical shock. It should operate within a temperature range from very cold to very hot and be resistant to common environmental hazards, such as dust, condensation, water splashes, and RF interference. | |
| Weight and Form Factor | Device should not distract or hinder the officer wearing the device from performing other job functions, especially ones related to officer safety. Cameras are designed with widely varying mounting methods and options. Device should be selected for maximum usability and safety. | |