FAST Policing by Telephone: a Randomised Controlled Trial

Abstract

Abstract Research Question Can caller satisfaction and trust in police be improved (or equalled), after police agree to send a police car to meet with a caller face-to-face, by the alternative of immediate transfer of the call to a police officer who speaks to the caller at length by telephone? Data A total of 1016 calls for police service to 999 or 101 assigned by call takers as falling into a “medium” priority category were checked for eligibility, including consent of the caller to speak immediately to a police officer by phone if possible. Eligible offence types excluded domestic abuse but included a variety of other matters. A majority (57.7%) of eligible cases were about threats made by neighbours, workplace colleagues or others known to the caller. A total of 450 cases were selected as eligible for the test sample out of a total of assessed as potentially eligible. Methods Eligible cases were randomly assigned to either a control group (N = 225) of business as usual (BAU) attempts to provide a face-to-face meeting with a police officer, or the experimental group (N = 225) receiving immediate telephone transfer to a police officer who talked with the caller for over an hour as the initial police response. Analyses were done by intention-to-treat. While 99.75% (N = 249/250) of the experimental cases were treated as assigned, only about half of the 225 control cases actually received a face-to-face meeting with a police officer. All 450 assigned callers who gave consent to enter the experiment were contacted for a satisfaction survey at least 14 days following random assignment of the cases, from which the completion rate was 72.5% (almost identical in the two treatment groups). Findings Eligible, consenting callers reported substantially higher levels of being “satisfied” or “very satisfied” with the experimental police response by telephone (92.6%) than with the BAU efforts to arrange a face-to-face meeting between a police officer and the caller (68.9%). Trust and confidence in Kent police declined among 21% of callers receiving BAU service, but only 9% of callers given immediate telephone service. The median time from the initial call to a conversation between police and caller was under 1 min for the experimental treatment vs. 2721 min for the 80% of BAU control treatments in which any conversation between an officer and a caller occurred within 96 h after the call. Conclusion This first experiment in a research collaboration on FAST (Finding Alternative and Speedier Tactics) policing has opened the door to further tests of immediate response by remote communications (Rothwell, et al. Cambridge Journal of Evidence-Based Policing, 6, 1–24, 2022). For the kinds of cases included in this experiment, there is a clear preference by callers for the speedier service by a simple phone call over much slower attempts to provide a face-to-face meeting. If broadly adopted across many other high-volume, low-harm categories of requests for police service, fast policing by phone, video or silent live-chat online could improve public approval of policing while allowing more time for police to prevent more serious crimes.