Florida Law Enforcement Policies for and Experience With Tarasoff-Like Reporting

Discussion

To our knowledge, this is the first study to evaluate the presence of policy, knowledge, or experience with Tarasoff-type warnings coming from mental health professionals to Florida law enforcement agencies. The encouraging news is that most law enforcement agencies who responded had policies in place to address Tarasoff concerns. Given that the survey was brief to encourage participation, the specifics of the policies are not fully known, and, given some of the responses received from departments, specifics on policies may not have been shared even with more direct and expansive inquiries. One key element that was identified is that the majority of the policies are similar to, if not the same as, how a department would respond to a call regarding a patient expressing self-harm (68%). This observation raises the question of whether unique policies and procedure are needed for explicit Tarasoff situations, such as a criminal background check for every Tarasoff-type call as part of the risk assessment for the officers, more in-depth domestic violence screening of significant others living with the individual, or follow-up contact with a treatment provider if one is known. A 1998 study by McNeil and colleagues¹⁵ examining characteristics of individuals about whom Tarasoff warning calls were made by mental health care providers to the police indicated that over 50 percent of the patients had a criminal history; the majority had made threats against a family member, friend, or a former partner; and 52 percent were civilly committed. Although this study is older, it does highlight that about 50 percent of patients about whom a mental health care provider had concerns, and theoretically was not in a position to civilly commit, were not detained after the warning.

What also became evident with the results of our survey is that there is a lack of uniformity in centralized data processing, at least among the respondents. Whereas 90 percent of respondents would document a call from a mental health provider, only 38 percent of agencies pass the information to a centralized hub. Although the documentation and response rate seem much improved from the earlier studies discussed,^11,12 this lack of centralized collection of data may hinder the formation of sound public policy and implementation of surveillance measures (e.g., improved mental health care funding, training for law enforcement and mental health care providers, or targeted mental health interventions for certain populations). The lack of centralized data from law enforcement also makes it difficult to determine metrics to track the frequency and outcomes of these calls on a state level and to compare one state’s approach to another’s.

Most agencies that responded have policies regarding notifying the potential victim (89.1%) or locations (91.4%). This finding is important information for psychiatrists and policy makers to be aware of. Given that law enforcement has more resources and capacity to identify potential victims’ contact information (e.g., driver’s license databases) than mental health providers in general, it seems appropriate to delegate notification to the law enforcement community. In addition, law enforcement may be better able to answer potential victims’ questions on how to handle or address the situation, such as what will be done, how to protect oneself, and legal options (e.g., how to obtain a restraining order or initiate a trespass complaint). It was also important to learn that 59.6 percent and 80.6 percent of departments had policies about monitoring a suspected victim or location, respectively, which indicates that notifying law enforcement did have positive value for safety to the threatened person or location beyond just notifying potential victims.

Although many agencies said they had policies in place and would document calls that come in, there was difficulty gauging how many of these warnings the departments received in a year. The only free-answer question (i.e., the number of Tarasoff calls an agency received in the preceding year) was the most frequently unanswered question. Those who did provide numbers reported few if any known calls (< 10), as indicated by the mean ± SD being 2.8 ± 7.8. It is important to note that this relatively low number of calls occurred when Florida was still primarily a permissive Tarasoff state. The authors hope to follow up in the coming years to determine if the change from permissive to mandatory reporting will have an impact on this number.

Similar to the study by Huber et al¹¹ in North Carolina and Michigan, a minority of law enforcement respondents (43.4%) had knowledge, education, or experience with the Tarasoff ruling. One could argue that this finding is not surprising because Tarasoff is a court ruling that is only applicable in California. But most mental health providers are trained on the duty and obligations that have resulted from Tarasoff and the majority of states have a Tarasoff-type legal obligation.⁶ For the average mental health provider, Tarasoff has become the shorthand way to refer to the duty or obligation to keep a third party safe. The fact that past researchers thought that law enforcement in different jurisdictions should know about a decades-old California court case highlights the potential communication problem that may exist between mental health providers and law enforcement. Mental health providers may know the duty and obligation, but not the fine points of the legal system⁶; similarly, law enforcement may be well versed in their jurisdiction’s laws but unaware of the duty or expectations of outside professions.

The Florida statute was intentionally written to require a treatment provider to notify law enforcement without specifying which law enforcement agency (i.e., the agency where treatment is provided, where the patient lives, or where the potential victim lives). In part, this was done to decrease the burden on the care provider (e.g., one call rather than three; not having to decide which agency to call). In Florida, it can be as simple as the provider dialing 911, with the assumption being that the dispatcher would better know which law enforcement agency or agencies should respond, how to contact them, and what information law enforcement would need to know to make the wellness check and to contact the potential victim or location.

As to specifics of education that may benefit law enforcement in the state of Florida, that may vary depending on the situation and resources of each department. For example, 34 Florida sheriff’s departments had special Crisis Intervention Team (CIT) officers in 2014, and it is estimated that more than 3,000 law enforcement individuals had received CIT training by 2015.^25,26 There have been attempts to increase the number of counties participating, with the state allocating additional funds for counties not already participating.²⁵ We are not aware of how many police departments in the state of Florida have undergone CIT training, although it is known that some larger metropolitan departments have done so. The CIT training program involves at least 40 hours of additional training designed to help law enforcement and other first responders effectively render services to individuals with mental illness.^25,–,28 The Florida CIT training program includes education about mental health symptoms, risk assessment for self-harm and violence, de-escalation techniques, and resource utilization (e.g., follow-up with family, case workers, and facilitating contact with crisis intervention units if needed).²⁷ There is abbreviated CIT training available for dispatchers, but we have no data about how many dispatchers have participated in this program or how the training is modified.²⁷ Although the official Florida CIT program model does not specifically make reference to Tarasoff, the fact that this training is available and has been utilized by at least half of the Florida sheriff’s departments may indicate why the sheriffs’ offices had the highest familiarity rate with the Tarasoff ruling and the concepts it incorporates.

One complaint that many health care providers voice when dealing with a Tarasoff situation is the lack of feedback related to outcome from a law enforcement perspective. As noted in some of the anecdotal examples above, such as the 2014 mass murder in Isla Vista, law enforcement identified that there was no imminent risk, so they did not feel they were able to initiate an involuntary commitment process at the time.¹⁸ As noted in the official report on the incident, the police talked with the perpetrator’s mother but did not provide any feedback or have a discussion with the mental health counselor, who also called in about a month before the event. This is the description of the interaction in the report:After [the Officers] were finished speaking with the suspect, one of the deputies called [the mother] and briefed her on the situation. The deputy asked [the mother] if the YouTube videos she watched were suicidal or homicidal in nature. [The Mother] said they were not . . . . At the end of the conversation, the deputy asked [the mother] if she needed further assistance, or if there was anything else that needed to be done. [The Mother] did not request any follow-up action (Ref. 18, p 47).

Without suggesting that the officers were at fault or did anything wrong in this case, this is an example where feedback or additional discussion with the mental health provider who had called in might have resulted in a different outcome (e.g., more frequent visits or change in therapeutic approach, better therapeutic interaction due to outside collateral information, deeper assessment of potential risk at next visit). California did make changes in aspects of its laws related to wellness checks after this event, such as giving law enforcement more discretion to confiscate weapons.¹⁸ An additional change that may be considered in the future, especially regarding individuals with a known mental health history who are in active treatment, is for law enforcement to contact the treater about the outcome.

Although an increased level of communication between law enforcement and treaters for the purpose of patient and public safety can have potential benefits, there is also the prospect of additional ethics concerns (e.g., confidentiality) or legal concerns (e.g., HIPAA). Although a full discussion of these areas is beyond the scope of this article (e.g., dangerous-patient exception, emergency exceptions), a brief discussion is warranted. The American Psychiatric Association’s ethics guidelines note, “Psychiatrists should not provide third parties with more information than is needed under the circumstances and they should stick to the facts” (Ref. 29, p 5). Adhering to this guidance may be challenging when it comes to interactions with law enforcement where it may be difficult to identify what information is relevant (e.g., potential marital or sexual problems leading to patient stress if the suspected threat is to the general public or workplace). Most states have laws that allow for clinicians to provide information in good faith when there is concern of serious risk to third parties based on clinical judgment. In addition, emergent or dangerous situations are exceptions to HIPAA.³⁰

Although there are concerns related to the provider’s giving out information, there are usually no concerns about their receiving factual outside collateral information. If law enforcement officers are able to identify a mental health provider from the original call or learned from the patient, then it might be important for future policies or statutes to allow feedback from law enforcement to treaters as part of a Tarasoff or wellness call situation. Receiving factual information that a safety check has occurred, the context of the interaction, and the outcome would likely aid the clinician in providing follow-up care.

The new Florida statute mandates notice to law enforcement but is permissive for notification to a potential victim. This requirement makes it difficult to predict the law’s effect on the frequency with which providers will try to call potential victims in the future. Considering the state was permissive before and remains permissive after, it may have no effect. It could result in increased attempts to call because providers may make the erroneous assumption that the new statute is mandatory in all areas. It could also lead to a decreased number of calls because the provider may feel confident that law enforcement has been notified and, therefore, the potential victim will be notified through law enforcement. An older study looking at potential victims contacted regarding threats from patients indicated that most individuals were already aware of the threat and were perceived as having “anxiety mixed with thankfulness” for being contacted (Ref. 14, p 1212).

We also note that more than half of surveyed agencies did not participate in this survey even when invited to do so by varying means and multiple invitations. Those who did participate preferred answering a paper questionnaire and mailing it back over filling out an online survey. Although this finding was somewhat unexpected, it is important information for those who plan to conduct future research on the topic. Therefore, when considering any future surveys among law enforcement agencies, it would be advisable to include both paper- and web-based questionnaire invitations.