In what appears to be a first for a health system, Parkland Health & Hospital System in Dallas recently implemented suicide screenings for all patients.
The program is the first of its type in the nation, according to Kimberly Roaten, PhD, director of quality for safety, education, and implementation in the department of psychiatry at Parkland and associate professor of psychiatry at The University of Texas Southwestern Medical Center, also in Dallas. A clinical psychologist working with Parkland patients, Roaten is one of the leaders who developed the new program.
“The Joint Commission requires healthcare providers screen all patients with psychological problems for suicide risk,” Roaten notes. “But we believe it’s important to screen everyone because some of this risk may go undetected in a patient who presents for treatment of non-psychiatric symptoms. We decided to move beyond what is required by The Joint Commission and screen all patients who enter our system for care based on the evidence that patients who die by suicide often present for treatment to EDs or primary care in the months leading up to their deaths. The timing was right in the fall of 2014 to reconsider the screening.”
In 2014, Parkland dedicated the resources needed to make this possible, which included hiring 12 psychiatric social workers, selecting a standardized and validated suicide screening instrument, building an algorithm in the electronic health record that triggers the appropriate clinical intervention depending on the patient’s answers to a few simple questions, and training all nursing staff to implement the program.
Parkland implemented suicide risk screening with all emergency department patients and hospital inpatients in February 2015, says Celeste Johnson, DNP, APRN, PMH CNS, director of nursing in psychiatric services at Parkland.
“In late May, we transitioned from the previous screening program to the standardized suicide risk screening at all Parkland community-oriented primary care health centers and also at the correctional health division for all inmates at the Dallas County Jail,” Johnson says. “Our goal is to screen every patient using proven screening tools that can help us save lives.”
Parkland has screened more than 100,000 patient encounters at the hospital and emergency department, and more than 50,000 patient visits in outpatient settings. Parkland uses the Columbia Suicide Severity Rating Scale (C-SSRS), a validated screening tool, with adults 18 and over and the ASQ (Ask Suicide Screening Questionnaire) with 12- to 17-year-olds.
The Parkland algorithm sorts patients into three suicide risk categories based on their answers to the screening questions: no risk identified, moderate risk identified, and high risk identified. Those at high risk are immediately placed under one-to-one supervision, suicide precautions are implemented, and an evaluation by a behavioral health clinician is initiated. Patients at moderate risk are automatically referred to a psychiatric social worker and usually are seen during the same visit. If a patient chooses not to speak with a psychiatric social worker during the visit, they will receive a follow-up phone call to provide additional support and resources.
For instance, a patient may come in with a sprained ankle or sore throat, but if his or her suicide risk screening shows moderate risk, Parkland’s clinical algorithm immediately alerts a member of the behavioral health team to come and speak with the patient. Before discharge, both moderate- and high-risk patients also are given information about suicide warning signs, suicide crisis center hotline numbers, and Dallas County community mental health resources.
So far, the suicide risk screening in the emergency department and inpatient units at Parkland has found 1.8% of patients to be at high risk and approximately 4% at moderate risk for suicide.
“To our knowledge we are the first big hospital system in the U.S. to implement a universal screening program for suicide risk and the data we are gathering will be significant for other organizations in the future,” Roaten says.
Implementing such a plan required far more than simply writing a new policy, Roaten notes. Parkland leaders considered screening completion time, who would complete the screening, when the screen would be completed, behavioral health provider response to patients who were identified to be potentially at risk, whether the health system had enough resources, and the effect on ED throughput. The Parkland team did not want the plan to slow the flow of patients through the ED.
Roaten and her colleagues determined that there several positive aspects of the plan. The screening instruments are easy to use, in the public domain, reliable and valid, and educational modules and video are available online for standardized training. Using the tool does not require a mental health background.
“And it’s the right thing to do for patients,” she says.
The Parkland team identified no real downsides to screening all patients, as long as the appropriate resources were in place prior to initiation of a screening program.
“The current evidence suggests that there is no iatrogenic harm associated with screening for suicide risk. We are not making our patients feel worse by asking about suicide,” Roaten says. “The opposite is true: Patients experience providers as empathic and appreciate their concern.”
The planning phase required a significant time commitment from multiple stakeholders prior to implementation, Johnson notes. The most time-consuming phase of the process was developing a specialized clinical response algorithm and building the screening tool in the EHR. It was important that the screening items be implemented in a way that was simple, user-friendly, and efficient, Johnson says.
Additionally, the clinical response algorithm unfolds in the EHR on the back-in, significantly improving patient care without complicating the user/provider experience, she explains. The planning phase also involved rolling out the education to all clinicians. The planning and implementation team consisted of dedicated physicians, psychologists, nursing staff, IT support, clinical educators, social workers, nursing leaders, and hospital administration who supported this initiative.
Data from the initial phases are now being used to estimate the needs in the final phase of implementation. Parkland expects there will be additional cost associated with the final phase as clinicians are hired to respond to at-risk patients.
“The primary benefit is that we have identified patients who are potentially at risk that might have been missed without universal screening,” Johnson says.
Parkland officials considered the possibility that a formal policy of screening everyone creates more obligation to detect suicidal patients, raising the possibility that if someone does commit suicide, the policy could make it easy for a plaintiff’s attorney to say that Parkland should have detected the risk. That risk does not outweigh the positive reasons for screening patients, says Paul Leslie, JD, executive vice president and general counsel for Parkland.
“While we cannot comment on what the plaintiffs’ bar may or may not argue, we believe that use of the new process will enhance the opportunities for identifying patients potentially at risk and will lead to better outcomes,” he says.