METHODS
From August 2015 to November 2016 , the CAD-MDD-S screening tool was distributed on tablet computing devices to ambulatory patients ages 18 years and older who presented with concerns unrelated to mental health to two of our clinical locations . The locations consisted of an academic ED and an academic-affiliated student health services outpatient clinic . The academic , Level I tertiary care ED treats approximately 50,000 patients annually , while the student health services clinic cares for approximately 20,000 students and 2,000 community members annually . This study was approved by our university ’ s Institutional Review Board , and informed consent was obtained from each patient prior to participation .
Student research assistants ( SRAs ) enrolled in a university-based academic course available to undergraduate and graduate-level students administered the screening tool at both locations . A convenience sampling method was employed , enrolling between the hours of 10 a . m . and 7 p . m . on Tuesdays through Fridays , due to the availability of SRAs . The SRAs were responsible for working closely with clinical staff to identify eligible patients for the study . Patients were approached by SRAs in the privacy of their respective examination rooms when deemed appropriate by the provider and care team .
After patient consent was obtained , SRAs provided patients with the tablets to complete the screening tool . Each patient privately received a unique , randomly generated user login identification ( ID ) and password , with an associated subject ID . Once complete , data generated by the screening tool was stored in a secure database . The screening tool generated output for each patient in regards to their suicide risk and MDD diagnosis . A qualitative suicide warning was generated in the form of “ Yes ” or “ No ” and was recorded by the research assistants . MDD output consisted of two data points : diagnosis in the form of “ Yes ” or “ No ” and the percentage confidence of this diagnosis . Both data points were also recorded by the research assistants .
In addition to the output from the CAD- MDD-S , the following variables were also recorded by the SRAs : time of patient approach , time of patient completion of participation , and patient triage acuity level ( only applicable for ED patients ). Given that this was a pilot feasibility study , specific patient information such as age , gender , ethnicity , and other factors was not recorded . Upon notification of a patient ’ s positive suicide and / or depression screen , SRAs notified the health care provider so that additional evaluation , treatment , and / or referrals could be performed at the discretion of the treating provider . Additional history would be obtained from the patient by the care team in order to assess risk and determine follow-up to be scheduled with a primary care provider or specialist .
The frequency of positive screenings for suicide and MDD and the time to completion of the screening tool by patients in both settings were determined . The goal was to assess the feasibility and efficiency of the screening tool in order to assess the potential impact the tool may have in identifying at risk patients in these clinical settings .
RESULTS
From August 2015 to November 2016 , a total of 390 patients were approached by SRAs to participate in the study , with 345 ( 89 %) consenting to complete the screening protocol at one of the two acute care locations ( Figure 1 ). Approximately 55 % of the patients were enrolled at the student health services clinic , with the remaining 45 % enrolled in the ED setting . There were 13 positive suicide screenings ( 4 %) and 76 positive MDD screenings ( 22 %) during the study ; 2 % of patients screened positive for both suicide and MDD . Patients were more likely to screen positive for suicide in the ED ; however , there were equal rates of positive MDD screening at both locations .
Time estimates of patient participation , beginning from the initial SRA approach for consent to the time of completion of the survey , were available for 337 patients ( 98 %). The median time to completion for patients was six minutes . Figure 2 depicts groupings of total elapsed time of patient participation during the study . Approximately 87 % of patients experienced a total participation time of ten minutes or less .
DISCUSSION
To our knowledge , this was the first pilot study to present specific , real-time data with suicide and MDD screenings in Appalachian acute care settings of an ED and a student health clinic . Our study demonstrated that the CAD-MDD-S was a feasible method to rapidly screen for suicide and MDD in both settings , with the majority of patients having a time to completion estimate of less than 10 minutes ( 87 %), and of those in the less than 10 minute time frame , the majority finished in less than five minutes ( 52 %). The median time to completion was six minutes in the chaotic environment of an ED .
By having patients complete the screening tool during their idle care time with non-clinicians , we were able to effectively maximize provider time with patients , while providing minimal interruptions and delays in the patient ’ s overall length of stay . For the purposes of this pilot study , we chose not to enroll patients in the waiting room areas of both locations , in order to maximize patient privacy concerns . However , after the initial analysis of this study , the waiting area could become an optimal idle time to use for this type of screening , if the proper privacy concerns be addressed .
Although the majority of research in this field being focused on patients with mental health concerns as the primary inclusion criteria , it is vital to also screen those patients who are presenting with other medical issues , just like in our study , given that the overall prevalence of depression in the general ED population is about 20 %. 14 Our results are consistent with prior studies describing the under-diagnosis of depression and other mental health disorders among ED patients . 15
An aspect of the our pilot study that may have allowed for a 89 % participation rate is the usage of a tablet to administer the screening tool , in combination with allowing patients to maintain their privacy by completing the questions alone in their
West Virginia Medical Journal • June 2021 • 27