Transitions between inpatient or residential mental health (MH) and outpatient MH care are high-risk periods for patients. During the postdischarge period, patients are vulnerable to poor outcomes, including higher risk of suicide, care attrition, and hospital readmission.1–3 Incorporating evidence on these risks and the positive impact of outpatient MH follow-up postdischarge, community standards of postdischarge quality of care now include measurement of outpatient MH engagement within 7 and 30 days of inpatient psychiatric discharge,4 aimed to link inpatient and outpatient care plans, reduce readmissions, and improve longer-term care engagement.
Recent work suggests that postdischarge outpatient follow-up within 30 days may reduce early readmissions5 and better link patients to additional care during the next several months.6 Evidence also suggests that early outpatient engagement following residential treatment may lower mortality risk in the few years following discharge.7
Building on this, the Department of Veterans Affairs (VA) Office of Mental Health and Suicide Prevention developed performance metrics delineating stratified postdischarge engagement goals based on the discharge setting and suicide risk, with recommendations for 2 to 4 follow-up encounters in the 30 days following VA residential or inpatient MH discharges.8 Performance across VA sites has varied, with a median national performance of 74% in 2019, and facilities ranging from connecting approximately 65% to 85% of patients to care. At the site of this pilot, a gap in transitioning patients from residential MH to outpatient MH care was identified using performance metric data.
Efforts to address transitions of care require engaging multiple stakeholders across various settings with interdisciplinary collaboration, building out or adjusting processes on the acute side and the outpatient setting, and developing bridging interventions.3,9–11 The effort described here included a combination of strategies that previously have been found to be successful in improving transition processes. Notable components include predischarge scheduling of follow-up appointments, a dedicated transition team to guide the patient between care settings, and postdischarge telephone follow-up and tracking of outpatient MH care engagement.
The primary aim of this effort was to develop, pilot, and formatively evaluate a transition management process for patients discharged from VA residential MH care at the Greater Los Angeles VA, a large VA health system, to enhance outpatient MH engagement during the first 30 days postdischarge. In line with previous literature, we aimed to address numerous aspects of the transition process by developing new processes across various settings.12 Notably, the pilot period included the start of the COVID-19 pandemic and therefore needed to be feasible in the context of largely remote outreach and follow-up methods. Both pilot and maintenance data are described in this report.
METHODS
Patient description/context
Patients included in this pilot study were individuals being discharged from the Greater Los Angeles VA MH Residential Rehabilitation Treatment Program (RRTP) to the community setting. The RRTP provides residential care for patients with MH conditions, substance use disorders, homelessness, and rehabilitative care needs with the goal of fostering patient independence prior to discharge into the community.13 Interdisciplinary clinical care at the RRTP includes medical, psychiatric, vocational, educational, and social services. The RRTP is a 296-bed facility. Prior to implementation of this transitions-of-care effort, RRTP social workers were responsible for selecting and connecting patients with appropriate MH outpatient follow-up appointments postdischarge. The outpatient MH network at this VA encompasses more than 25 MH programs, including a number of subspecialty MH clinics representing a complex and geographically dispersed system.
Framework
The Quality Enhancement Research Initiative's (QUERI's) synthesis and recommendations on transitions of care from hospital to home provide a framework for improving transitional care in the Veterans Health Administration.12 The key recommendations of this framework address multiple components of care transition: ability to assess and respond to individual patient needs; bridging interventions between pre- and postdischarge settings; and interventions designed to address the needs of patients with specific complex, chronic conditions.12
Although this framework does not focus on MH transitional care, many of its components mapped onto identified challenges and thus opportunities for improvement at the site of this pilot. For instance, the breadth and complexity of the outpatient MH system with more than 25 outpatient MH programs significantly complicated anticipatory discharge planning for inpatient/residential providers, and patients lacked a clear point of contact after discharge before connecting with outpatient care.
Program design
Prior to designing an intervention, the team examined 5 months of local data on the rates of patients connecting to 2 or more outpatient MH visits within 30 days of RRTP discharge: rates ranged from 40% to 69%, below the locally set target of 80%. In investigating key drivers of low success rates using medical record reviews and focus groups with RRTP providers and outpatient MH leaders, findings indicated lack of anticipatory communication between discharging providers and outpatient programs, including about inclusion and exclusion criteria for various outpatient specialty MH programs; a lack of standard workflow to refer patients to outpatient MH care upon discharge; lack of awareness and processes on the outpatient clinic side to facilitate rapid visits for recently discharged patients at the clinic lead and scheduler levels; and lack of standard outreach and care management in the period between discharge and outpatient care to encourage and monitor successful patient engagement with outpatient care. Many of these drivers reflected elements of the QUERI framework, which provided guidance on developing processes to improve these gaps. Outreach to other VA sites with greater success in connecting patients with outpatient MH care postdischarge also helped identify intervention components that might fit local needs. One described intervention from a site with significant improvement over time included a dedicated transition management team and noted the benefits of points of contact bridging the residential/inpatient and outpatient settings.
To address the identified barriers, a new transitions-of-care program was piloted for RRTP discharges, iteratively adjusted through weekly Plan-Do-Study-Act (PDSA) cycles during the pilot period, and then continued in a postpilot maintenance period. A multidisciplinary team designed the following initial program to improve the transition process.
In the preimplementation stage, the multidisciplinary team developed a new one-stop electronic MH Postdischarge Consult in conjunction with the facility informatics team. Each version of the consult, one version for outpatient MH follow-up care at the main campus and another version for MH care at 2 large ambulatory care centers or community clinics, included all potential outpatient MH programs on a single form. This list encompassed nearly 30 MH programs in all. The second new component of the intervention was the development of a Postdischarge Management Team (PDMT), comprising 2 Transition Care Managers, who were shifted from an existing consult management RN team and thus were knowledgeable about most outpatient MH programs' admissions criteria. Prior to implementation, outpatient leaders involved in designing the process communicated to outpatient programs that discharged patients would have priority status for scheduling, and they asked every program to develop potential open access slots to book patients into more quickly if the next available regularly scheduled appointment was more than 14 days from discharge.
In the new process, with buy-in from RRTP leadership, providers at the RRTP were asked to place an MH Discharge Consult ideally at least 3 to 5 days prior to discharge for every discharging patient, but later if needed. The Transition Care Managers reviewed the Discharge Consults and communicated via messaging or phone with the discharging providers about questions or suggestions regarding potential outpatient MH options. When the appropriate outpatient MH program location was confirmed, the Transition Care Managers would place an electronic consult to that program. They would then call the patients prior to discharge to introduce themselves as a bridging contact during the transition period. Often during these calls, the RN would simultaneously conference in schedulers from the outpatient program to secure an appointment, with an effort made to confirm appointment dates and times prior to patient discharge. In cases that Transition Care Managers were not directly involved in scheduling, they monitored the date of the first postdischarge outpatient appointment and outreached to clinics if appointments did not fall within the 14 days postdischarge. As the process stabilized, this was later extended to 21 days.
Postdischarge, the Transition Care Managers called patients soon after discharge and often made additional outreach and reminder calls as necessary. In conjunction with the department data analyst, the Transition Care Managers monitored patients for 30 days postdischarge to track outpatient MH care engagement.
Transition Care Managers, a backup consultative physician, and the department data analyst huddled weekly to review outcome data and weekly process data and to adjust processes iteratively (Table). For instance, to increase use of the Discharge Consult, providers who were not placing the consult for their patients were identified and individually encouraged to do so as early as possible. The addition of balancing outcome measures in the data collection and analysis (discussed later) is another example of iterative adjustments to the process.
Table. -
Weekly Summary Data
|
Substrate—Number of discharges (weekly collection)
|
| Total # of weekly residential MH discharges |
|
Process—Use of consult (weekly collection)
|
| # of MH Discharge Consults placed last week |
| % of discharges with MH Discharge Consult placed |
| Average # of days between MH Consult placed and discharge date |
|
Process—Success in scheduling (weekly collection)
|
| # and % of discharges with appointment date and time in hand prior to discharge |
| # and % of patients with Discharge Consult placed during prior week with at least 1 MH appointment scheduled, with SSN-level data for fallouts |
| # and % of patients with Discharge Consult placed who have appointment date within 14 d of anticipated discharge date, with SSN-level data for fallouts |
|
Process—RN transition management: 30 d (weekly collection)
|
| # and % of discharges no-showing to first outpatient encounter, with SSN-level data for fallouts |
| # and % of discharges with zero MH encounters, with SSN-level data for fallouts |
| # and % of discharges with at least 1 MH encounter and where (program) |
| # and % of discharges with at least 2 MH encounters and where (program) |
|
Outcome—Success in transitioning patients to outpatient care (biweekly collection, 30 d after specified time period)
|
| # of discharges during this 2-wk period |
| # and % of all discharges with ≥2 outpatient MH visits within 30 d of discharge |
|
Outcome—Balancing measures to ensure quality patient care (biweekly collection, 30 d after specified time period)
|
| % of discharges meeting criteria during 2-wk period who met measure through homeless services or telephone care only, with SSN-level data for fallouts and associated visits (to assess for care gaps) |
Abbreviations: MH, mental health; SSN, social security number.
Weekly team huddles also allowed 3 data sources to be cross-referenced: the Corporate Data Warehouse, which is the VA national data warehouse, used primarily for appointment data and electronic consult data; the National Postdischarge Engagement Dashboard, which displays all MH discharges from the study site, allowing the team to “catch” discharged patients for whom Discharge Consults were not placed; and the Transition Care Managers' manually tracked data, which included information about patient no-shows and other outreach attempts.
Measures
The team used process and outcome measures to evaluate the impact of the pilot intervention. The number of discharges from the RRTP was tracked weekly. Process measures included the percentage of discharged patients for whom a Discharge Consult was placed, time from Discharge Consult placement to discharge date, and the percentage of discharges with a visit scheduled and “in hand” prior to discharge. Data were tracked weekly and were used to identify needed actions (eg, outreach to RRTP providers to place consult; outreach to clinics to request earlier appointments).
Outcome measures included the number of discharged patients with 2 or more MH encounters within 30 days of discharge, using data reported in the national MH postdischarge dashboard. A target of 80% success was set by the stakeholder team.
Balancing measures were included as a check to identify potential “false” successes or unintended consequences of the intervention. Since the national dashboard counted certain non-MH services such as homeless services, as well as the Transition Care Manager phone calls, as MH encounters, an additional balancing measure was developed to identify patients who met criteria through these services but lacked encounters with target outpatient MH programs. These data were drawn from VA Corporate Data Warehouse visit data. This measure allowed Transition Care Managers to flag such patients for additional review and outreach about MH needs. Data collection was refined for this measure in the course of the rollout. Data for these measures were collected in a 10-week preintervention baseline period for patients discharged from residential treatment, serving as a comparison.
Analysis
For the primary outcome of interest, the percentage of patients receiving 2 or more outpatient MH visits within 30 days of discharge from residential care, time series data were analyzed using a Statistical Process Control (SPC) XmR chart and P chart. In the SPC literature, XmR charts have been noted to provide a valid statistical approximation for attribute, or classification, data including percent pass/fail data for subgroups of unequal size traditionally graphed in P charts.14
SPC charts plot a series of measurements over time and use the inherent variation in the data to calculate the mean and limits of expected variation (upper and lower control limits, equal to mean ± 3 SDs).15 A stable process is that for which all measurements fall within the upper and lower control limits. As SPC charts plot measurements over a period of time, they can show whether a changing value is stable, as opposed to occurring due to chance.16 Descriptive data were also used to examine average weekly success rate pre- and postintervention, as well as to examine consult use and other measures.
Ethical considerations
This effort did not go through the institutional review board process as this was designed as a clinical improvement effort.
RESULTS
One-hundred forty MH Discharge Consults were placed during the initial 3-month pilot period starting in January 2020, and a total of 443 MH Discharge Consults were placed from program initiation through the end of the 55-week intervention period. On average, postdischarge consults were placed 2.7 days prior to discharge.
During a 10-week preintervention baseline period, the success rate across all weeks in connecting residential MH patients to 2 or more MH outpatient encounters in 30 days postdischarge was 52%; in the 55 weeks following the intervention, the success rate across all weeks was 89%. Weekly success rates preintervention ranged from 30% to 70%, while weekly success rates following the intervention ranged from 57% to 100%.
Weekly average success rate in patients having postdischarge MH appointments in hand prior to discharge was 80%. Data were tracked over 37 weeks, with tracking discontinued after stability. Patients were connected with 22 different outpatient MH programs within the health system and with programs at 4 other VA health systems.
Data were collected repeatedly over time, on a weekly basis. SPC X-bar and mR charts (see Supplemental Digital Content, Figures 1 and 2, available at: https://links.lww.com/JNCQ/A934 and https://links.lww.com/JNCQ/A935, respectively) and P chart data demonstrated sustained improvement soon after implementation. New control limits were calculated to identify the pattern established over the course of the 3-month pilot, showing a change in the mean from 53% to 90% and showing stable variation rather than uncommon cause variation following the pilot throughout the sustainment period. Upper and lower control limits were 1.2 and 0.54, respectively.
Balancing measures identified patients meeting the measure through clinic categorization codes indicating homeless care or general MH telephone care. After refining the data collection process, 10 weeks of preintervention data and 24 weeks of postintervention data were available. Prior to the intervention, the average weekly percentage of patients identified as potentially meeting criteria through homeless services and telephone care from Transition Care Managers rather than target outpatient MH care was 49%. Following the intervention, the weekly average percentage identified as potentially meeting the criteria through these sources was 18%. As part of the PDMT process, every chart identified weekly within this cohort was reviewed during the team huddle; some were identified as false-positives (ie, psychiatric care that occurred within a homeless services category code), while other patients were identified as having not yet connected with scheduled care or not responding to outreach calls. For those patients, additional outreach to patients and/or providers as appropriate was attempted.
DISCUSSION
In this effort, we found that implementation of a one-stop electronic MH Discharge Consult paired with a team of 2 MH Transition Care Manager RNs led to substantial improvement in rapid outpatient MH care engagement. Over 55 weeks, 89% of patients discharged from residential MH met criteria of 2 or more outpatient MH encounters within 30 days of discharge during the study period compared with 52% preintervention. SPC charts helped visualize the positive impact of the intervention and demonstrate maintained stability over time.
The development of a Transition Care Manager team created a designated source of bridging support for patients during the transition process as well as accountability to track patient engagement postdischarge. In addition to these strengths, data demonstrated high acceptability of a new “one-stop” Discharge Consult among residential staff, as shown by high utilization of the MH Discharge Consults and informal positive qualitative feedback. SPC charts ensured that the positive changes in postdischarge outpatient follow-up were not due to chance but rather were statistically meaningful changes over time.
Weekly scheduled review of process and outcome measures not only served as an accessible tool to evaluate success of the program regularly but also allowed for quickly identifying any steps in the process needing adjustment, in line with the continuous quality improvement approach. During weekly huddles, the team iteratively identified and planned to address problem areas in the transition process. For instance, the Transition Care Managers were able to identify individual providers not placing the consults and perform outreach to encourage use for more efficient scheduling by Transition Care Managers on the outpatient side. In addition, team members were able to identify certain outpatient clinics with long wait times and worked specifically with the programs to develop improved rapid access options.
In developing and adjusting this effort, we drew both from the QUERI framework for improving transitional care and from other quality improvement literature focused on transitions of MH care.9,10,17 A unifying theme in the literature was bridging contact pre- and postdischarge, which was a major strength of our program. As seen in prior transitions-of-care efforts, our program involved a multidisciplinary approach with the involvement of multiple stakeholders in development of the process.
Although not directly elucidated by the data, this project demonstrated emotional value to patients in creating greater continuity during the transition period; patients provided positive feedback about having a reliable touch point postdischarge. In addition, the Transition Care Managers supported patients to address barriers to care and provided appropriate resources outside of outpatient MH care in certain circumstances, including connecting patients with homeless services, dental services, neurology, and primary care.
While this effort was accomplished through reassignment of staff duties without adding additional staff members, replication of this effort might be hindered by lack of personnel available to fill PDMT roles as well as lack of buy-in from stakeholders, including inpatient/residential providers, outpatient providers and schedulers, or patients.
This effort presents a growth opportunity for MH nursing to take advantage of their specific skill set in care management of complex MH patients and play a leadership role in using data to effectively narrow gaps in care. The role of a Transition Care Manager could be applied across many different fields, settings, and care teams in health care. Locally, our team expanded this program to our inpatient MH settings after data analysis showed sustained improvement in the number of patients rapidly and successfully engaging with outpatient MH services after residential MH discharge.
REFERENCES
1. Valenstein M, Kim HM, Ganoczy D, et al. Higher-risk periods for suicide among VA patients receiving depression treatment: prioritizing suicide prevention efforts. J Affect Disord. 2009;112(1/3):50–58. doi:10.1016/j.jad.2008.08.020
2. Pfeiffer PN, Valenstein M, Ganoczy D, Henry J, Dobscha SK, Piette JD. Pilot study of enhanced social support with automated telephone monitoring after psychiatric hospitalization for depression. Soc Psychiatry Psychiatr Epidemiol. 2017;52(2):183–191. doi:10.1007/s00127-016-1288-2
3. Vigod SN, Kurdyak PA, Dennis C-L, et al. Transitional interventions to reduce early psychiatric readmissions in adults: systematic review. Br J Psychiatry. 2013;202(3):187–194. doi:10.1192/bjp.bp.112.115030
4. NCQA. Follow-up after hospitalization for mental illness. Published July 20, 2018. Accessed April 20, 2021.
https://www.ncqa.org/hedis/measures/follow-up-after-hospitalization-for-mental-illness
5. Marcus SC, Chuang C-C, Ng-Mak DS, Olfson M. Outpatient follow-up care and risk of hospital readmission in schizophrenia and bipolar disorder. Psychiatr Serv. 2017;68(12):1239–1246. doi:10.1176/appi.ps.201600498
6. Beadles CA, Ellis AR, Lichstein JC, et al. First outpatient follow-up after psychiatric hospitalization: does one size fit all? Psychiatr Serv. 2015;66(4):364–372. doi:10.1176/appi.ps.201400081
7. Harris AHS, Gupta S, Bowe T, et al. Predictive validity of two process-of-care quality measures for residential substance use disorder treatment. Addict Sci Clin Pract. 2015;10:22. doi:10.1186/s13722-015-0042-5
8. Schmidt EM, Wright D, Cherkasova E, Harris AHS, Trafton J. Evaluating and improving engagement in care after high-intensity stays for
mental health or substance use disorders. Psychiatr Serv. Published online June 9, 2021. doi:10.1176/appi.ps.202000287
9. Matarazzo BB, Forster JE, Hostetter TA, et al. Efficacy of the home-based
mental health evaluation (HOME) program for engaging patients in care after hospitalization. Psychiatr Serv. 2019;70(12):1094–1100. doi:10.1176/appi.ps.201900002
10. Wray AM, Hoyt T, Welch S, et al. Veterans Engaged in Treatment, Skills, and Transitions for Enhancing Psychiatric Safety (VETSTEPS). Psychiatr Rehabil J. 2019;42(3):277–283. doi:10.1037/prj0000360
11. Smith TE, Haselden M, Corbeil T, et al. Relationship between continuity of care and discharge planning after hospital psychiatric admission. Psychiatr Serv. 2020;71(1):75–78. doi:10.1176/appi.ps.201900233
12. Kansagara D, Chiovaro JC, Kagen D, et al.
Transitions of Care From Hospital to Home: An Overview of Systematic Reviews and Recommendations for Improving Transitional Care in the Veterans Health Administration. US Department of Veterans Affairs; 2015.
13. VA Office of
Mental Health. Domiciliary care for homeless veterans program. VA Homeless Programs. Accessed May 15, 2021.
https://www.va.gov/homeless/dchv.asp
14. Mohammed MA, Worthington P, Woodall WH. Plotting basic control charts: tutorial notes for healthcare practitioners. Qual Saf Health Care. 2008;17(2):137–145. doi:10.1136/qshc.2004.012047
15. Benneyan JC, Lloyd RC, Plsek PE. Statistical process control as a tool for research and healthcare improvement. Qual Saf Health Care. 2003;12(6):458–464. doi:10.1136/qshc.2004.012047
16. Clark M, Young T, Fallon M. Systematic review of the use of Statistical Process Control methods to measure the success of pressure ulcer prevention. Int Wound J. 2018;15(3):391–401. doi:10.1111/iwj.12876
17. Dixon L, Goldberg R, Iannone V, et al. Use of a critical time intervention to promote continuity of care after psychiatric inpatient hospitalization. Psychiatr Serv. 2009;60(4):451–458. doi:10.1176/ps.2009.60.4.451
