Senator Isakson, thank you for the opportunity to discuss the work of the Office of Inspector General (OIG) regarding mental health services provided by VA through the Atlanta VA Medical Center (VAMC) in Decatur, Georgia. The focus of our statement will be two reports issued on April 17, 2013, Healthcare Inspection – Mismanagement of Inpatient Mental Health Care, Atlanta VA Medical Center, Decatur, Georgia, and Healthcare Inspection – Patient Care Issues and Contract Mental Health Program Mismanagement, Atlanta VA Medical Center, Decatur, Georgia. These two reports highlight deficiencies in administration of the acute Mental Health (MH) inpatient unit and deficiencies in administration, tracking, and monitoring of contract mental health services and the clinical impact. I am accompanied by Mr. Murray Leigh, Director of the Healthcare Financial Analysis Division, Office of Inspector General.
The Atlanta VAMC has a 40-bed acute inpatient mental health unit that is locked and admits only voluntary patients. During fiscal year (FY) 2012, the unit’s average daily census was 34 with an occupancy rate of 84 percent and an 11-day average length of stay. Unit leadership consisted of a Medical Director, Nurse Manager, and Social Work Supervisor. The unit Medical Director reports directly to the Chief, Mental Health Service Line (MHSL) who in turn reports to the facility Chief of Staff. The Nurse Manager reports directly to the Associate Nurse Executive, Nursing Care Unit/MH, and the unit’s lead Social Worker reports to the unit Medical Director.
With regards to outpatient care for mental health services, in 2008, Veterans Integrated Service Network (VISN) 7, which includes the Atlanta VAMC, established a contract with Select Systems LLC (SELECT), an affiliate of the Georgia Association of Community Service Boards (CSB) to provide general outpatient mental health services, crisis stabilization, and psychosocial rehabilitation/day treatment to patients referred by any of the eight VAMCs in VISN 7. Twenty-six CSBs provide mental health care as subcontractors under the contract with SELECT.
In 2011, the OIG substantiated allegations that several mental health clinics at the Atlanta VAMC had significantly high numbers of patients on their electronic wait lists (EWL) over a period of months in FY 2010. The facility managers were aware of the high number of patients in need of mental health services which would further add to the EWL, but were slow in taking actions to address the situation.
Due to that report, the facility increased utilization of the SELECT contract. The patients on the General Mental Health Clinic EWL appeared to initially decrease. The facility reported referring 4,000–5,000 patients to CSBs for mental health treatment since the initiation of the contract in 2008. In FY 2012, the facility estimated that the community provided 25–33 percent of all outpatient care for the 15,000 mental health outpatients enrolled at the facility. Administratively, patients would remain on the EWL until the facility received confirmation that a patient attended their initial CSB appointment.
MISMANAGEMENT OF INPATIENT MENTAL HEALTH CARE
The OIG conducted this inspection based on an allegation received by the OIG Hotline that the Atlanta VAMC MHSL leadership’s failure to establish effective unit policies, ensure monitoring of unit patients, staff the unit appropriately, and care about patients contributed to the death of a patient on the inpatient mental health unit at the facility. We substantiated that facility and MHSL policies for the inpatient mental health unit did not sufficiently address patient care safety. Specifically, we found that the facility did not have adequate policies or practices for contraband, visitation, urine drug screening, or provider notification of clinical changes in a patient’s condition.
The patient attended an ophthalmology appointment at a clinic on a different floor from the locked inpatient mental health unit during which he was unsupervised for several hours. After returning to the unit, he appeared drowsy. A urine drug screen was obtained but was unobserved. Early the next morning the patient was found dead from an overdose of medications that had not been prescribed for the patient while admitted to the unit.
Drug Screening Policy
Substance use is a common co-morbid diagnosis for patients admitted to inpatient mental health units. Therefore, the policies and practices needed for monitoring patients on locked mental health units should also be inclusive of those needed for treatment of substance use, such as observed urine drug screens (UDS) and thorough contraband searches. The validity of UDS results is dependent on specimen integrity. Patients who know their urine will be positive for drugs might attempt to alter the specimen to be tested through dilution or substitution with a clean urine specimen from another person. Staff observation of urine collection is most likely to ensure an accurate sample. Non-witnessed collections can be effective if safeguards, such as searching the donor and collection site, are in place to prevent alteration of the specimen.
We determined the facility had not provided staff with a UDS policy or training prior to the subject patient’s death. The day after the patient’s death, another inpatient reported to staff that he had provided the patient with a urine sample, after the subject patient asked him for a clean urine specimen for his probation officer. The facility implemented supervised UDS collection after this event.
Following the patient’s death, another patient acknowledged that he obtained medication from visitors and that he had shared this medication with the patient. Prior to the patient’s death the unit had no written MHSL policies addressing patient visitation. Staff would ask visitors to sign in on a log, limit items brought onto the unit, and limit visitor’s presence to a large group room. Staff reported inconsistent adherence to these practices. Subsequent to the patient’s death, a new policy was introduced and unit nurses changed their practice to require that visitors be limited in number, sign in on the unit log, and remain in smaller group rooms under staff observation. The facility designated two existing lockers on the unit to secure visitor items with the plan to purchase additional lockers for outside the unit.
Observation levels for patients on the inpatient mental health unit range from 1:1 observation in which staff is at arm’s length from the patient at all times; direct observation in which staff have direct visualization of a patient at all times; checks every 15 minutes, checks every 30 minutes, and at the lowest level of observation, checks every 60 minutes. Patients are escorted to appointments by mental health unit staff or patient escort staff. Patients on 30 or 60-minute checks can be left unmonitored off unit, while patients on 15-minute checks or higher patient observation levels must be accompanied by staff when off unit.
Regardless of patient observation level, unit staff are required to document their observations using paper flow sheets that are later scanned into the electronic health record. We found discrepancies in staff documentation regarding patient observations that raised questions regarding the credibility of the information entered. On the day of the subject patient’s death, he was off the unit at an ophthalmology appointment from 3:00 p.m. to 5:00 p.m. but the observation flow sheet showed him to be on the unit with 30-minute checks and that staff offered him a meal at 4:30 p.m.
Following the patient’s death, nursing leadership implemented three initiatives to improve flow sheet accuracy. First, they attached a photo of each patient to the flow sheets; second, they began random quality checks of the unit’s observation flow sheets; and third, conducted a time study to identify multi-patient observation by employees, and subsequently adjusted policies and secured approval for additional staff.
During our September 2012 site visit, unit staff reported that patient observation level practices for off-unit appointments remained unchanged, in spite of their requests for change. However, during our February 2013 site visit, unit leadership reported that practices were changed shortly after the subject patient’s death to limit off-unit appointments to medical emergencies and discharge needs, and to require that only mental health unit staff escort patients. The noted discrepancy between staff and leadership’s understanding of changes to unit practices is a reflection on the state of communication on the unit.
Root Cause Analysis and Patient Incidents
We had concerns about the document review and timeliness of follow-up actions recommended by the facility’s internal Root Cause Analysis (RCA). As VA requires, the facility initiated an RCA following the subject patient’s death. The RCA was conducted in accordance with guidance set forth by VA’s National Center for Patient Safety (NCPS). The appointed RCA team members included mental health unit staff who were either peers or in an authoritative role on the unit. While we understand that such a team would have insights into the unit’s operation, it is our opinion that some team members, as process owners, could have been limited in their ability to recognize unit problems. The RCA team identified root causes that may have contributed to this adverse event. We found that in the 2 months following the completion of the RCA, the facility leaders responsible for the recommendations had not initiated the follow-up action for all items.
We also found inadequate program oversight including a lack of follow-up actions by leadership in response to patient incidents. We interviewed numerous employees who voiced frustration in the general lack of leadership action taken on the unit when adverse events occurred, particularly following the subject patient’s death. The day after the patient’s death, the unit Medical Director led an interdisciplinary staff meeting that identified areas of improvement for patient safety. Staff described the meeting as productive but expressed concern that leadership had not implemented many of the suggested recommendations or updated them on progress.
We recommended that the Under Secretary for Health ensure that VHA develops national policies that address contraband, visitation, urine drug screens, and escort services for inpatient mental health units.
We also made seven recommendations to the VISN 7 Director and the VAMC Director to ensure that:
• The facility inpatient mental health unit develop and implement policies that adequately address contraband, visitation, urine drug screening, and escort service.
• The facility inpatient mental health unit employ safeguards for documentation that accurately reflect staff observation of patients.
• The facility inpatient mental health unit strengthen program oversight including follow-up actions taken by leadership in response to patient incidents.
• The facility strengthen and improve the RCA process to ensure that all information and documentation related to the event are reviewed and that follow up actions are completed and timely.
• The facility improve communication with staff regarding debriefings and planned actions to address identified deficiencies.
• The facility inpatient mental health units are equipped with functional and well-maintained life support equipment.
• The facility evaluate the care of the subject patient with Regional Counsel for possible disclosure(s) to the appropriate surviving family member(s) of the patient.
The Under Secretary for Health and the Veterans Integrated Service Network and Facility Directors concurred with our recommendations. Three of eight recommendations remain open as of August 1, 2013. The OIG will monitor VA’s planned actions to implement outstanding recommendations until they are completed.
PATIENT CARE ISSUES AND CONTRACT MENTAL HEALTH PROGRAM MISMANAGEMENT
In this report, the OIG reviewed allegations reported to the OIG Hotline that the Atlanta VAMC mismanaged and did not provide oversight of a mental health contract. During our inspection, staff brought forth additional allegations that there was inadequate coordination, monitoring, and staffing for oversight of contracted mental health patient care. We substantiated the original allegation of mismanagement in the administration of the contract as well as the additional allegations raised when we were onsite.
Contract Liaisons between the Atlanta VAMC and the CSBs
The facility’s three contract liaisons were to serve as coordinators for patients referred for mental health treatment while in transition between the facility and CSBs. According to the facility’s functional statement, contract liaison responsibilities include clinical oversight, treatment plan development in collaboration with the CSBs, site visits, problem resolution, and intermediary duties between the facility and the CSBs.
Contract Administration and Management
This was a complex contract that involved the VISN 7 Contracting Officer, VISN 7 Mental Health Administrative Officer, facility contracting officer technical representative (COTR), contract liaisons, the contractor, and staff from the 16 CSBs used by the facility (as of June 2012). As more patients were referred on the contract, the administrative and managerial difficulties increased. The contract was not specific on documentation requirements for patient coordination, invoicing, or notification. The lines of communication and responsibilities were not sufficiently clear to handle the many issues that frequently occurred.
The three contract liaisons spent most of their time reacting to emergent CSB questions and patient coordination problems, and were not able to do many of their other responsibilities such as tracking referrals, reviewing patient records, approving the authorizations, and certifying monthly invoices for each of the CSBs. As the number of patients referred to the CSBs continued to grow, the facility staff had increasing difficulty processing the claims and invoices. Furthermore, the facility did not promptly communicate the reasons for payment denial to the CSBs. This delayed resolution of the billing problems and prompted some CSBs to refuse acceptance of new patients.
In some cases when the CSBs provided clinical progress notes along with invoices, the facility did not consistently scan and upload available CSB progress notes into the patient’s EHR. However, the staff at the four CSBs whom we interviewed reported that the facility had significant communication gaps regarding changes in requirements or notification of missing documentation.
Because the facility did not reliably know the number of patients receiving services on the SELECT contract, facility administrators could not adequately estimate the required funding. The facility’s budget tracking sheet showed that $6.7 million was originally budgeted for FY 2012, but an additional $3.2 million was required to fund the contract for the remainder of the fiscal year. However, we did not find evidence that the available funding had been an issue in making payments to the CSBs.
Oversight of Patients Referred to CSBs for Mental Health Care
The focus in the inspection was on what the facility did or did not do in regards to coordination and oversight of the contracted mental health patient care. We did not assess the quality of mental health care provided by the CSBs, which maintain accreditations from the Commission on Accreditation of Rehabilitation Facilities.
The facility did not establish an effective coordination, tracking, and monitoring system for oversight of patients referred under the contract and MHSL managers did not adequately oversee or monitor contracted patient care services to ensure safe and effective treatment. Most noticeably, program managers were unable to sufficiently identify and track the enrolled CSB patients. Many times the facility did not know who was scheduled for CSB appointments or if patients had or had not been seen by the CSB providers.
Among patients referred but not seen, we found a patient who committed suicide approximately 6 months after initial referral and a second patient who had reported suicidal ideation and subsequently committed suicide within 3 weeks of referral. A third patient was hospitalized approximately 4 months after initial referral, then re-hospitalized 1 month later. The patient was incarcerated a few weeks later.
The process for scheduling initial CSB-appointments was ambiguous in that appointments could be scheduled by the patient, the facility, or by the CSBs. We reviewed a sample of 85 EHRs from a list provided by the facility of approximately 1,500 CSB-referred patients. We found that in 21 percent of our random sample the facility failed to coordinate necessary mental health services for this at-risk population in that care had not been established at the CSBs and no mental health follow-up was provided by the facility.
In addition, 74 percent of the sample of CSB-referred patients had wait times greater than 14 days, with a wait time average of 92 days and a median of 56 days (ranging from 5 to 432 days). VHA policy requires that an initial mental health appointment be scheduled within 14 days of a referral. We also found that during the wait period for the initial CSB appointment, there was typically no clinical contact between the patient and the Atlanta VAMC, thereby further increasing the potential for patients to “fall through the cracks.” The facility subsequently developed a process to track CSB-referred patients in January 2012.
Our review also confirmed that facility managers did not provide adequate staff, training, resources, support, and guidance for effective oversight of the contracted mental health program. Facility managers estimated that the Mental Health Assessment Team was referring up to 60 new patients each week to the CSBs. At the time of our review, the facility had assigned approximately 10 employees (some with collateral duties) to manage and provide oversight for over 4,000 patients referred to CSB programs. With limited staff, contract liaisons were unable to monitor patient progress, provide adequate clinical oversight, and collaborative treatment planning with the CSB staff.
Quality Assurance for Contracted Mental Health Care
We found the facility’s contract program lacked an integrated and effective Quality Assurance (QA) program and did not have a CSB-specific QA process. For example, VA facility program managers did not track and trend patient complaints or conduct oversight visits to the CSB sites as required by VA directives and the contract. On January 31, 2013, the contract with SELECT expired. Facility managers negotiated a short-term (8-month) contract to cover the need for community mental health services until a longer-term contract is negotiated and new staff is recruited and additional mental health clinics are opened at the facility. A QA matrix was proposed for possible inclusion in the new contract.
We recommended the Under Secretary for Health take note and rectify the deficiencies described in the report with respect to the provision of quality mental health care and contract management, with the goal that veterans receive the highest quality medical care from either the VA or its partners. We also recommended the Facility Director evaluate the care of patients discussed in this report with Regional Counsel for possible disclosure(s) to the appropriate surviving family member(s) of the patients.
The Under Secretary for Health, the VISN Director, and the VAMC Director concurred with our recommendations. The OIG will monitor VA’s planned actions to implement outstanding recommendations until they are completed.
Our 2011 report found a large number of patients waiting for mental health treatment and the facility's slow response to address the problem. We found the results of these recent inspections extremely troubling. We understand that since the issuance of these reports the Atlanta VAMC has instituted several changes. We will actively follow up on VA’s planned actions to implement our recommendations and monitor progress until all corrective actions have been taken.
Senator Isakson, thank you again for this opportunity to testify. We would be pleased to answer questions that you may have.
Table of Contents