Voluntary retirement and resignations primary drivers of VHA loss of clinical employees

While the Veterans Health Administration (VHA) provided health care to about 6.7 million veterans in fiscal year (FY) 2015, it continues to face growing demand, a trend that is expected to continue. For example, the total annual outpatient medical appointments the VHA provided increased by 17.1 million visits (20 percent) from FY 2011 through 2015. To serve these veterans, in 2015, the VHA had 195,900 employees in 45 types of clinical occupations. And despite a number of target hiring initiatives by the VHA, the U.S. Senate’s Committee on Veterans’ Affairs and others have expressed concern about the VHA’s ability to maintain the appropriate clinical workforce to meet the needs of veterans, due to factors such as national shortages and increased competition for clinical employees.

After the Veterans Access, Choice, and Accountability Act of 2014 (P.L. 113-146) required the VA Office of Inspector General (OIG) to identify and report annually on the five VHA clinical occupations with the largest staffing shortages, in January 2015, the VA OIG reported that these occupations were physicians, registered nurses, physician assistants, psychologists, and physical therapists. The Committee on Veterans’ Affairs then asked the Government Accountability Office (GAO) to review the issues related to retention of clinical employees at VHA. Specifically, the GAO was asked to examine: (1) how the VHA collects and uses information on employees’ decisions to leave clinical occupations; and (2) what trends, if any, exist over the last five years in the number of, and reasons for, VHA employee losses from the five shortage occupations.

The GAO report found that the VHA collects data on the reasons clinical employees leave the agency and uses that data for planning purposes. For example, the VHA collects personnel data on the number of and reasons for clinical employee losses, including voluntary resignations, retirements, or removals due to adverse actions, in its HR databases and through a voluntary exit survey. The VHA uses this information to evaluate its workforce needs and inform its recruitment and retention efforts.

The GAO reported that the VHA losses for the five shortage occupations increased from about 5,900 employees in FY 2011 to about 7,700 in FY year 2015. Voluntary resignations and retirements were the primary reasons for the VHA’s losses for these occupations, with resignations accounting for about 54 percent, and retirements 36 percent of losses annually. Physician assistants consistently had the highest loss rate among the five shortage occupations. The loss rate for physician assistants increased from 9.3 to 10.9 percent during this period. The loss rate for physical therapists decreased from FY 2011 to 2012 (from 8.3 to 6.4 percent), but then increased to 8.0 percent in FY 2015. The GAO found a similar trend for all clinical occupations across the VHA.

In addition to its review of the VHA’s five shortage occupations, the GAO also identified the 10 clinical occupations within the VHA with the highest loss rates as of FY 2015.  These occupations included: physician assistant, medical support assistant, medical supply aide and technician, optometrist, nursing assistant, medical records technician, health technician (optometry), physician, medical records administration, and practical nurse. The loss rates for these 10 occupations ranged from 5.3 percent to 10.9 percent from fiscal year 2011 through 2015. Two of the five shortage occupations—physician assistants and physicians—were among the 10 highest loss-rate occupations each year from FY 2011 through 2015. In addition, two other occupations—medical support assistants and nursing assistants—were also consistently among the 10 highest loss-rate occupations each year during this period. The six remaining occupations were technical positions that were generally small in overall number, such as medical supply aides and technicians. Because employees in these occupations generally do not require specialized education or licensing, the VHA told the GAO that they tend to be more easily replaced than those in the five shortage occupations.

The GAO also found the VHA’s exit survey indicated that advancement issues or dissatisfaction with certain aspects of the work were commonly cited as the primary reasons respondents in the five shortage occupations left VHA. According to the exit surveys, the reasons why respondents in the five shortage categories left the VHA can be summarized as follows:

  • 28 percent said that advancement and 21 percent said that dissatisfaction with certain aspects of the work, such as concerns about management and obstacles to getting the work done, was the primary reason they were leaving.
  • 71 percent said that a single event generally did not cause them to think about leaving, while 28 percent reported that it did.
  • 50 percent indicated that they were generally satisfied with the quality of senior management, while 31 percent were not.
  • 73 percent felt that their immediate supervisors treated them fairly at work, while 15 percent reported that they did not.
  • 67 percent felt that they were treated with respect at work, while 19 percent reported they were not.
  • 50 percent reported that one or more benefits would have encouraged them to stay, such as alternative or part-time schedules (25 percent) or student loan repayment or tuition assistance (12 percent).

The GAO made no recommendations in its report. The VA, however, provided written comments, citing a historical context for loss rate trends. For example, they noted that loss rates decreased during the economic downturn of 2008 to 2009 and that current loss rates represent a return to the rates seen in FYs 2006 to 2007. The VA also noted that the VHA’s workforce challenges mirror those of the health care industry, including the growing national shortage of physicians and nurses and increased competition for health care professionals in hard-to-fill occupations. The VA further suggested that the GAO use other, non-VA data sources to provide additional context.