This is the accessible text file for GAO report number GAO-09-17 entitled 'VA Health Care: Improved Staffing Methods and Greater Availability of Alternate and Flexible Work Schedules Could Enhance the Recruitment and Retention of Inpatient Nurses' which was released on October 24, 2008. This text file was formatted by the U.S. Government Accountability Office (GAO) to be accessible to users with visual impairments, as part of a longer term project to improve GAO products' accessibility. Every attempt has been made to maintain the structural and data integrity of the original printed product. Accessibility features, such as text descriptions of tables, consecutively numbered footnotes placed at the end of the file, and the text of agency comment letters, are provided but may not exactly duplicate the presentation or format of the printed version. The portable document format (PDF) file is an exact electronic replica of the printed version. We welcome your feedback. Please E-mail your comments regarding the contents or accessibility features of this document to Webmaster@gao.gov. This is a work of the U.S. government and is not subject to copyright protection in the United States. It may be reproduced and distributed in its entirety without further permission from GAO. Because this work may contain copyrighted images or other material, permission from the copyright holder may be necessary if you wish to reproduce this material separately. Report to Congressional Requesters: United States Government Accountability Office: GAO: October 2008: VA Health Care: Improved Staffing Methods and Greater Availability of Alternate and Flexible Work Schedules Could Enhance the Recruitment and Retention of Inpatient Nurses: GAO-09-17: GAO Highlights: Highlights of GAO-09-17, a report to congressional requesters. Why GAO Did This Study: Registered nurses (RNs) are the largest group of health care providers employed by VA’s health care system. RNs are relied on to deliver inpatient care, but VA medical centers (VAMC) face RN recruitment and retention challenges. VAMCs use a patient classification system (PCS) to determine RN staffing on inpatient units by classifying inpatients according to severity of illness to determine the amount of RN care needed. GAO reviewed VAMC inpatient units for (1) the usefulness of information generated by VA’s PCS; (2) key factors that affect RN retention; and (3) factors that contribute to delays in hiring RNs. GAO performed a Web-based survey of all VAMC nurse executives; interviewed VA headquarters officials and VAMC nursing officials, and conducted RN focus groups at eight VAMCs visited by GAO. The findings of GAO’s survey are generalizable to all nurse executives; however, findings from the focus groups at the eight VAMCs are not generalizable. What GAO Found: VAMC nursing officials—nurse executives who are responsible for all nursing care at VAMCs and nurse managers who are responsible for supervising RNs on VAMC inpatient units—GAO interviewed reported that although VA inpatient RNs are required to input patient data into VA’s PCS, they do not rely on the information generated by PCS because it is outdated and inaccurate. These nursing officials noted that VA’s PCS does not accurately capture the severity of patients’ illnesses or account for all the nursing tasks currently performed on inpatient units. Because of the shortcomings of VA’s PCS, nurse managers use data from a variety of sources to help set RN staffing levels for their inpatient units. At four of the eight VAMCs GAO visited, nurse managers told GAO that they set RN staffing levels for their inpatient units by adhering to the historical staffing levels that had been established for the units. Three VAMCs GAO visited set their RN staffing levels using data on the RN staffing levels found in inpatient units in other hospitals with similar characteristics. VA reported it is proposing to develop a new RN staffing system. However, VA has not developed a detailed action plan that includes a timetable for building, testing, and implementing the new nurse staffing system. VA nursing officials reported that VA’s ability to retain its RNs is adversely affected by two main factors. First, inpatient RNs reported that they spend too much time performing non-nursing duties such as housekeeping and clerical tasks. Second, even though VAMCs were authorized in 2004 to offer RNs two alternate work schedules that are generally desired by nurses—such as working three12-hour shifts within a week that would be considered full-time for pay and benefits purposes—few nurse executives reported offering these schedules; therefore, few RNs work these schedules. Specifically, according to nurse executives GAO surveyed only about 1 percent of many inpatient units offered alternate schedules and less than 1 percent of RNs actually worked these schedules. The availability of flexible work schedules, for example, working eight 10-hour shifts over a 2-week period, are more widely available among VAMCs but are still limited, according to GAO’s survey of nurse executives. Nursing officials and RNs noted other factors affecting retention such as reliance on supplemental staffing strategies—for example, RN overtime—and insufficient professional development opportunities. Both VA nurse executives and nursing officials identified limitations in VA’s process for hiring RNs and VA-imposed hiring freezes and lags as major contributing factors causing delays in hiring RNs to fill inpatient vacancies at VAMCs. VA nursing officials reported that hiring freezes and lags at VAMCs and delays resulting from limitations in VA’s hiring process can discourage prospective candidates from seeking or following through on applications for employment at these facilities. Although VA has recently taken steps to address some of the factors that are reported to contribute to RN hiring delays, it is too early to determine the extent to which these steps have been effective in reducing hiring delays. What GAO Recommends: GAO recommends that VA develop an action plan to implement a new nurse staffing system that ensures an accurate account of patient care needs and tasks performed by RNs and that VA assess the barriers to wider availability of alternate and flexible work schedules and explore ways to overcome these barriers. VA concurred with GAO’s findings and recommendations and plans to address GAO’s three recommendations. To view the full product, including the scope and methodology, click on [hyperlink, http://www.gao.gov/cgi-bin/getrpt?GAO-09-17]. For more information, contact Randall B. Williamson at (202) 512-7114 or williamsonr@gao.gov. [End of section] Contents: Letter: Results In Brief: Background: VAMC Nursing Officials Cannot Rely on Information from VA's PCS to Determine RN Staffing Levels for VA Inpatient Units: VAMC Nursing Officials and RNs Identify Several Factors That Could Adversely Affect Retention of Inpatient RNs: VAMC Nursing Officials Identified Limitations in VA's Hiring Process and VA-Imposed Hiring Freezes as Contributing Factors to RN Hiring Delays: Conclusions: Recommendations for Executive Action: Agency Comments and Our Evaluation: Appendix I: Scope and Methodology: Appendix II: Analysis of GAO Survey of VA Medical Center Nurse Executives: Appendix III: Summary of RN Focus Group Questions and Responses at the Eight VAMCs We Visited: Appendix IV: Comments from the Department of Veterans Affairs: Appendix V: GAO Contact and Staff Acknowledgments: Related GAO Products: Tables: Table 1: Percentage of VA Nurse Executives Reporting Ancillary Support Staff's Availability around the Clock: Table 2: Summary of Actions Taken by Eight VAMCs We Visited to Reduce Delays in Filling RN Vacancies: Table 3: Demographic Profile of RN Focus Group Attendees at Eight VAMCs We Visited: Abbreviations: AACN: American Association of Colleges of Nursing: ANCC: American Nurses Credentialing Center: AHRQ: Agency for Healthcare Research and Quality: BCMA: bar code medication administration: CPRS: computerized patient record system: EK: Gelectrocardiogram: HPPD: hours per patient day: HR: human resources: LPN: licensed practical nurse NA nursing assistant: OIG: Office of Inspector General: ONS: Office of Nursing Services: OPM: Office of Personnel Management: PAID: personnel accounting integrated data: PCS: patient classification system: RN: registered nurse: VA: Department of Veterans Affairs: VAMC: Veterans Affairs medical center: VISN: Veterans Integrated Service Network: United States Government Accountability Office: Washington, DC 20548: October 24, 2008: The Honorable Daniel K. Akaka: Chairman: Committee on Veterans' Affairs: United States Senate: The Honorable Ken Salazar: United States Senate: In 2007, the Department of Veterans Affairs (VA) employed almost 43,000 registered nurses (RN) who provided care to veterans in VA medical centers (VAMC) across the country. RNs are the largest group of health care providers employed by VA's health care system, and VA relies on the services provided by these nurses. Maintaining its RN workforce is critical to VA's provision of care to its veteran population. Studies in general have shown, for example, that a shortage of RNs, especially when combined with increased patient workload, can adversely affect patient outcomes and therefore the quality of care patients receive.[Footnote 1] Hospitals with lower levels of RN staffing have been shown to have higher rates of adverse events such as urinary tract infections and pneumonia. Conversely, an increase in RN staffing has been associated with a reduction in adverse patient care events and with better quality outcomes such as fewer infections.[Footnote 2] To ensure adequate staffing at VAMCs, Congress passed legislation in 2002 requiring VA to develop a national staffing policy, including a policy on nurse staffing. The law required that VA establish a policy on the staffing levels and expertise required to provide care to veterans at VAMCs. The need for a new nurse staffing system was also highlighted in 2004 when the VA Office of Inspector General (OIG) issued a report that raised concerns about the adequacy of RN staffing levels on VA's inpatient units and identified instances in which bed closures or diversions of patients to other inpatient facilities in the community resulted in part because of inadequate staffing.[Footnote 3] The VA OIG further found that staffing challenges at VAMCs resulted from the lack of a consistent staffing methodology and a lack of effective recruitment and retention strategies.[Footnote 4] To address the staffing issues raised in the OIG report and in light of the 2002 law, VA committed to design and implement a new nurse staffing system, which it has not yet completed. As noted in the 2004 VA OIG report, to avoid inadequate RN-staffing levels and the consequences for patient care, two issues are important: determining the right number and skill mix of RNs required to care for veterans, and recruiting and retaining a sufficient supply of RNs to meet required staffing levels. In determining the right number and skill mix of RNs needed to care for patients on inpatient units, nurse managers in hospitals typically consider various factors, including the number of patients and their care needs based on the seriousness of their illnesses, experience and qualifications of the RN staff, availability of support services such as housekeeping and patient transport, and budgetary factors. In considering these factors, many hospitals use a patient classification system (PCS) as a key tool to help nurse managers estimate the amount of nursing care needed by inpatients. By using such a system, nurse managers can more accurately and consistently predict RN staffing.[Footnote 5] Over time, more of the patient care at VAMCs has shifted from inpatient to outpatient settings, and veterans are older and sicker and therefore have more intensive nursing care needs. Like other hospitals with inpatient care units, the VA has a PCS that is intended to be used to help determine appropriate RN staffing levels to meet the nursing care needs on VAMC inpatient units. Established in the 1980s for use by its VAMCs, VA's PCS is a computer-based system that is designed to predict the RN staffing levels needed on inpatient units based on patient acuity, which is a measure of the type and severity of illness. Specifically, VA's PCS produces an estimate of RN hours per patient day (HPPD) needed by inpatient units to care for their patients. The PCS estimates are based on data that is entered into the PCS system on a daily basis on the number of patients in each unit and each patient's level of acuity. This estimate can then be used to set RN staffing levels. With respect to retaining the RNs it currently employs and recruiting new RNs, VA faces substantial challenges. In congressional testimony in 2007, a VA official discussed the challenges the department faces in retaining RNs in VA's workforce, in part, because many VA RNs are nearing retirement eligibility age.[Footnote 6] VA data show that in fiscal year 2005, 13 percent of VA's RNs were eligible for retirement, and VA projects this number to increase to 22 percent by 2012.[Footnote 7] According to the Office of Personnel Management (OPM), flexible work schedules allow workers to balance their work and personal commitments. To enhance retention of VA RNs, the Congress passed legislation authorizing VA in 2004 to allow its VAMCs to offer RNs two alternate work schedules that would allow RNs either to work three 12-hour shifts per week (and get paid for 40 hours) or work 9 months each year with 3 months off (and get paid 75 percent of their salary year round). Further, VA, like other health care providers, also faces challenges recruiting new RNs in order to maintain its RN workforce levels. In its 2007 to 2011 workforce succession strategic plan, VA identified recruiting RNs as its top recruitment priority, and in 2007 congressional testimony, a VA official reported that the department had taken steps to streamline VA's hiring process to address delays in filling vacant RN positions. VA's focus on recruiting RNs has occurred in the context of a nationwide RN shortage, which has further exacerbated VA's RN recruitment challenges. For example, the Health Resources and Services Administration--an agency within the Department of Health and Human Services--has projected that by the year 2020, 44 states will experience a shortage of nurses, and that by 2020, the nation's RN shortage will grow to more than one million nurses. This report provides information on current RN staffing practices on inpatient units at VAMCs.[Footnote 8] We specifically discuss (1) how useful the information generated by VA's PCS is for determining RN staffing levels on inpatient units, (2) key factors that VAMC nursing officials and RNs identify that affect RN retention on inpatient units, and (3) factors that VAMC nursing officials identify as contributing to delays in hiring RNs to fill vacant positions. To fulfill our objectives, we conducted a Web-based survey of VAMC nurse executives[Footnote 9] and interviewed VA headquarters officials in the Office of Nursing Services (ONS), VAMC nursing executives, VAMC inpatient unit nurse managers,[Footnote 10] and VAMC human resources (HR) officers. In addition, we conducted inpatient RN focus groups at eight VAMCs, obtained data as of April 2008 on the number of VA RNs who use alternate work schedules, interviewed representatives of state hospital associations, and reviewed and analyzed industry and agency reports and studies. For the purpose of this report, we use the term VAMC nursing officials to include VAMC nurse executives and inpatient unit nurse managers. In conducting the Web-based survey, we surveyed all VA nurse executives at VAMCs. The survey was sent to 140 VAMC nurse executives and obtained a 63 percent response rate, which allows us to generalize the results to all nurse executives at VAMCs. Based on our review of the completed surveys, we determined that the data were sufficiently reliable for our purposes. In selecting VAMCs as sites to conduct our interviews and focus groups, we judgmentally selected eight VAMCs located in Denver, Colorado; Houston, Texas; Minneapolis, Minnesota; New York, New York; Portland, Oregon; Seattle, Washington; Tampa, Florida; and Togus, Maine. We selected these VAMCs in order to include in our review the various types of inpatient units found at these medical centers such as intensive care, surgery, and medicine units. At the eight VAMCs we visited, 219 inpatient RNs from three shifts (day, evening, and night) attended the focus groups. Attendees at our focus groups included RNs of different ages, nurse experience levels, and length of tenure at VA. The results of our analyses at these VAMCs are not generalizable because the VAMCs selected are not necessarily representative of all VAMCs. The information presented in our focus group summaries accurately capture the opinions provided by the inpatient RNs who attended the focus groups at the eight VAMCs we visited. However these opinions cannot be generalized to all inpatient RNs at the eight VAMCs we visited or to all inpatient RNs at VAMCs. We performed a systematic review of the completed Web-based survey of VA nurse executives to assess the reliability of the data obtained from the survey. We checked each survey for problems such as key questions left unanswered, patterns of skipped questions, unclear written responses, and out of scope entries. We also assessed the reliability of data obtained from VA headquarters officials related to the number of VA RNs who use alternate work schedules. We contacted VA headquarters officials, who provided information on the quality checks they performed on these data. Based on our review and the information provided to us from VA officials, we determined that the data we used in our report were adequate for our purposes. For a detailed description of our scope and methodology, see appendix I. Selected results from our survey and a summary of responses from our focus groups are provided in appendix II and appendix III. We conducted this performance audit from May 2006 through September 2008 in accordance with generally accepted government auditing standards. Those standards require that we plan and perform the audit to obtain sufficient, appropriate evidence to provide a reasonable basis for our findings and conclusions based on our audit objectives. We believe that the evidence obtained provides a reasonable basis for our findings and conclusions based on our audit objectives. Results In Brief: Although VAMC nursing officials on VAMC inpatient care units said they are required to enter patient data into VA's PCS, they stated that the information generated by the PCS cannot be relied on because the PCS is outdated and produces inaccurate results. PCS was developed in the 1980s based on nursing practices and technology at that time, and according to nurse managers, it does not account for all of the nursing tasks currently performed on inpatient units and does not always accurately capture the level of nursing care required for patients with serious illnesses. For example, PCS does not account for the administration of certain intravenous medications or monitoring of a patient's abnormal heart activity, which were once limited largely to intensive care units that cater to sicker patients but are now performed on other inpatient units. As a result, nurse managers across all VAMCs use various other data in different degrees to help set RN staffing levels for their inpatient units. Nurse managers at four VAMCs we visited told us that they set RN staffing levels for their inpatient units by adhering to long-standing historical staffing levels that were established for the units sometimes more than a decade ago. In contrast, at three other VAMCs we visited, RN staffing levels are set by benchmarking their RN staffing levels against the levels found in inpatient units in other hospitals with similar characteristics. VA headquarters officials reported that they are developing a better and more standardized system to determine the RN staffing levels needed on inpatient units at VAMCs. In developing this system, VA is considering updating or replacing its current PCS, according to a VA official. However, VA did not provide a detailed action plan with specific timelines for the development, testing, and implementation of the new staffing system. VAMC nursing officials and inpatient RNs we interviewed reported that two main factors adversely impact RNs' job satisfaction and morale, and could ultimately adversely impact VA's ability to retain RNs. First, according to inpatient nurse managers we interviewed, some RNs are dissatisfied about spending too much time performing clerical and non- nursing tasks, such as answering telephones, changing bed linen, and drawing blood samples. Second, both VAMC nursing officials we interviewed and inpatient RNs in our focus groups told us that the limited availability of alternate and flexible work schedules affects the ability of RNs to balance work and personal commitments. While VAMCs are authorized to offer two alternate work schedules that are generally desired by RNs, according to nurse executives, only about 1 percent of many inpatient units reported offering these schedules; consequently, few RNs (less than 1 percent) work these schedules. The availability of flexible work schedules--for example, working 10-hour shifts in a 2-week period--are more widely available at VAMCs but are still limited, according to our survey of nurse executives. VAMC nursing officials and inpatient RNs also cited other factors that affect job satisfaction and morale and could ultimately adversely affect retention. These factors include reliance on supplemental staffing strategies, such as RN overtime, and insufficient professional development opportunities. VAMC nursing officials we interviewed reported three factors that contribute to delays in hiring RNs to fill vacancies at VAMCs--(1) delays in securing necessary approvals from medical center officials to fill RN vacancies, (2) poor coordination between nursing and human resources (HR) officials involved in the RN hiring process, and (3) a shortage of experienced and well-trained HR officials. Collectively, these factors result in significant delays in filling RN vacancies. For example, 44 percent of the nurse executives we surveyed reported that it took 45 to 80 days to fill inpatient RN vacancies at VAMCs in 2007 compared to the 24-to 45-day target that VA set for 2007. In contrast, VAMC nursing officials we interviewed said that local hospitals usually hired RNs in less than 21 days. The time it takes to fill inpatient RN vacancies at VAMCs can also be affected by hiring freezes, which are sometimes caused by medical center budgetary constraints. Nursing officials at VAMCs we visited told us that these delays contribute to VAMCs' losing applicants to local hospitals and increasingly adopting supplemental staffing strategies, such as RN overtime, to maintain RN staffing levels on inpatient units. In 2007, a VA Recruitment Process Redesign Workgroup identified barriers and delays in hiring and proposed several changes to streamline the hiring process, including changes to help coordinate and streamline the verification of applicants' education and professional credentials. In addition, each of the eight VAMCs we visited reported that it had begun local efforts to better manage and coordinate the RN hiring process. To improve the ability of VAMCs to determine the RN staffing levels needed for inpatient units and to recruit and retain inpatient RNs, we recommend that the Secretary of Veterans Affairs direct the Under Secretary for Health to implement the following three recommendations: * develop a detailed action plan that includes a timetable for building, testing, and implementing a new nurse staffing system, * ensure that the proposed nurse staffing system provides RN staffing estimates that accurately account for both the actual inpatient acuity levels and current nursing tasks performed on inpatient units and take into account the level of ancillary and nursing support that is available on VAMC inpatient units, and: * assess the barriers to wider availability of alternate and flexible work schedules for RNs at VAMCs and explore ways to overcome these barriers. In commenting on a draft of this report, VA concurred with our findings and recommendations. With regard to our recommendations, VA said that it plans to develop a new nurse staffing system--one which recognizes patient acuity levels and tasks currently performed by inpatient nurses. VA also said that it plans to convene a task force to more fully explore opportunities to offer alternate and flexible work schedules to its nurses. Background: RNs, along with other nursing support staff, provide care to patients on inpatient units at VAMCs. VA RNs are responsible for assessing and providing care to patients, administering medications, documenting patients' medical conditions, analyzing test results, and operating medical equipment. To obtain an RN license, an individual must complete a nursing education program, meet state licensing requirements, and pass a nursing licensing examination.[Footnote 11] Several types of clinical and ancillary support staff assist RNs in caring for patients on inpatient units. Nursing support staff--such as licensed practical nurses (LPN) and nursing assistants (NA)--perform nursing duties such as recording patient vital signs and assisting with bathing, dressing, and personal hygiene. In addition, ancillary support staff perform housekeeping, patient transport, and food service duties. Clinical support staff--such as lab technicians--also assist RNs in their patient care duties, for example, by drawing and testing blood or performing electrocardiograms (EKG). Recognizing that developing and maintaining a strong cadre of RNs at VAMCs is vital to providing high quality of care to our nation's veterans, the Congress and VA have both made efforts to better ensure that RN staffing levels at VAMCs are appropriate and to enhance recruiting and retention of RNs. These efforts include: * To improve RN staffing, retention and job satisfaction, and patient outcomes, VA is actively encouraging its medical centers to take part in a nationwide program called the Magnet Recognition Program®. This program was developed by the American Nurses Credentialing Center (ANCC) to recognize health care organizations that provide nursing excellence and quality patient care.[Footnote 12] In order to attain Magnet™ status, hospitals must meet certain requirements, including requirements related to staffing practices and quality monitoring.[Footnote 13] Some research indicates that facilities that have attained Magnet™ status have better patient outcomes, significantly higher percentages of baccalaureate-prepared nurses, and higher nurse job satisfaction rates.[Footnote 14] As of 2008, three VAMCs have attained Magnet™ status, four VAMCs have completed the application process, and 22 VAMCs are in the process of applying for Magnet™ status. * With respect to recruiting and hiring, VA has taken several actions. For example, in 2007, VA launched the VA Nursing Academy, a program designed to develop a pool of RN candidates for employment in VAMCs. Similarly, VA's Learning Opportunities Residency program is designed to attract baccalaureate nursing students to work as RNs at VAMCs upon graduation. VA also has several other initiatives to enhance the educational preparation of its health care staff and scholarships for current employees pursuing degrees in nursing. These initiatives, which serve as recruitment and retention tools, include an education loan repayment program and scholarships for employees seeking health care careers.[Footnote 15] In August 2007, a VA Recruitment Process Redesign Workgroup made recommendations to redesign the recruiting and hiring of health care practitioners within VA, including RNs.[Footnote 16] The work group analyzed VA's hiring process and identified barriers and delays in hiring. The work group's findings and recommendations included a timeline for nurse hiring. As discussed later in this report, VA is in the process of implementing actions recommended by this work group. * To enhance recruiting and retention of RNs at VAMCs, the Congress passed legislation in 2004 authorizing two alternate work schedules for RNs employed by the VA.[Footnote 17] One of these alternate schedules allows RNs to work three 12-hour shifts that are considered a 40-hour work week for pay and benefits purposes. The other alternate work schedule allows RNs to work full-time for 9 months with 3 months off duty within a fiscal year and be paid 75 percent of the full-time work rate for each pay period of that fiscal year. In addition to these alternate work schedules, executive branch government agencies-- including the VA--are authorized by OPM to offer flexible work schedules. A flexible work schedule is an 80-hour biweekly basic work requirement that allows an employee to determine his or her own schedule--arrival and departure times--within the limits set by the agency.[Footnote 18] The standard work schedule for full-time VA employees is ten 8-hour work days within a 2-week period.[Footnote 19] VAMC Nursing Officials Cannot Rely on Information from VA's PCS to Determine RN Staffing Levels for VA Inpatient Units: VAMC nursing officials reported that although VA RNs are required to input patient data into VA's PCS, many said they cannot rely on the information generated by PCS because the PCS is outdated and inaccurate. Because of the shortcomings of VA's PCS, nurse managers use various other data to help set RN staffing levels for their inpatient units, such as historical staffing levels and benchmarking RN staffing levels to inpatient units in hospitals with similar characteristics. VA is proposing action to develop a new nurse staffing system but did not provide a detailed action plan and milestones for building and implementing such a system. VAMC Nursing Officials Report That Information from VA's PCS Is Outdated and Inaccurate: VAMC nursing officials we interviewed told us that VA's PCS does not generate reliable information that would allow them to better predict the RN staffing levels required for their inpatient units. These nursing officials cited two key limitations of this information--it is outdated and inaccurate.[Footnote 20] VA headquarters officials in the Office of Nursing Services (ONS) and VA's OIG also reported that PCS has significant limitations. In 2004, VA's OIG also recommended that VA develop a new standardized nurse staffing methodology capable of accurate staffing estimates and VA concurred with the OIG's recommendation. According to nursing officials we interviewed, VA's PCS was developed in the 1980's based on time and motion studies of RNs conducted over 20 years ago; as a result, the information the system produces does not account for all the tasks currently performed by RNs on inpatient units. For example, VA's PCS does not account for certain recent RN tasks--such as the administration of certain intravenous medications or monitoring of a patient's abnormal heart activity--that were once limited largely to intensive care units that cater to sicker patients but are now performed on other inpatient units. Similarly, VA's PCS generates estimates that do not reflect tasks associated with VA's computerized bar code medication administration (BCMA) system that was fully implemented in 2003, more than a decade after the development of VA's PCS.[Footnote 21] These RN tasks include tracking, monitoring, and reporting medication administration performed using the BCMA on an inpatient unit. VAMC nursing officials also told us that VA's PCS produces inaccurate data with respect to patient acuity levels, which in turn can generate erroneous HPPD estimates. Specifically, a key piece of data VA nurses enter into the VA PCS is the acuity level for each patient on an inpatient unit. To do this, RNs use one of five PCS categories, with category 1 representing patients requiring the lowest level of care and category 5 the highest level of care. Nursing officials we interviewed at VAMCs we visited and officials with VA's ONS reported that VA's PCS does not accurately capture the actual acuity level of patients on inpatient units: According to VAMC nursing officials we interviewed, nursing staff at VAMCs are required to classify patients by acuity level on a daily basis using VA's PCS. However, nursing officials reported that classifying patients by acuity level using the PCS is not a productive use of their time because the information output from the PCS is not useful for RN staffing purposes. In addition, officials with the VA OIG and ONS told us that the information contained in VA's Computerized Patient Record System (CPRS)[Footnote 22]--concerning a patient's illness, medical condition, and treatments--is not integrated with, or available within VA's PCS when nurses assess and assign patients to one of the five acuity levels. VA's ONS is proposing to convene an interdisciplinary team--consisting of headquarters and field staff--to develop a more effective RN staffing system for VA by 2012, according to VA's Chief of Nursing Services. The Chief told us that the new RN staffing system will include a database that reflects up-to-date nursing tasks as well as information from patients' computerized medical records, and that this database will also be used to evaluate the effectiveness of nursing care at VAMCs. The Chief also reported that as part of the new RN staffing system, VA may upgrade or replace its PCS. In developing a new or upgraded PCS, VA needs to ensure that all current nursing tasks and patient acuity are accurately captured. However, ONS did not provide a VA charter for the interdisciplinary team or a detailed action plan with specific timelines for the building, testing, and implementation of an updated system for staffing RNs on inpatient units in its VAMCs. Instead of Relying on Information from PCS, VA Medical Centers Use a Variety of Data Sources to Set RN Staffing Levels: Instead of relying on the information generated by VA's PCS, VAMCs use various other data to determine RN staffing for inpatient units.[Footnote 23] Our survey of nurse executives coupled with our visits to VAMCs provides insights into the types of data used and shows that information used for staffing RNs varies considerably among VAMCs. Results from the survey show that nurse managers typically consider a combination of data to estimate both the number of RNs and the RN skill levels their inpatient units require. The survey results also show that the types of data most commonly used by inpatient unit nurse managers across VAMCs are the average number of patients typically cared for on the unit, HPPDs, the acuity level of patients on the unit, the number of RN staff historically assigned to the unit, and the ratio of RNs to patients on the unit.[Footnote 24] Nurse managers at the eight VAMCs we visited told us how they use various data to help set RN staffing levels for their inpatient units. At four VAMCs we visited, nurse managers told us that they set RN staffing levels for their inpatient units by adhering to the historical staffing levels that had been established for the units. According to these nurse managers, they inherited their RN staffing levels when they assumed their position as manager of the unit. These nurse managers told us that the staffing levels for their inpatient units were established more than a decade ago. Nurse managers at another facility we visited told us that they consider data on the number of patients on the unit, HPPDs, nurse-to-patient ratios, and historic staffing levels to estimate the RN staffing needs on their units. Nurse managers using historical RN staffing levels to set current RN levels told us that this method does not adequately match RN staffing levels to the needs of inpatient units. Nurse managers at the four VAMCs that use such historical data said that historical RN staffing levels had not matched the acuity levels of their patients, which has increased over time. The other three VAMCs we visited have attained Magnet™ status and are required to set their RN staffing levels by benchmarking them against data on RN staffing levels found in non-VA facilities that have attained Magnet™ status. VA and non-VA Magnet™ facilities are grouped for benchmarking based on inpatient units with similar characteristics. Magnet™ facility RN-staffing data are available to facilities participating in the Magnet Recognition Program®. The nurse executive at one Magnet™ VAMC that benchmarks told us that it had not experienced RN staffing problems, and unit nurse managers at this VAMC expressed general satisfaction with RN staffing levels. VAMC Nursing Officials and RNs Identify Several Factors That Could Adversely Affect Retention of Inpatient RNs: VAMC nursing officials and inpatient RNs reported that two main factors adversely impact RN's job satisfaction and ultimately could impact VA's ability to retain RNs. First, according to these groups, some inpatient RNs are dissatisfied about spending too much time performing non- nursing duties, such as cleaning beds after a patient is discharged or answering unit telephones. Second, even though VAMCs were authorized in 2004 to offer RNs two alternate work schedules, few nurse executives reported offering these schedules; as a consequence, few RNs work these schedules. Both nursing officials and inpatient RNs working on inpatient units told us that the limited availability of flexible and alternate work schedules affects the ability of RNs to balance work and personal commitments. In addition to these two main factors, inpatient RNs cited other factors affecting retention, such as reliance on supplemental staffing strategies, for example RN overtime, and insufficient professional development opportunities. VAMC Nursing Officials Report That Performing Non-Nursing Tasks Could Negatively Impact RN Retention: VAMC nursing officials at five VAMCs we visited reported that RNs on inpatient units routinely perform non-nursing tasks, such as housekeeping tasks and transporting patients to other areas of the hospital for tests, because inpatient units often lack ancillary support staff or nursing support staff to perform these tasks. In 2001, we reported that job dissatisfaction because of inadequate support staff was reported to be a major contributor to retention problems in the nursing workforce.[Footnote 25] According to VAMC nurse executives we surveyed many of their VAMCs did not have access to ancillary support services around the clock. For example, as table 1 shows, the percentage of nurse executives at VAMCs who had support staff available around the clock ranged from 13 percent--for staff available to administer electrocardiograms--to 53 percent, for staff available to draw and test samples. Table 1: Percentage of VA Nurse Executives Reporting Ancillary Support Staff's Availability around the Clock: Ancillary support staff: Lab and blood drawing; Percent available: 53. Ancillary support staff: Housekeeping; Percent available: 50. Ancillary support staff: Maintenance; Percent available: 34. Ancillary support staff: Unit/ward clerk; Percent available: 31. Ancillary support staff: Transport/escort; Percent available: 19. Ancillary support staff: Electrocardiogram (EKG) technician; Percent available: 13. Source: GAO survey of VA nurse executives. [End of table] Nursing officials at VAMCs we visited indicated that RNs performing non- nursing tasks reduces RN job satisfaction and has caused some RNs to leave VA to accept jobs at other hospitals where RNs are required to perform fewer non-nursing tasks. Nursing officials also reported that RNs prefer to focus on providing nursing care to patients and that RNs performing non-nursing tasks could adversely affect the retention of RNs on inpatient units. Nursing officials from VAMCs and a representative of a state hospital association we interviewed cited three main factors--budgetary constraints, institutional practices, and retention and recruiting-- that contributed to insufficient ancillary and nursing support staff to assist RN on inpatient units. * Budgetary constraints can delay hiring ancillary and nursing support staff mainly through hiring freezes or lags.[Footnote 26] A VA official told us that hiring freezes and hiring lags are used for budgetary reasons or to manage personnel costs. According to nursing officials, a hiring freeze may be initiated by the regional network[Footnote 27] or imposed by a single VA medical center. During a hiring freeze, inpatient units typically require authority from a VAMC resource board[Footnote 28] to fill a vacant position. * Institutional practices at some VAMCs lead to some categories of ancillary and support staff's being unavailable during evening or weekend shifts, resulting in the need for RNs to perform additional tasks during these shifts. For example, housekeeping staff and laboratory staff who draw blood samples are not always available. In other cases, support staff do not perform all of the tasks associated with a certain function, resulting in the need for RNs to perform the tasks. For example, patient escort staff do not always assist in getting patients onto a stretcher or into a wheelchair for transport. * Recruiting and retaining ancillary and support staff can be difficult because a limited supply of support staff can lead to competition among local hospitals. According to a representative of a state hospital association we interviewed, there is a national shortage of allied health professionals--such as NAs, clinical laboratory technicians, radiology technicians, and physical therapists--in the hospital setting that can affect the workload of RNs. VAMC Nursing Officials and RNs Report That Limited Use of Alternate and Flexible Work Schedules Has the Potential to Adversely Impact Retention: VAMC nursing officials we interviewed and inpatient RNs who attended our focus groups at VAMCs reported that the limited availability and use of alternate and flexible work schedules at VAMCs limit the ability of RNs to balance their work and personal life needs and could adversely impact the retention of RNs. In 2008, we reported on the importance of work schedules that offer flexibility being available for older employees, who are nearing or are at the age when they may consider retirement as an incentive to remain working.[Footnote 29] While VA received legal authority in 2004 to offer alternate work schedules to RNs, these schedules are rarely used at VAMCs.[Footnote 30] Available 2007 data from VA show that less than half of one percent of the approximately 43,000 RNs employed by VA use alternate work schedules.[Footnote 31] There is low usage mainly because inpatient units at VAMCs do not usually offer such alternate work schedules. According to our survey of nurse executives, one alternate work schedule--36 hours per week--was reported by only 1 percent of surgical, mental health, medical, polytrauma, and intensive care units, while the second alternate work schedule--working full time for 9 months with 3 month off duty--was not offered at all. Several nursing officials we interviewed noted that not offering alternate work schedules can be a deterrent to retaining RNs. Half of all nurse executives reported that the lack of alternate and flexible schedules at their VAMC was one of the primary reasons for difficulty competing with local hospitals in recruiting and retaining RNs. VAMC nursing officials noted, however, that the ability to implement alternate work schedules at their VAMCs may be constrained by factors such as limited RN staffing. Flexible work schedules are more widely available than alternate work schedules at VAMCs. VAMCs offer several types of flexible work schedules--such as 10 and 12-hour schedules--and the availability of these flexible work schedules vary by the type of inpatient unit. Nurse executives we surveyed reported that the most frequently used flexible schedule was the 12-hour schedule, which was reported for 68 percent of medical intensive care and critical care units and 30 percent for nursing home units.[Footnote 32] Other flexible work schedules were used less frequently: for example, according to nurse executives we surveyed the use of the 10-hour schedule was reported by 13 percent of medical units and only 1 percent of spinal cord injury units.[Footnote 33] As was the case with alternate work schedules, several nursing officials we interviewed noted that the ability to implement flexible work schedules at their VAMCs was constrained by the number of RNs available to cover the various shifts. According to VAMC nursing officials, offering flexible work schedules is an important factor in recruiting and retaining RNs. Half of VA nurse executives we surveyed reported that one of the primary reasons for the difficulty in competing with local hospitals in retaining inpatient RNs was that flexible work schedules were not offered on some units at their medical center. We were told that many private hospitals use flexible work schedules as a way to improve nurse retention. A nursing official reported that a survey by the American Organization of Nurse Executives--a professional organization for nurse leaders and executives--found that after salary, the top benefit desired by nurses was flexible work schedules. One state hospital association representative we interviewed reported that 42 percent of the hospitals surveyed in their state offered flexible work schedules. VAMC Nursing Officials and Inpatient RNs Reported That Reliance on Supplemental Staffing and Insufficient Professional Development Opportunities Could Adversely Impact RN Retention: VAMC nursing officials and inpatient RNs cited other factors that could adversely impact job satisfaction, and ultimately, the retention of RNs at VAMCs, including reliance on supplemental staffing strategies and insufficient professional development opportunities.[Footnote 34] * A reliance on supplemental staffing strategies, such as RN overtime because of inadequate RN staffing levels on their unit, are factors that could adversely impact RN job satisfaction and ultimately retention.[Footnote 35] For example, when there is an unplanned absence, nurse managers use supplemental staffing strategies or operate the units short-staffed. Forty-eight percent of nurse executives reported that inpatient units worked short-staffed at some point, and 41 percent of nurse executives reported that mandatory RN overtime was used as a supplemental staffing strategy. Nurse managers reported that in some instances, they get an RN to work a part of the next shift or the entire next shift, or float staff as a result of a staff vacancy, staffing shortages, or an increase in the number of inpatients on the unit. Moreover, one nurse manager reported that "burnout" can stem from a reliance on supplemental staffing strategies. * According to RNs who attended our focus groups, insufficient professional development and training opportunities for inpatient RNs are RN-retention issues. For example, inpatient RNs noted that access to training and professional development activities for RNs can be limited. For example, RNs on the evening and night shifts sometimes find it difficult to participate in professional development activities and education programs.[Footnote 36] VAMC Nursing Officials Identified Limitations in VA's Hiring Process and VA-Imposed Hiring Freezes as Contributing Factors to RN Hiring Delays: VAMC nursing officials we surveyed and interviewed reported that delays resulting from limitations in VA's hiring process and hiring freezes and lags at VAMCs can often discourage prospective RN candidates from seeking or following through on applications for employment at VAMCs. Although VA has recently taken steps to address some of the factors that contribute to RN hiring delays, it is too early to determine the extent to which these steps have been effective in reducing hiring delays. Limitations in VA's Hiring Process Include Delays in Securing Necessary Approvals, Poor Coordination, and a Shortage of Experienced and Well- Trained HR Officials: VAMC nurse executives we surveyed and nursing officials we interviewed identified limitations in VA's hiring process. Nursing officials identified three areas of limitations--delays in securing necessary approvals from medical center resource boards to fill RN vacancies; poor coordination between nursing and HR officials involved in hiring; and a shortage of experienced and well-trained HR officials.[Footnote 37] Collectively, these factors result in significant delays in filling RN vacancies. We surveyed VAMC nurse executives to estimate the time it typically takes VAMCs to fill RN vacancies and found that 44 percent reported it took 45 to 80 days to fill inpatient RN vacancies at VAMCs in 2007 compared to the 24-to 45-day target timelines that VA set in 2007. One-third of nurse executives we surveyed reported that it took more than 80 days to fill RN vacancies at their VAMCs. In contrast, local hospitals usually hired RNs in less than 21 days, according to nursing officials we interviewed.[Footnote 38] Nursing officials during our site visits reported that these delays contribute to VAMCs' losing applicants to local hospitals as well as reliance on supplemental staffing strategies to maintain RN staffing levels on inpatient units. Delays in gaining approval to fill RN vacancies: One factor VAMC nursing officials identified as contributing to hiring delays is the period of time during which officials wait to get approval from a VA medical center resource board--an internal board that controls the medical center's budget and the number of authorized staff positions-- to fill an RN vacancy.[Footnote 39] Poor coordination between nursing and HR officials: VAMC nursing officials identified poor coordination between nursing and HR officials as another factor that contributed to delays in filling RN vacancies. HR officials are involved in handling application paperwork, interviewing RN applicants, scheduling screening activities such as physical examinations and background checks, and verifying employment references. VA's Recruitment Process Redesign Workgroup recently issued several recommendations aimed at improving coordination in filling RN vacancies. Poor coordination can occur when nursing officials must wait for information from HR officials before a job offer can be made to an applicant. For example, VAMC nursing officials we interviewed stated that they may have to wait a few weeks for HR officials to determine an appropriate salary estimate based on an applicant's educational qualifications and experience, a process that must take place before a job offer can be made to an RN applicant.[Footnote 40] In our survey, about 65 percent of nurse executives cited the inability to provide a salary estimate promptly to an applicant as one of the primary reasons they lost RN applicants to competing, non-VA hospitals. Poor coordination can also occur during the pre-employment process. According to nursing officials, RN applicants may make multiple visits to the medical center for pre-employment physicals and verification of state licenses because these activities have not been coordinated into one visit for the RN applicant. Shortage of experienced HR officials: VA headquarters and VAMC nursing officials identified the shortage of experienced and well-trained HR officials who process RN employment applications and hiring paperwork as a factor that contributes to RN hiring delays. VAMC nursing officials we interviewed reported that VAMCs have suffered a "brain drain" of experienced HR officials through retirement or attrition. VA noted in its recent workforce succession strategic planning report that it faces a challenge caused by a "lack of trained HR staff and expertise in the area of human resources." VA's ability to address delays in filling RN vacancies depends, in part, on its ability to retain experienced HR officials and to recruit and train new ones. According to VA, new HR recruits must acquire a good grasp of the breadth and complexity of HR knowledge and skills required by the federal government and VA. For example, VA noted that an effective HR official must possess specific knowledge of the complex laws, rules, and regulations for more than 300 VA occupations. VAMC Nursing Officials Report Periodic Hiring Freezes Imposed by VAMCs Delay Initiation of Hiring Process to Fill RN Vacancies: VAMC nurse executives we surveyed reported that hiring freezes and lags delayed the initiation of the hiring process to fill RN vacancies.[Footnote 41] Forty-four percent of VA's nurse executives we surveyed reported that they experienced a medical center hiring freeze between 2002 and 2007. On the average, nurse executives we surveyed reported experiencing two hiring freezes during this period, and 45 percent of nurse executives reported that the hiring freezes they experienced lasted on average from 7 to 12 months. Furthermore, 67 percent of nurse executives we surveyed reported that they experienced a hiring lag--that is, a temporary delay in hiring or a recurring process intended to control expenditures by limiting hiring to a certain number of new employees in a given pay period. About one-third of the nurse executives we surveyed indicated that these hiring freezes contributed to delays in the hiring process, and nearly half of nurse executives reported that a lag in hiring also contributed to delays that may dissuade potential applicants. Some nursing officials we interviewed told us that once the word has spread in the local community that the VAMC has imposed a hiring freeze, the medical center has difficulty recovering from the effects of the hiring freeze. In some cases, nursing officials reported that it took up to 2 years for RNs to reapply for VA employment because some applicants were not aware that VA's hiring freeze had ended. Efforts Are Under Way to Address Some of the Factors Contributing to RN Hiring Delays: VA has a number of efforts under way at both the national level and at individual VAMCs to reduce shortages in its healthcare workforce, including RNs. On a nationwide basis, VA has authorized its medical centers to implement several changes recommended by the Recruitment Process Redesign Workgroup that studied recruitment and hiring at VA.[Footnote 42] These recommendations may address some of the factors that contribute to delays in filling RN vacancies. According to VA officials, these changes are designed to increase flexibility and efficiency without weakening the process of screening candidates' professional credentials. Collectively, VA's recent changes consist of ways to (1) complete applicant interviews and physical examinations on the same day, (2) make a job offer to an RN applicant within 30 days, (3) allow use of electronic education transcripts in lieu of paper transcripts sent through the mail, and (4) create additional HR positions to help meet VA's future needs as its experienced HR officials retire. In addition to implementing VA's nationwide efforts, nursing officials at eight VAMCs we visited cited a number of steps that have been taken at individual VAMCs to increase efficiency and reduce hiring delays. These steps include: * Improved communication and coordination between HR and nursing officials during the hiring process in various ways. For example, seven of the eight VAMCs we visited reported that they improved coordination and communication between nursing and HR officials involved in hiring RNs by better tracking an applicant's paperwork and coordinating other pre-employment activities, including scheduling interviews and physical examinations on the same day when possible. In addition, two of the eight VAMCs have increased their interactions through regular meetings. Nursing officials at two VAMCs we visited told us that they have implemented a program called "On the Floor in 24," an effort which allows the VAMC to bring an RN on board within 24 days by better coordinating and expediting steps in the hiring process. * Hiring RNs under temporary appointments until screening activities such as physical examinations, drug tests, and background checks are completed. Nursing officials at five VAMCs we visited told us that they have hired some RNs on a temporary appointment status until screening activities are completed. Moreover, another VAMC we visited implemented a program called "On-Demand Hiring," established by nursing and HR officials, which involves hiring an RN with the aid of a nurse recruiter. VAMC nursing officials reported that the nurse recruiter may then provide the applicant a salary range; afterwards, HR officials become involved by making a job offer to the RN applicant and proceed with screening activities such as arranging to have fingerprints taken for a background check, scheduling a physical examination, and drug tests. Seven VAMCs we visited implemented a computerized process to expedite the verification of RNs' professional credentials. * Hiring a nurse recruiter as a contact point between RN applicants and HR officials to handle application paperwork. Nursing officials at four VAMCs we visited reported that they have hired a nurse recruiter who will act as the focal point for coordinating with HR in posting RN vacancies and various steps in the RN hiring process. * Implementing new procedures to make the hiring process more efficient. These procedures include delegating authority to sign nursing personnel actions, creating an application tracking database, and delegating authority from HR to nursing officials to give provisional salary quotes to applicants. Nursing officials at one VAMC we visited told us that before these recent changes to the hiring process, they usually had to wait for HR to provide an estimated starting salary to interested applicants who may have other job offers to consider. Table 2 summarizes the actions taken by VAMCs we visited to reduce delays in filling RN vacancies. Table 2: Summary of Actions Taken by Eight VAMCs We Visited to Reduce Delays in Filling RN Vacancies: Actions taken: Used VA's expedited RN credentialing process; VAMC A: X; VAMC B: X; VAMC C: --; VAMC D: X; VAMC E: X; VAMC F: X; VAMC G: X; VAMC H: X; Number of VAMCs that implemented action: 7. Actions taken: Improved communication between VAMC officials involved in hiring RNs; VAMC A: X; VAMC B: X; VAMC C: X; VAMC D: X; VAMC E: X; VAMC F: --; VAMC G: X; VAMC H: X; Number of VAMCs that implemented action: 7. Actions taken: Used temporary appointments; VAMC A: X; VAMC B: X; VAMC C: X; VAMC D: --; VAMC E: X; VAMC F: X; VAMC G: --; VAMC H: --; Number of VAMCs that implemented action: 5. Actions taken: Used "On the Floor in 24" (days); VAMC A: X; VAMC B: X; VAMC C: --; VAMC D: --; VAMC E: --; VAMC F: --; VAMC G: --; VAMC H: -- ; Number of VAMCs that implemented action: 2. Actions taken: Hired a nurse recruiter; VAMC A: --; VAMC B: [A]; VAMC C: X; VAMC D: X; VAMC E: X; VAMC F: X; VAMC G: --; VAMC H: --; Number of VAMCs that implemented action: 4. Actions taken: Used "On-Demand" hiring; VAMC A: --; VAMC B: --; VAMC C: --; VAMC D: --; VAMC E: X; VAMC F: --; VAMC G: --; VAMC H: --; Number of VAMCs that implemented action: 1. Actions taken: Other; VAMC A: --; VAMC B: --; VAMC C: X[B]; VAMC D: X[B]; VAMC E: --; VAMC F: --; VAMC G: X[B]; VAMC H: --; Number of VAMCs that implemented action: 3. Source: GAO Analysis of VA data. Note: --means VAMC officials did not note this action was taken. [A] VAMC already had a nurse recruiter. [B] One VAMC hired a new HR chief and plans to initiate several other practices. Another VAMC increased classes of student nurses to have a larger pool to recruit for permanent jobs when these student nurses receive their RN licenses. The third VAMC hired two nurse recruiters to cover other campuses and outpatient clinics. [End of table] While VA's national and local efforts to reduce hiring delays are commendable, most are relatively recent, and it is too early to determine the extent to which these efforts will reduce RN hiring delays or whether they are sustainable. Conclusions: To better ensure that RN staffing levels on inpatient units in its VAMCs are adequate to provide quality care to its patients, VA must be vigilant on two fronts. First, it is important that the department expeditiously proceed with planning and implementing a nurse staffing system that accurately reflects patient needs and RN workload requirements. The inaccuracy of VA's current PCS limits its usefulness in helping to establish adequate RN staffing levels and reveals a larger problem--VA does not have a viable system to accurately determine the RN staffing needs for inpatient units. VA recognizes the need to develop a new standardized nurse staffing system capable of accurate staffing estimates. Without an accurate nurse staffing system, VAMCs and nursing officials do not have good information as a basis for making sound judgments about the RN staffing needs of inpatient units and often must use supplemental RN staffing strategies to match RN staffing levels with patient care needs. Excessive use of supplemental RN staffing strategies can in turn adversely impact RN morale and job satisfaction and may lead to RN retention problems. Second, as the number of VA RNs who may consider retirement increases and the nation continues to face an RN shortage, it is important that VA maximize its ability to hire new RNs, as well as to retain RNs that it currently employs. To help address RN retention at VAMCs, VA can use OPM-authorized flexible work schedules and congressionally authorized alternate work schedules; however, VAMC nursing officials reported limited availability and use of alternate and flexible schedules. VA's ability to maintain its RN workforce could be enhanced if it can expeditiously hire qualified applicants and offer more flexible and alternate work schedules for RNs. Further complicating VAMC RN staffing problems is that RNs often perform non-nursing tasks that lead to job dissatisfaction. Where RNs must perform non-nursing tasks because ancillary and nursing support staff are not available, it is important that nurse managers have the ability to adjust unit RN staffing levels so that nurses have adequate time to perform these duties. Developing a staffing approach that can accurately determine adequate RN staffing levels may also support VA's ability to help RNs balance their work and personal commitments through the offering of alternate work schedules to RNs. Recommendations for Executive Action: To improve the ability of VAMCs to determine RN staffing levels needed for inpatient units and to recruit and retain inpatient RNs, we recommend that the Secretary of Veterans Affairs direct the Under Secretary for Health to implement the following three recommendations: * develop a detailed action plan that includes a timetable for building, testing, and implementing the new nurse staffing system; * ensure that the new nurse staffing system provides RN staffing estimates that accurately account for both the actual inpatient acuity levels and current nursing tasks performed on inpatient units and adequately take into account the level of ancillary and nursing support that is available on VAMC inpatient units; and: * assess the barriers to wider availability of alternate and flexible work schedules for RNs at VAMCs and explore ways to overcome these barriers. Agency Comments and Our Evaluation: In commenting on a draft of this report, VA concurred with our findings and recommendations and provided a description of actions that it plans to take to address our recommendations. Regarding our first recommendation--that VA develop a detailed action plan for building, testing, and implementing a new nurse staffing system--VA stated that it has long recognized the need for an automated and data-driven nurse staffing methodology and noted some of the challenges in developing and implementing such a system. VA reported that the pilot implementation of the proposed nurse staffing system is expected to be completed next year and plans to implement the new system on all inpatient units by 2012. VA provided a copy of a three- phase plan for the creation, testing, and implementation of a new staffing methodology for nursing personnel on inpatient units. While the first phase of VA's plan appears to be in-process and the second phase addresses staffing in areas other than inpatient units, the third phase of VA's plan--which includes the development of an automated scheduler application and a patient acuity application--is pending approval by VA's Office of Information. This third phase in VA's plan is critical to developing an automated and data-driven nurse staffing methodology, and we would encourage VA to approve this phase of the plan as soon as possible. To address our second recommendation--that VA ensure its new nurse staffing system accurately account for factors including available ancillary and nursing support--VA stated that its new nurse staffing system will include indicators for ancillary and nursing support and that nurse staffing projections will be determined based on nurse responsibilities. Concerning our third recommendation--that VA assess the barriers to wider availability of alternate and flexible work schedules and explore ways to overcome these barriers--VA stated that it plans to convene a task force to fully assess barriers to the effective use of alternate and flexible work schedules for RNs and to identify potential solutions for overcoming these barriers. The task force will present its findings by June 2009. VA's written comments are reprinted in appendix IV. We are sending copies of this report to the Secretary of Veterans Affairs, appropriate congressional committees, and other interested parties. We will also provide copies to others upon request. In addition, the report is available at no charge on the GAO Web site at [hyperlink, http://www.gao.gov]. If you or your staff members have any questions about this report, please contact me at (202) 512-7114 or at williamsonr@gao.gov. Contact points for our Offices of Congressional Relations and Public Affairs can be found on the last page of this report. GAO staff members who made major contributions to this report are listed in appendix V. Signed by: Randall B. Williamson: Director, Health Care: [End of section] Appendix I: Scope and Methodology: We examined the Department of Veterans Affairs' (VA) inpatient registered nurses (RN) staffing practices at VA medical centers (VAMC) and the challenges VA faces in hiring and retaining RNs. Specifically, we identified (1) how useful the information generated by VA's patient classification system (PCS) is for determining RN staffing levels on VA inpatient units, (2) key factors that nursing officials and RNs identify that adversely affect RN retention on inpatient units, and (3) factors that nursing officials identify as contributing to delays in hiring RNs to fill vacant positions. To examine how VAMCs determine the RN staffing levels needed for inpatient units, we conducted a Web-based survey of all VA nurse executives at VAMCs. A nurse executive is a member of the executive leadership team at a VAMC and is responsible for all nursing care delivered at the VAMC. The survey was sent to 140 VA nurse executives and obtained a 63 percent response rate, which allows us to generalize the results to all VA nurse executives at VAMCs. To field the survey, we contacted a VA headquarters official in the Office of Nursing Services (ONS) to obtain a list of VAMC nurse executives, and the official provided email addresses for the nurse executives. See appendix II for the results of our VA nurse executive survey. To gain further information on RN staffing, we interviewed VAMC nurse executives and VAMC inpatient unit nurse managers[Footnote 43] responsible for determining inpatient RN staffing levels at eight VAMCs we visited located in Denver, Colorado; Houston, Texas; Minneapolis, Minnesota; New York, New York; Portland, Oregon; Seattle, Washington; Tampa, Florida; and Togus, Maine.[Footnote 44] We selected these VAMCs because of their geographic variation and to capture various types of inpatient units including medical intensive care, critical care, surgical intensive care, surgery, medicine, behavioral health, nursing home, and spinal cord injury. The findings from these eight VAMCs we visited cannot be generalized to all VAMCs. To assess VA's PCS and the information reported by nursing officials we reviewed the literature to identify relevant best practices in nurse staffing and the design of systems used to classify patients and interviewed representatives from state hospital associations in states where we conducted visits to VAMCs about the use of staffing methodologies and supplemental staffing strategies. To identify the factors that nursing officials and RNs identify as adversely affecting RN retention, we obtained April 2008 VA data on the number of VA RNs who use alternate work schedules. In addition, we interviewed nurse managers responsible for supervising RNs on inpatient units at the eight VAMCs we visited and conducted focus groups with inpatient RNs to get their perspectives on retention issues. The inpatient RNs in our focus groups typically deliver care at the patient bedside on inpatient units. The 219 RNs who participated in our focus groups were from three shifts (day, evening, and night) at the eight VAMCs we visited. Attendees at our focus groups included RNs of different ages, nurse experience levels, and lengths of tenure at VA. (See table 3 for a demographic profile of VA RNs who attended our focus groups.) During each focus group session, we provided RNs an opportunity to offer their opinions on a variety of issues related to their experience working at VA. For each focus group we utilized a series of structured questions to gain RNs' opinions on nurse staffing, recruitment, and retention issues. A summary of the responses from our focus groups are provided in appendix III. We also interviewed representatives from state hospital associations in states where we conducted visits to VAMCs about the local retention challenges affecting RNs. Table 3: Demographic Profile of RN Focus Group Attendees at Eight VAMCs We Visited: Demographics: Total number of RN participants; VAMC A: 22; VAMC B: 21; VAMC C: 22; VAMC D: 35; VAMC E: 17; VAMC F: 38; VAMC G: 40; VAMC H: 24. Demographics: Mode of age range (years); VAMC A: 51-60; VAMC B: 51-60; VAMC C: 41-50; VAMC D: 41-50; VAMC E: 41-50; VAMC F: 51-60; VAMC G: 51- 60; VAMC H: 51-60. Demographics: Median number of years worked at current VAMC; VAMC A: 3; VAMC B: 7; VAMC C: 6; VAMC D: 2; VAMC E: 3; VAMC F: 5; VAMC G: 5; VAMC H: 14. Demographics: Number of years worked at VA (range); VAMC A: 1-30; VAMC B: 0-34; VAMC C: 1-25; VAMC D: 0-35; VAMC E: 0-20; VAMC F: 0-26; VAMC G: 0-32; VAMC H: 0-31. Demographics: Median number of years worked at VA; VAMC A: 3; VAMC B: 8; VAMC C: 6; VAMC D: 5; VAMC E: 7; VAMC F: 5; VAMC G: 5; VAMC H: 15. Demographics: Median number of years worked as an RN; VAMC A: 18; VAMC B: 25; VAMC C: 8; VAMC D: 15; VAMC E: 5; VAMC F: 14; VAMC G: 22; VAMC H: 17. Demographics: Years worked as an RN (range); VAMC A: 1-43; VAMC B: 2- 42; VAMC C: 2-37; VAMC D: 1-42; VAMC E: 1-30; VAMC F: 0-44; VAMC G: 1- 43; VAMC H: 1-46. Demographics: Work shifts; VAMC A: D, E, N; VAMC B: D, E, N; VAMC C: D, E, N; VAMC D: D, E, N; VAMC E: D, N; VAMC F: D, E, N; VAMC G: D, E, N; VAMC H: D, E, N. Source: GAO focus groups of VA RNs. Legend: D = day: E = evening: N = night: [End of table] To identify the factors that contribute to delays in hiring RNs to fill vacant positions, we used results from our Web-based survey of nurse executives and interviewed VA headquarters officials, human resources (HR) officials, and nurse managers who recruit RNs at the eight VAMCs we visited. In addition, we reviewed VA policies, guidance, and reports related to RN staffing, retention and hiring issues and obtained, from VA headquarters officials, work schedule data on VA RNs who use alternate work schedules contained in VA's Personnel Accounting Integrated Data (PAID) System which houses VA's payroll and human resources information. We also interviewed representatives from state hospital associations in states where we visited VAMCs about RN recruitment challenges. We assessed the reliability of the data obtained from our Web-based survey of VA nurse executives and from VA headquarters officials. We performed a systematic review of the completed surveys by checking each survey for problems such as key questions left unanswered, patterns of skipped questions, unclear written responses, and out-of-scope entries. The information presented in our focus group summaries accurately capture the opinions provided by inpatient RNs who attended the focus groups at the eight VAMCs we visited. However these opinions cannot be generalized to all inpatient RNs at the eight VAMCs we visited, or to all inpatient RNs at VAMCs. We contacted VA headquarters officials responsible for VA's PAID system to gain an understanding of the completeness and accuracy of the data and whether quality checks were performed on these data. Based on this assessment we determined that these data were adequate for our purposes. We conducted this performance audit from May 2006 through September 2008 in accordance with generally accepted government auditing standards. Those standards require that we plan and perform the audit to obtain sufficient, appropriate evidence to provide a reasonable basis for our findings and conclusions based on our audit objectives. We believe that the evidence obtained provides a reasonable basis for our findings and conclusions based on our audit objectives. Selected results from our survey and a summary of responses from our focus groups are provided in appendix II and appendix III. [End of section] Appendix II: Analysis of GAO Survey of VA Medical Center Nurse Executives: To obtain the views of VA nurse executives on various staffing, recruitment, and retention issues, we conducted a Web-based survey of VA nurse executives employed at VAMCs. The survey contained questions on topics such as nurse executives' views on the use of supplemental staffing strategies at the VA medical center where the executives work, RN vacancies, recruitment and retention challenges, and the use of hiring freezes or lags in hiring. Some of these questions are listed below. Not all column totals add to 100 percent because of rounding, multiple answers to some questions that ask respondents to check all that apply, or no response checked by VA nurse executives for some questions. Q1: As of March 31, 2007, which of the following services or departments at this facility employed support staff for all shifts (day, evening, and night)? 1. Lab and blood drawing; Checked (percentage): 53; Number of respondents: 47. 2. Housekeeping; Checked (percentage): 50; Number of respondents: 44. 3. Maintenance; Checked (percentage): 34; Number of respondents: 30. 4. Unit/ward clerk; Checked (percentage): 31; Number of respondents: 27. 5. Transport/escort; Checked (percentage): 19; Number of respondents: 17. 6. Electrocardiogram (EKG) tests; Checked (percentage): 13; Number of respondents: 11. Source: GAO. [End of table] Q2: What were the effects on inpatient units of having bedside non- management RN vacancies? 1. RN overtime increased; Checked (percentage): 83; Number of respondents: 72. 2. RN floated to the unit with vacancies; Checked (percentage): 78; Number of respondents: 68. 3. Staff on units worked with fewer staff (short staffed); Checked (percentage): 49; Number of respondents: 43. 4. Number of patient beds were capped; Checked (percentage): 48; Number of respondents: 42. 5. Used contract/agency RNs; Checked (percentage): 45; Number of respondents: 39. 6. Used fee-basis RNs; Checked (percentage): 45; Number of respondents: 39. 7. Patients were diverted to non-VA facilities; Checked (percentage): 39; Number of respondents: 34. 8. RN turnover on the unit; Checked (percentage): 39; Number of respondents: 34. 9. Increased use of leave by RNs; Checked (percentage): 33; Number of respondents: 29. 10. Patients were diverted to other inpatient units; Checked (percentage): 32; Number of respondents: 28. 11. Patients were diverted to other VA facilities; Checked (percentage): 28; Number of respondents: 24. 12. Patient admissions were delayed; Checked (percentage): 26; Number of respondents: 23. 13. Increased patient incidents; Checked (percentage): 15; Number of respondents: 13. 14. Other effects; Checked (percentage): 8; Number of respondents: 7. 15. No effects; Checked (percentage): 5; Number of respondents: 4. Source: GAO. [End of table] Q3: What strategies did this facility use to supplement bedside non- management RN staffing? 1. Use of voluntary overtime; Checked (percentage): 92; Number of respondents: 80. 2. Float RNs from one unit to another; Checked (percentage): 85; Number of respondents: 74. 3. Employ student technician; Checked (percentage): 53; Number of respondents: 46. 4. Employ part-time RNs; Checked (percentage): 52; Number of respondents: 45. 5. Use of fee-basis RNs; Checked (percentage): 49; Number of respondents: 43. 6. Use of intermittent RNs; Checked (percentage): 49; Number of respondents: 43. 7. Staff on units worked with fewer staff (short staffed); Checked (percentage): 48; Number of respondents: 42. 8. Use of contract RNs; Checked (percentage): 46; Number of respondents: 40. 9. Use of mandatory overtime; Checked (percentage): 41; Number of respondents: 36. 10. Unit managers routinely took on patient assignments; Checked (percentage): 38; Number of respondents: 33. 11. Use of retired annuitants; Checked (percentage): 25; Number of respondents: 22. 12. Use of a formal staffed float pool; Checked (percentage): 15; Number of respondents: 13. 13. Other strategies; Checked (percentage): 10; Number of respondents: 9. 14. Did not employ any supplemental staffing strategies; Checked (percentage): 3; Number of respondents: 3. Source: GAO. [End of table] Q4: Based on your experience recruiting for bedside non-management RNs, what has been the average length of time it has taken to fill a position from the time you were authorized to recruit and fill the vacancy by an approved form SF 52 to the time an RN comes on board? 1. 6 weeks or less; Checked (percentage): 2.3; Number of respondents: 2. 2. 7 to 8 weeks; Checked (percentage): 13.8; Number of respondents: 12. 3. 9 to 16 weeks; Checked (percentage): 43.7; Number of respondents: 38. 4. More than 16 weeks; Checked (percentage): 33.3; Number of respondents: 29. 5. Other; Checked (percentage): 3.5; Number of respondents: 3. 6. Not checked; Checked (percentage): 3.5; Number of respondents: 3. Source: GAO. [End of table] Q5: What steps did this facility take in the last 5 years to simplify or shorten the hiring process for new RNs? 1. Improved communication to keep applicant informed about steps and paperwork required in hiring process; Checked (percentage): 72; Number of respondents: 47. 2. Brought in RNs under temporary appointments; Checked (percentage): 66; Number of respondents: 43. 3. Used VA's expedited VetPro credentialing process; Checked (percentage): 42; Number of respondents: 27. 4. Hired a nurse recruiter; Checked (percentage): 42; Number of respondents: 27. 5. Initiated boarding process after RN was hired; Checked (percentage): 35; Number of respondents: 23. 6. Instituted new procedures; Checked (percentage): 23; Number of respondents: 15. 7. Other; Checked (percentage): 22; Number of respondents: 14. 8. Hired additional staff to assist nurse recruiter; Checked (percentage): 17; Number of respondents: 11. Source: GAO. [End of table] Q6: Why did this facility hire bedside non-management RNs on temporary appointments? 1. To expedite the hiring process; Checked (percentage): 56; Number of respondents: 49. 2. To board RN at later date when paperwork is complete; Checked (percentage): 44; Number of respondents: 38. 3. To fill staffing need because hiring process takes too long; Checked (percentage): 39; Number of respondents: 34. 4. To evaluate RN to determine if good fit; Checked (percentage): 18; Number of respondents: 16. 5. Other; Checked (percentage): 3; Number of respondents: 3. Source: GAO. Note: "Boarding" is the process by which a VAMC determines the appropriate starting nurse grade level for a RN. [End of table] Q7: Did this facility experience a Veterans Integrated Service Network (VISN)-imposed or facility-imposed hiring freeze that affected its ability to hire bedside non-management RNs in the last 5 years? (During a hiring freeze units typically require authority from a resource board or other similar entity within the facility or VISN to fill a vacant RN position.) 1. VISN imposed hiring freeze; Checked (percentage): 9; Number of respondents: 8. 2. Medical center imposed hiring freeze; Checked (percentage): 15; Number of respondents: 13. 3. Both VISN and medical center imposed hiring freeze; Checked (percentage): 20; Number of respondents: 17. 4. No hiring freeze; Checked (percentage): 54; Number of respondents: 47. 5. Not checked; Checked (percentage): 2; Number of respondents: 2. Source: GAO. [End of table] Q8: What was the average length of the VISN or facility imposed hiring freezes that affected your ability to hire bedside non-management RNs? 1. 1 month or less; Checked (percentage): 5; Number of respondents: 2. 2. 2 to 3 months; Checked (percentage): 20; Number of respondents: 8. 3. 4 to 6 months; Checked (percentage): 18; Number of respondents: 7. 4. 7 to 12 months; Checked (percentage): 45; Number of respondents: 18. 5. More than 12 months; Checked (percentage): 8; Number of respondents: 3. 6. Not checked; Checked (percentage): 5; Number of respondents: 2. Source: GAO. [End of table] Q9: Did this facility experience a lag in hiring that affected its ability to hire bedside non-management RNs in the last 5 years? (A lag in hiring is a temporary delay in hiring RNs or can be a recurring process that delays hiring, i.e., only allowing a certain number of RNs to be brought on board per pay period or requiring vacant positions to be approved by an entity such as a resource board.) 1. Yes; Checked (percentage): 67; Number of respondents: 58. 2. No; Checked (percentage): 30; Number of respondents: 26. 3. Not checked; Checked (percentage): 3; Number of respondents: 3. Source: GAO. [End of table] Q10: What type of restrictions imposed during a hiring freeze or lag affected your ability to fill bedside non-management RN positions? 1. Recruitment for vacant RN positions needed approval by the equivalent of a resource board; VISN freeze (percentage): 26 (16); Medical center freeze (percentage): 51(31); Hiring lag (percentage): 62 (38). 2. Recruitment for each RN position needed approval by the equivalent of a resource board; VISN freeze (percentage): 31 (19); Medical center freeze (percentage): 49 (30); Hiring lag (percentage): 61(37). 3. Hiring for vacant RN positions deferred for a period of time; VISN freeze (percentage): 25 (15); Medical center freeze (percentage): 38 (23); Hiring lag (percentage): 46 (28). 4. Overtime for RNs increased; VISN freeze (percentage): 21 (13); Medical center freeze (percentage): 38 (23); Hiring lag (percentage): 59 (36). 5. Limits placed on recruitment for a number of RN positions; VISN freeze (percentage): 18 (11); Medical center freeze (percentage): 33 (20); Hiring lag (percentage): 36 (22). 6. Limited number of RN positions filled; VISN freeze (percentage): 15 (9); Medical center freeze (percentage): 31(19); Hiring lag (percentage): 38 (23). 7. Did not recruit for certain RN positions on specific units; VISN freeze (percentage): 15 (9); Medical center freeze (percentage): 28 (17); Hiring lag (percentage): 33 (20). 8. Limits were placed on number of RNs that could be hired in a pay period; VISN freeze (percentage): 11 (7); Medical center freeze (percentage): 23 (14); Hiring lag (percentage): 31(19). 9. Use of contract/agency RNs increased; VISN freeze (percentage): 10 (6); Medical center freeze (percentage): 15 (9); Hiring lag (percentage): 34 (21). 10. RNs hired under temporary appointments; VISN freeze (percentage): 5 (3); Medical center freeze (percentage): 13 (8); Hiring lag (percentage): 13 (8). 11. No vacant RN positions were filled; VISN freeze (percentage): 8 (5); Medical center freeze (percentage): 7 (4); Hiring lag (percentage): 7 (4). 12. Other; VISN freeze (percentage): 2 (1); Medical center freeze (percentage): 3 (2); Hiring lag (percentage): 8 (5). Source: GAO. Note: The number of respondents is indicated in parentheses. [End of table] Q11: How difficult has it been to compete with local health care establishments in recruiting bedside non-management RNs? 1. Very difficult; Checked (percentage): 24; Number of respondents: 21. 2. Generally difficult; Checked (percentage): 25; Number of respondents: 22. 3. Moderately difficult; Checked (percentage): 24; Number of respondents: 21. 4. A little difficult; Checked (percentage): 10; Number of respondents: 9. 5. Not at all difficult; Checked (percentage): 14; Number of respondents: 12. 6. Not checked; Checked (percentage): 2; Number of respondents: 2. Source: GAO. [End of table] Q12: What are the primary reasons for the difficulty in competing with local health care establishments in recruiting bedside non-management RNs? 1. Hiring process was lengthy; New Graduates (percentage): 76 (57); Experienced RNs (percentage): 89 (67). 2. RNs were required to rotate shifts; New Graduates (percentage): 65 (49); Experienced RNs (percentage): 68 (51). 3. Unable to make position offers promptly; New Graduates (percentage): 65 (49); Experienced RNs (percentage): 71 (53). 4. Shortage of RNs in the local market area; New Graduates (percentage): 48 (36); Experienced RNs (percentage): 67 (50). 5. Salary level low compared to locality pay area; New Graduates (percentage): 48 (36); Experienced RNs (percentage): 63 (47). 6. Alternate or flexible work schedule was not offered on the unit; New Graduates (percentage): 51 (38); Experienced RNs (percentage): 57 (43). 7. Tuition reimbursement not available at time of employment; New Graduates (percentage): 36 (27); Experienced RNs (percentage): 32 (24). 8. Recruitment incentives were not sufficient; New Graduates (percentage): 29 (22); Experienced RNs (percentage): 36 (27). 9. Difficult to recover from hiring freeze; New Graduates (percentage): 28 (21); Experienced RNs (percentage): 35 (26). 10. Reimbursement for continuing education not sufficient; New Graduates (percentage): 25 (19); Experienced RNs (percentage): 28 (21). 11. Determining salary was delayed by professional standards board; New Graduates (percentage): 21 (16); Experienced RNs (percentage): 24 (18). 12. Recruitment incentives were not available; New Graduates (percentage): 27 (20); Experienced RNs (percentage): 25 (19). 13. Tuition reimbursement not sufficient; New Graduates (percentage): 19 (14); Experienced RNs (percentage): 17 (13). 14. Reimbursement for continuing education not available; New Graduates (percentage): 16 (12); Experienced RNs (percentage): 17 (13). 15. VA benefit package was not as attractive as local competitors; New Graduates (percentage): 13 (10); Experienced RNs (percentage): 13 (10). 16. Other reasons; New Graduates (percentage): 13 (10); Experienced RNs (percentage): 15 (11). Source: GAO. Note: The number of respondents is indicated in parentheses. [End of table] Q13: How difficult has it been to compete with local health care establishments in retaining bedside non-management RNs? 1. Very difficult; Checked (percentage): 14; Number of respondents: 12. 2. Generally difficult; Checked (percentage): 9; Number of respondents: 8. 3. Moderately difficult; Checked (percentage): 30; Number of respondents: 26. 4. A little difficult; Checked (percentage): 32; Number of respondents: 28. 5. Not at all difficult; Checked (percentage): 13; Number of respondents: 11. 6. Not checked; Checked (percentage): 2; Number of respondents: 2. Source: GAO. [End of table] Q14: What are the primary reasons for the difficulty in competing with local health care establishments in retaining bedside non-management RNs? 1. RNs were required to rotate shifts; Checked (percentage): 60; Number of respondents: 52. 2. Alternate or flexible work schedule was not available; Checked (percentage): 49; Number of respondents: 43. 3. Salary level for experienced RN was low compared to local area; Checked (percentage): 43; Number of respondents: 37. 4. Patient to nurse ratios often high compared to local non-VA hospitals; Checked (percentage): 40; Number of respondents: 35. 5. Salary level for new graduate RN was low compared to local area; Checked (percentage): 31; Number of respondents: 27. 6. VA patients are sicker and more complex than patients at local non- VA hospitals; Checked (percentage): 30; Number of respondents: 26. 7. Retention incentives were not sufficient; Checked (percentage): 26; Number of respondents: 23. 8. Other reasons; Checked (percentage): 21; Number of respondents: 18. 9. Retention incentives were not available; Checked (percentage): 17; Number of respondents: 15. 10. Reimbursement for continuing education not sufficient; Checked (percentage): 17; Number of respondents: 15. 11. Reimbursement for continuing education not available; Checked (percentage): 8; Number of respondents: 7. 12. VA benefit package was not as attractive as local competitors; Checked (percentage): 7; Number of respondents: 6. Source: GAO. [End of table] Question 15: As of March 31, 2007, which of the following staffing methodologies were used to determine staffing levels for inpatient care for inpatient units at this facility? 1. Professional judgment; Checked (percentage): 73; Number of inpatient unit respondents: 357. 2. RN skill mix; Checked (percentage): 68; Number of inpatient unit respondents: 336. 3. Average daily census; Checked (percentage): 62; Number of inpatient unit respondents: 303. 4. Hours per patient day; Checked (percentage): 61; Number of inpatient unit respondents: 302. 5. VA PCS; Checked (percentage): 58; Number of inpatient unit respondents: 287. 6. Typical patient acuity; Checked (percentage): 58; Number of inpatient unit respondents: 284. 7. Historic full-time equivalent employees (FTEEs); Checked (percentage): 50; Number of inpatient unit respondents: 245. 8. Nurse to patient ratios; Checked (percentage): 44; Number of inpatient unit respondents: 218. 9. Benchmark against other VAMCs; Checked (percentage): 39; Number of inpatient unit respondents: 192. 10. VA expert panel; Checked (percentage): 39; Number of inpatient unit respondents: 191. 11. Facility or VISN budget; Checked (percentage): 37; Number of inpatient unit respondents: 184. 12. National benchmarks; Checked (percentage): 33; Number of inpatient unit respondents: 162. 13. American Nurses Association guidelines; Checked (percentage): 27; Number of inpatient unit respondents: 133. 14. Guidance from the VISN; Checked (percentage): 14; Number of inpatient unit respondents: 70. 15. Minimum Data Set (MDS); Checked (percentage): 14; Number of inpatient unit respondents: 70. 16. VA directive based on staffing recommendations from Paralyzed Veterans of America (PVA); Checked (percentage): 3; Number of inpatient unit respondents: 15. 17. Directive on polytrauma units; Checked (percentage): 1; Number of inpatient unit respondents: 4. Source: GAO. Note: Responses for 492 inpatient units were reported by VA nurse executives. [End of table] [End of section] Appendix III: Summary of RN Focus Group Questions and Responses at the Eight VAMCs We Visited: The table lists the questions we asked RNs at the eight VAMCs we visited and their five most frequent responses. Table 4: Question: What attracted you to work at VA?; Five most frequent focus group participants' responses: * Benefits; * Like working with veterans; * Education assistance; * Salary; * Job security, positive prior experience, and ability to transfer to other VAMCs. Question: What keeps you working at VA?; Five most frequent focus group participants' responses: * VA's patient population; * Retirement benefits; * Positive work environment; * Benefits; * Professional development. Question: What is not at your facility now that would keep you working here?; Five most frequent focus group participants' responses: * Adequate staffing; * More opportunity for promotion and advancement; * More ancillary and non-nursing support staff; * Flexible schedule; * More funding for educational programs. Question: What contributes to the RN shortage?; Five most frequent focus group participants' responses: * Hiring process; * Inadequate number of staff; * Absenteeism; * Lack of support for new hires; * Lack of support staff (i.e., RNs performing non-nursing tasks). Question: How would you improve recruitment and hiring?; Five most frequent focus group participants' responses: * Shorten VA's hiring process; * Better communication with applicants; * Better advertisement for RN positions; * Offer recruitment bonuses; * Increase outreach (i.e., to nursing schools). Question: How is staffing determined?; Five most frequent focus group participants' responses: * Available full-time equivalent employees (FTEE); * Nurse-to-patient ratios; * Patient acuity; * Patient census; * Patient classification system. Question: What types of supplemental staffing strategies are used?; Five most frequent focus group participants' responses: * Float staff among units; * Voluntary overtime; * Work short staffed; * Contract/ Agency nurses; * A combination of voluntary and mandatory overtime. Source: GAO focus group sessions with VA RNs. Note: RNs raised several miscellaneous issues that were excluded from our analysis tabulating the five most frequent responses at the focus groups. [End of table] [End of section] Appendix IV: Comments from the Department of Veterans Affairs: The Deputy Secretary Of Veterans Affairs: Washington: October 10, 2008: Mr. Randall Williamson: Director, Health Care: U.S. Government Accountability Office: 441 G Street, NW: Washington, DC 20548: Dear Mr. Williamson: The Department of Veterans Affairs (VA) has reviewed your draft report, VA Health Care: Improved Staffing Methods and Greater Availability of Alternate and Flexible Work Schedules Could Enhance the Recruitment and Retention of Inpatient Nurses, (GAO-09-17) and concurs with your findings and recommendations. VA has long recognized the need to develop an automated, data driven staffing methodology and patient acuity system that uses patient workload and other indicators to reliably determine nursing staff requirements and accurately reflect the levels of acuity in acute patient units. Considering the innumerable variables that impact the day-to-day patient care demands on nursing staff, especially in a system as massive and varied as VA, the development of such a system continues to be a challenge. Balancing the prescriptiveness that an automated national nursing staffing methodology will necessarily entail with the unique patient care needs of individual facilities and individual nursing units will require that flexibility be built into the system. It will also be necessary to develop an infrastructure that allows integration with other appropriate VA databases. The Veterans Health Administration (VHA) has laid the groundwork to implement such a standardized, fully automated staffing methodology for nursing personnel that is anticipated to be accomplished in three phases. Our ultimate goal is to implement an automated, data-driven, nationally standardized staffing methodology for VHA nursing personnel at all points of care by 2012. I am also in agreement with your observations about the barriers that frequently hinder optimal levels of recruitment and retention of VA registered nurses, and with your recommendation to more fully assess how these barriers might be overcome. At the same time, I am pleased to report that 2008 was a banner year for VA in hiring new registered nurses. Nevertheless, the Office of Nursing Service plans to charter a special task force to address this recommendation and to identify options for expanding alternate and flexible work schedules. This process is anticipated to be completed by mid-June 2009. of the process. I am confident, however, that steps are being taken in the right direction, and that this goal will be accomplished. The enclosures specifically address GAO's recommendations and provide comments to the draft report. VA appreciates the opportunity to comment on your draft report. Sincerely yours, Signed by: Gordon H. Mansfield: Enclosures: Enclosure I: Department of Veterans Affairs Comments to Government Accountability Office (GAO) Draft Report VA HEALTH CARE: Improved Staffing Methods and Greater Availability of Alternate and Flexible Work Schedules Could Enhance the Recruitment and Retention of Inpatient Nurses (GAO-09-17): GAO Recommendations: Recommendation 1: Develop a detailed action plan that includes a timetable for building, testing, and implementing the new nurse staffing system. Response: Concur. In January 2008, the Office of Nursing Services chartered a Staffing Methodology Action Team (SMAT) to revise the current staffing methodology process for nursing personnel. The SMAT is composed of two subgroups, a Development Team and a Review Team. Throughout fiscal 2008, the SMAT has developed a plan for the creation, testing, and implementation of a new staffing methodology system for nursing personnel on inpatient units. A new draft Veterans Health Administration (VHA) directive, Staffing Methodology for VHA Nursing Personnel, has also been developed. The plan has three phases. The initial pilot implementation phase focuses on nursing staff requirements for all inpatient points of care. During this phase, nurse managers from participating facilities will be provided with a staffing methodology workbook, which includes the necessary tools and methods to identify, aggregate, and compare data with other nursing units, ultimately leading to national staffing benchmarks. The pilot will involve the participation of appropriate nursing and administrative personnel from Veterans Integrated Service Networks (VISNs) 1, 6, 15, 16 and 20 and will test the new process. Training tools have been developed in conjunction with the Employee Education System. A national pilot site training session for participants will be held in Dallas, Texas on October 7 and 8, 2008. Throughout the rest of October, extensive efforts will be devoted to preparing the facilities for launch of the pilot test, which will occur between November 3, 2008 and January 31, 2009. With feedback from the participating facilities, the process will then be carefully evaluated and refinements to the methodology will be addressed as necessary. This process is expected to be completed by the end of March 2009. (See Enclosure II for a detailed description of the pilot implementation.) The second phase in the development of a nationally-standardized methodology process will provide guidance to include additional points of care, other than inpatient, and the third and final phase will provide the guidance for a fully automated system that encompasses initial data entry through report production. This will include automated methods that integrate with VA databases, including the Decision Support System (DSS), the pay system and the VA Nursing Outcomes Database (VANOD). VANOD reports enable personnel to correlate nursing sensitive indicators and patient outcomes to evaluate staffing effectiveness. In addition, a statement of work for a nurse staffing system that consists of an automated scheduler application and a patient acuity application has been completed. The software will include flexibility options to be customized based on the facilities need and the ability to generate reports to optimize staffing levels. This project is pending approval, and the Office of Nursing Service will work with the Office of Information on the very high level needs that are required to complete Phase 3. Based on the results of the pilot implementation plan, the draft Directive will be revised as needed and submitted for concurrence and approval. This action is expected to be completed by April 2009, after which time the new staffing methodology will be incrementally implemented throughout the remaining VISNs. Recommendation 2: Ensure the new nurse staffing system provides Registered Nurse (RN) staffing estimates which accurately account for both the actual inpatient acuity levels and current nursing tasks performed on inpatient units and adequately take into account the level of ancillary and nursing support that is available on Veterans Affairs Medical Center inpatient units. Response: Concur. VHA considers the non-nursing tasks and level of ancillary/nursing support to be part of the indicators that units recognize to calculate full-time equivalents (FTE). The process addresses these tasks by identifying them as indicators in the new staffing methodology process that is described in the draft directive and on the Excel spreadsheet for the FTE calculator. The staffing projections will be determined based on direct care staff responsibilities, and will include indicators for ancillary and nursing support in the work environment. Recommendation 3: Assess the barriers to wider availability of alternate and flexible work schedules for RNs at VAMCs and explore ways to overcome them. Response: Concur. The Office of Nursing Service, in conjunction with the Office of Workforce Management and Consulting, will convene a task force composed of relevant program managers, as well as facility directors, nurse executives and nurse managers, to fully assess barriers to the effective use of alternate and flexible work schedules for RNs and to identify potential solutions for overcoming the barriers. The task force will present its findings by June 2009, and plans for the initiation of implementation actions will be established as feasible. [End of section] Appendix V: GAO Contact and Staff Acknowledgments: GAO Contact: Randall B. Williamson at (202) 512-7114 or williamsonr@gao.gov. Acknowledgments: In addition to the contact named above, Marcia A. Mann, Assistant Director; N. Rotimi Adebonojo; Mary Ann Curran; Linda Diggs; Martha A. Fisher; Krister Friday; Susannah Bloch; and Suzanne Worth made major contributions to this report. [End of section] Related GAO Products: VA Health Care: Recruitment and Retention Challenges and Efforts to Make Salaries Competitive for Nurse Anesthetists. GAO-08-647T. Washington, D.C.: April 9, 2008. VA Health Care: Many Medical Facilities Have Challenges in Recruiting and Retaining Nurse Anesthetists. GAO-08-56. Washington, D.C.: December 13, 2007. Nursing Workforce: HHS Needs Methodology to Identify Facilities with a Critical Shortage of Nurses. GAO-07-492R. Washington, D.C.: April 30, 2007. Nursing Workforce: Emerging Nurse Shortages Due to Multiple Factors. GAO-01-944. Washington, D.C.: July 10, 2001. [End of section] Footnotes: [1] Robert L. Kane et al, "Nurse Staffing and Quality of Patient Care," Pub. No. 07-E005, March 2007, Agency for Healthcare Research and Quality (AHRQ); and Kaiser Permanente Institute for Health Policy, "Nurse Staffing and Care Delivery Models: A Review of the Evidence" (March 2002). [2] Mark W. Stanton, and Margaret K. Rutherford, "Hospital Nurse Staffing and Quality of Care." Agency for Healthcare Research and Quality, Research in Action, Issue 14, AHRQ, Pub. No. 04-0029, Rockville, Md.: 2004. [3] Office of Inspector General, Department of Veterans Affairs, "Evaluation of Nurse Staffing in Veterans Health Administration Facilities" (Aug. 13, 2004). [4] VA concurred with all the findings and recommendations of the OIG's report. [5] See Holly A. DeGroot, "Patient Classification System Evaluation, Part 1: Essential System Elements," Journal of Nursing Administration, Vol. 19, No. 6, June 1989; Holly A. DeGroot, "Patient Classification System Evaluation, Part 2: System Selection and Implementation," Journal of Nursing Administration, Vol.19, No. 7, July/August 1989; Jean A. Seago, "A Comparison of two Patient Classification Instruments in an Acute Care Hospital," Journal of Nursing Administration, Vol. 32, No. 5, May 2002; Kelle Harper and Crystal McCully, "Acuity Systems Dialogue and Patient Classification System Essentials," Nursing Administration Quarterly, October-December 2007, Vol. 31, No. 4. One study on PCS identifies key features of a PCS, which include a means to accurately predict individual patient nursing care requirements; a method for periodically validating the amount and type of care delivered to each type of patient; and a system for monitoring the reliability of the PCS over time. [6] Statement of William F. Feeley, Deputy Under Secretary for Health for Operations and Management, Department of Veterans Affairs, House Committee on Veterans' Affairs, Subcommittee on Health, October 18, 2007. [7] Department of Veterans Affairs, Workforce Succession Strategic Plan, Fiscal Year 2007-2011 (Washington, D.C.: October 2006). [8] For our study, VA inpatient units include medicine, surgery, behavioral health, medical intensive care, critical care, spinal cord injury, polytrauma, and nursing home care units. [9] A VAMC nurse executive is a member of the executive leadership team at a VAMC and is involved in strategic planning, organizational assessment and program development at the VAMC. The nurse executive is also responsible for all nursing care delivered at the VAMC. [10] These VAMC nurse managers are typically responsible for supervising the RNs who provide care on an inpatient unit as well as managing operation of the unit. [11] VA requires medical center officials to verify the state licenses and national certificates of all health care practitioners--including RNs--hired by VA using its Web-based database of practitioners' credentials, VetPro. [12] For more details on Magnet™ status see [hyperlink, http://nursecredentialing.org/magnet/] (accessed on Sept. 4, 2008). [13] The ANCC considers 12 nurse sensitive quality indicators, including data on the level of nursing care needed, patient injuries and nurse satisfaction. As part of the Magnet™ status review process ANCC evaluates a written proposal submitted by the hospital, conducts site visits to the hospital and receives public comments from the local community. [14] See Robert L. Kane et al. [15] VA's Employee Incentive Scholarship Program which began in March 2000, pays employees' tuition for nursing degree programs. See 38 U.S.C. §7621 (2000). VA's Education Debt Reduction Program, a student loan repayment program began in May 2002, and is available to newly hired clinicians, such as RNs. See 38 U.S.C. §7681 (2000). VA's National Nursing Education Initiative is a scholarship program for VA RNs enrolled in nationally accredited education programs and the VA Nursing Education for Employees Program provides salary replacement to employees enrolled full-time in LPN or RN nursing education programs. [16] Members of the work group included a medical center director, an HR officer, and a nurse recruiter. [17] Pub. L. No. 108-445, Department of Veterans Affairs Health Care Personnel Enhancement Act of 2004, §4, 118 Stat. 2636, 2643-45. [18] In this report, we use the term flexible work schedule to refer to both flexible and compressed work schedules. A compressed work schedule is an 80-hour biweekly basic work requirement that is scheduled for less than the usual 10 workdays--for example, a schedule consisting of eight days of 10-hour shifts. [19] Certain health professionals at VA are exempt from this basic work schedule. [20] We did not independently assess the HPPDs generated by PCS to determine whether they were adequate for the needs of various inpatient units. [21] BCMA allows nurses to document and track the prescribed medication given to patients using bar code technology. [22] CPRS is VA's electronic patient medical record that was implemented in 1998 and contains all medical record documentation including lab test results, radiologic images, and patient appointments. VA clinicians can access CPRS from anywhere within VA's health care system. [23] The VA OIG found that nurse managers do not follow a standardized methodology and instead use a variety of factors for determining nurse staffing. [24] VA nurse executives also reported that VA's PCS is one data source used by nurse managers when determining RN staffing; however, as noted earlier, nurse executives and nurse managers we interviewed reported that the data from PCS were outdated, inaccurate, and not very useful. [25] See GAO, Nursing Workforce: Multiple Factors Create Nurse Recruitment and Retention Problems, GAO-01-912T (Washington, D.C.: June 27, 2001). See the end of this report for a list of related GAO products. [26] During a hiring freeze, vacancies can only be filled after receiving authority from a medical center resource board or similar VA medical center entity. During a hiring lag, only a limited number of vacancies can be filled at the medical center or vacant positions can only be filled with approval from a resource board. [27] VA delegates decision making authority regarding financial and delivery of health care services to its 21 regional health care networks, including most budget and management responsibilities concerning VAMC operations. [28] The resource board usually consists of VA medical center officials--such as an associate director, chief of staff, HR chief, chief fiscal officer, and nurse executive--that recommend funding for medical center positions including nursing service positions. [29] See GAO, Older Workers: Federal Agencies Face Challenges, but Have Opportunities to Hire and Retain Experienced Employees, GAO-08-630T (Washington, D.C.: Apr. 30, 2008). [30] The two schedules are: working three 12 hours shifts each week or working 9 months out of every 12 months to count as full-time employment. [31] GAO estimated the use of alternate work schedules by VA RNs using data from the VA Personnel Accounting and Integrated Data files, which house VA's payroll and HR information. The data files did not contain information that could be used to estimate the extent of the use of all flexible work schedules by VA RNs. [32] The 12-hour flexible work schedule is a combination of one 8 and six 12-hour shifts totaling 80 hours within a 2-week period. [33] The 10-hour flexible work schedule is comprised of eight 10-hour shifts within a 2-week period. [34] In addition to the challenges cited by nurse managers and inpatient RNs, nurse executives we surveyed cited staffing issues and salaries as issues affecting retention. [35] Nurse executives we surveyed reported that voluntary overtime (92 percent), floating RNs between inpatient units (85 percent), the use of part-time RNs (52 percent), and working short-staffed (48 percent) are strategies used to supplement RN staffing on VA inpatient units. [36] To provide greater leadership opportunities for RNs, VA implemented the clinical nurse leader role, which is sponsored by the American Association of Colleges of Nursing. As of August 2007, over 80 VAMCs have requested permission to participate in this initiative. [37] The other issues cited by nurse executives in our survey and in our interviews that contribute to difficulty recruiting RNs were the required rotation of RNs from one shift to another, poor coordination in making a prompt salary offer, a shortage of RNs, and lower VA RN salary levels compared to the local market area. [38] A 2005 industry survey of hospitals in New York reported that 60 percent streamlined hiring to interview and evaluate candidates as well as make job offers on the same day. [39] A medical center resource board may not grant approval to fill a vacant RN position on the first request, thus resulting in delays in posting RN vacancy notices. [40] VA policy states that an RN's starting salary is determined after the individual's educational qualifications and nursing experience are evaluated by the nurse professional standards board at the VAMC. According to VA, the medical center's nurse professional standards board recommends the appropriate grade for appointment and advancement. The board is comprised of members who are at a grade that is equal to or higher than that of the applicant being considered, covers the range of practice within an occupation being considered, and serves a term of 2 or more years, according to VA. [41] Hiring to fill RN vacancies cannot typically begin during a hiring freeze or lag in hiring. During a hiring freeze, RN vacancies can only be filled after receiving authority from a medical center resource board or similar VA medical center entity. Hiring freezes and hiring lags are used for budgetary reasons or to manage personnel costs. [42] Members of the work group included a medical center director, HR officer, and nurse recruiter. [43] These VAMC nurse managers are typically responsible for supervising RNs who provide care on an inpatient unit--such as intensive care and nursing home units--as well as managing operation of the unit. [44] Three of the eight VAMCs we visited have attained Magnet™ status. GAO's Mission: The Government Accountability Office, the audit, evaluation and investigative arm of Congress, exists to support Congress in meeting its constitutional responsibilities and to help improve the performance and accountability of the federal government for the American people. GAO examines the use of public funds; evaluates federal programs and policies; and provides analyses, recommendations, and other assistance to help Congress make informed oversight, policy, and funding decisions. GAO's commitment to good government is reflected in its core values of accountability, integrity, and reliability. Obtaining Copies of GAO Reports and Testimony: The fastest and easiest way to obtain copies of GAO documents at no cost is through GAO's Web site [hyperlink, http://www.gao.gov]. Each weekday, GAO posts newly released reports, testimony, and correspondence on its Web site. To have GAO e-mail you a list of newly posted products every afternoon, go to [hyperlink, http://www.gao.gov] and select "E-mail Updates." Order by Mail or Phone: The first copy of each printed report is free. Additional copies are $2 each. A check or money order should be made out to the Superintendent of Documents. GAO also accepts VISA and Mastercard. Orders for 100 or more copies mailed to a single address are discounted 25 percent. Orders should be sent to: U.S. Government Accountability Office: 441 G Street NW, Room LM: Washington, D.C. 20548: To order by Phone: Voice: (202) 512-6000: TDD: (202) 512-2537: Fax: (202) 512-6061: To Report Fraud, Waste, and Abuse in Federal Programs: Contact: Web site: [hyperlink, http://www.gao.gov/fraudnet/fraudnet.htm]: E-mail: fraudnet@gao.gov: Automated answering system: (800) 424-5454 or (202) 512-7470: Congressional Relations: Ralph Dawn, Managing Director, dawnr@gao.gov: (202) 512-4400: U.S. Government Accountability Office: 441 G Street NW, Room 7125: Washington, D.C. 20548: Public Affairs: Chuck Young, Managing Director, youngc1@gao.gov: (202) 512-4800: U.S. Government Accountability Office: 441 G Street NW, Room 7149: Washington, D.C. 20548: