This Article Abstract Full Text (PDF) Services Download to citation manager Citing Articles Citing Articles via HighWire PubMed PubMed Citation

Prospects and Pitfalls: Use of the RAI-HC Assessment by the Department of Veterans Affairs for Home Care Clients

Correspondence: Address correspondence to Catherine Hawes, PhD, Department of Health Policy and Management, School of Rural Public Health – Mail Stop 1266, Texas A&M University System Health Science Center, College Station, TX 77843-1266. E-mail: hawes{at}srph.tamhsc.edu

	Abstract

TOP Abstract Rationale for Standardized... Content of the RAI-HC Testing the RAI-HC for... Potential DVA Applications of... Challenges to Using the... Conclusions References

 
Purpose: The U.S. Department of Veterans Affairs has adopted two functional assessment systems that guide care planning: one for nursing home residents (the Resident Assessment Instrument [RAI]) and a compatible one for home care clients (RAI-HC). The purpose of this article is to describe the RAI-HC (often referred to as the Minimum Data Set–Home Care or MDS-HC) and its uses and offer lessons learned from implementation experiences in other settings. Design and Methods: We reviewed implementation challenges associated both with the RAI and the RAI-HC in the United States, Canada, and other adopter countries, and drew on these to suggest lessons for the Department of Veterans Affairs as well as other entities implementing the RAI-HC. Results: Beyond its clinical utility, there are a number of evidence-based uses for the assessment system. The resident-level data can be aggregated and analyzed, and scales identify clinical conditions and risk for various types of negative outcomes. In addition, the data can be used for other programmatic and research purposes, such as determining eligibility, setting payment rates for contract care, and evaluating clinical interventions. At the same time, there are a number of implementation challenges the Department of Veterans Affairs and other organizations may face. Implications: Policy makers and program managers in any setting, including state long-term-care programs, who wish to implement an assessment system must anticipate and address a variety of implementation problems with a clear and consistent message from key leadership, adequate training and clinical support for assessors, and appropriate planning and resources for data systems.

Key Words: Minimum Data Set • MDS-HC • • Home health

This article focuses on the RAI-HC system, its potential applications, the likely challenges involved in using such a standardized assessment tool, and how the DVA might address these obstacles. However, our conclusions may well apply to other entities and governments considering implementation of the RAI-HC, including U.S. states and other countries.

	Rationale for Standardized Assessment and Genesis of the RAI-HC

TOP Abstract Rationale for Standardized... Content of the RAI-HC Testing the RAI-HC for... Potential DVA Applications of... Challenges to Using the... Conclusions References

 
The demand for a standardized home care assessment system has emerged from a variety of developments in health care. These include changes in the Medicare hospital reimbursement policy and resulting concerns about the continuity and quality of post-acute care (Kosecoff et al., 1990); the efficacy of geriatric assessment and subsequent work to employ these principles more broadly (Applegate, Akins, Vanderzwagg, Thoni, & Baker 1983; Hendricksen, Lund, & Stromgard, 1984; Rubenstein, Rhee, & Kane, 1982; Tulloch & Moore, 1979); the growing demand to achieve a better balance between community-based and institutional care options; rising state expenditures for Medicaid long-term-care services; and the promotion of computerized data systems to monitor quality, evaluate policies, and forecast future system needs.

As a result of these factors, researchers and clinicians have developed several standardized tools that have been adopted by state and federal policy-makers to improve various aspects of home health and home care services (Tonner, LeBlanc, & Harrington, 2001). One was the Outcome and Assessment Information Set (OASIS), developed under a mandate from the what is now the Centers for Medicare & Medicaid Services (CMS) with cofunding from the Robert Wood Johnson Foundation (CMS, 2006; Hittle et al., 2003). The original purpose of OASIS was to develop a set of items that would form the basis for measuring outcomes of Medicare home health clients, and this grew into a set of outcome-based quality improvement indicators. OASIS data are used to report 41 home health quality measures, 30 of which are risk adjusted and 11 of which are descriptive (CMS, 2006). Thus, OASIS was developed as a group of data elements that represented a core set of sociodemographic, environmental, support, and functional status items rather than a comprehensive assessment tool to be used for clinical assessment and care planning. Since then, OASIS items have been adopted for use in case-mix adjustment of Medicare's prospective payment system for home health care. In addition, over the past decade, additional items were added to OASIS that a workgroup of home care experts deemed essential for clients, although the goal "was not to produce a comprehensive assessment instrument, but to provide a set of data items necessary for measuring patient outcomes and essential for assessment—which HHAs in turn could augment as they judge necessary" (CMS, 2006, background page; see also links to the main OASIS components and general application, http://www.cms.hhs.gov/OASIS/Downloads/maincomponentsandgeneralapplication.pdf). The OASIS assessment is performed on every patient who receives services from home health agencies that participate in the Medicare program.

Unlike OASIS, the RAI-HC was developed for a clinical purpose—to form the basis of a comprehensive client assessment and to guide individualized client service planning for home health and home care services. As such, it mirrored the development and intent of the nursing home RAI, which was the first of what has become a family of instruments developed to improve long-term care for elderly and disabled individuals through better assessment and care planning. The Institute of Medicine (IOM) embraced the goal of improving care through geriatric assessment in its 1986 report on improving the quality of care in nursing homes. The IOM envisioned a uniform, comprehensive functional assessment of a person's strengths, preferences, and needs, systematically linked to an individualized plan of care, as a cornerstone of improved quality (IOM, 1986). CMS funded the development of the nursing home RAI pursuant to Congressional nursing home reforms in the Omnibus Budget Reconciliation Act of 1987. This development has been extensively discussed elsewhere (Hawes, Morris, Phillips, Mor, & Fries, 1995; Hawes et al., 1997; Mor et al., 1997; Morris et al., 1990; Phillips et al., 1997).

Seeing the benefits of the original RAI, its developers undertook the design of a similar assessment system for home care as part of a research consortium, interRAI, that has since created an integrated family of assessment instruments (www.interrai.org). The RAI-HC (Morris et al., 1997) was developed in 1994, and Version 2.0 was released in 1999 (Morris, Fries et al., 2000). As we discuss below, the RAI-HC is a more comprehensive assessment system, with boarder clinical and quality improvement intent and uses, than OASIS. Reflecting their different intents, OASIS and the RAI-HC are not equivalent. For example, a cross-walk of the two instruments in the mid-1990s found that only 15 of the 30 RAI-HC Clinical Assessment Protocols (CAPS; which are key mechanisms for linking assessment information to problem identification and care plan decisions; described below) could be replicated in OASIS, whereas there were no major OASIS areas lacking in the RAI-HC (Morris, 1997).

	Content of the RAI-HC

TOP Abstract Rationale for Standardized... Content of the RAI-HC Testing the RAI-HC for... Potential DVA Applications of... Challenges to Using the... Conclusions References

 
The RAI-HC system (often referred to by the name of the assessment instrument, the MDS-HC) highlights issues related to functioning and quality of life for individuals living in the community. It thus provides a common language for assessing an individual's health status, care needs, and preferences. The RAI-HC also systematically links a person's assessment information to care-planning choices and guides the development of individualized service plans. It is intended for use with adults in home care settings and is generally used with frail elders and individuals with disabilities who receive formal health care or supportive home care services.

The RAI-HC system comprises two parts. The first is an assessment instrument (the MDS-HC) designed to collect standardized information on a broad range of domains critical to caring for individuals in the community, as displayed in Table 1.

The second part of the RAI-HC system consists of 30 problem-focused CAPs (see Table 2). The CAPs cover conditions that are common risks for home care clients, such as risk for functional decline, nursing home placement, or abuse.

The RAI-HC is designed for use at a client's admission to a home health or home care program or prior to hospital discharge, but its utility is augmented by reassessment at some standard interval (e.g., 90 days), depending on the type of client being assessed, when it can identify changes, support revision of care plans, and provide outcome measures. In addition, the variability among home care clients, from those needing clinically complex care or intensive rehabilitation to those who need only ongoing assistance with ADLs affects the time needed to complete an assessment. However, on average, the RAI-HC assessment takes about 60 min to complete for a new client (Morris et al., 1997).

	Testing the RAI-HC for Reliability and Validity

TOP Abstract Rationale for Standardized... Content of the RAI-HC Testing the RAI-HC for... Potential DVA Applications of... Challenges to Using the... Conclusions References

 
The RAI-HC has passed critical tests for both interrater reliability and validity in multiple trials. The initial tests showed good consistency (weighted kappa of 0.7 or more) in dual independent assessments of randomly selected home care clients in five countries, including the United States, Canada, Japan, Australia, and the Czech Republic (Morris et al., 1997). These reliabilities were achieved using nurses, social workers, and physicians who worked for home care agencies, rather than research nurses. Function, health status, and service items, which are shared between the RAI-HC and the nursing home RAI, had excellent reliability (average weighted kappa of 0.74). Subsequent reliability trials have confirmed these findings (Kwan, Chi, Lam, Lam, & Chou, 2000; Landi et al., 2000).

The RAI-HC has also been tested for validity. Content validity was studied using stakeholder opinions and expert clinicians as the gold standard for whether RAI-HC items were clinically relevant and focused on key issues for home care clients. In addition, a group of experts developed each CAP and validated it through clinical focus groups and ongoing empirical research, including research on the ability of the trigger to predict the risk of a change in that CAP area (Morris, Fries et al., 2000).

Convergent validity was established by determining whether items from the RAI-HC would detect levels of physical functioning and cognitive status as defined by analogous well-established instruments. Trained nurses assessed home care clients independently of and with nurses blinded to the results from the research rating scales. RAI-HC items and scales compared well with ratings from the Barthel ADL Index (Pearson's correlation of.74), the Lawton Instrumental Activities of Daily Living Scale (also.74), and the Mini-Mental State Examination (MMSE;.81; Landi et al., 2000).

These findings show that the RAI-HC is very reliable, with good item reliability across nursing home and home- and community-based service settings, types of clients, cultures, and assessor qualifications. The trials also demonstrate that RAI-HC scale scores represent valid measures of physical function and cognitive status among home care clients, and that clinicians and stakeholders see the RAI-HC as clinically relevant and useful (Kwan et al., 2000; Landi et al., 2001; Leung et al., 2001).

	Potential DVA Applications of the RAI-HC

TOP Abstract Rationale for Standardized... Content of the RAI-HC Testing the RAI-HC for... Potential DVA Applications of... Challenges to Using the... Conclusions References

 
The Millennium Act required the DVA to adopt a uniform process to enable comprehensive assessment and care planning for its home care clients. Thus, the DVA's main purpose for implementing the RAI-HC is to improve clinical care. A key strength of the RAI-HC system is that it was designed for use with a variety of home care populations, ranging from individuals with complex clinical care needs to those who receive palliative care. This is particularly important in the DVA, which serves a diverse veteran population both directly through its own home-based primary care program and indirectly through services provided through contracts with external agencies. Furthermore, the use of a uniform assessment system increases the likelihood that all DVA clients will receive a comprehensive assessment of their needs and preferences and a plan of care systematically related to those characteristics, rather than ones determined by the characteristics and preferences of the person conducting the assessment or the agency providing services. A uniform mandatory assessment system is needed because DVA clients receive services from a wide variety of agencies and providers that may differ in their skills and the consistency of their assessments. With responsibility for the monitoring the cost of care and ensuring care quality, the DVA has an incentive to ensure that veterans receive the right care for the right cost in the right setting.

Although the DVA's initial goal may be to use the RAI-HC for assessment and service planning, there are other uses of the RAI-HC data that the DVA may find worthwhile. For example, the IOM cited the DVA as having a strong tradition of synthesizing empirical evidence and applying it to clinical care (Kohn, Corrigan, & Donaldson, 2000). Thus, the DVA may want to use RAI-HC client-level data to assess various aspects of process and outcome quality. Similarly, the DVA could use the data to generate profiles of its clients and make comparisons across different types of home care programs or long-term-care settings and services (e.g., to compare nursing home or assisted living residents to home care clients). The DVA could also use the data to examine the relative performance of different providers or the effect of different mixes of services on client outcomes. Such uses of the data could inform program and policy design for the DVA system. Furthermore, with common data elements across home care and nursing home populations, the DVA could examine factors that predict or explain why some veterans are referred for residential programs whereas others receive home care. In the remainder of this section, we discuss the multiple potential uses of RAI-HC data beyond its primary clinical uses.

Summary Scales
Embedded within the RAI-HC are scales that summarize functional status in the areas of ADLs, cognition, communication, pain, and mood. These scales generate validated summary measures of individual functioning in the specified areas and can be used for a variety of purposes, including to (a) summarize the status of individual clients or groups of clients, (b) identify change in an individual's clinical or functional status over time, (c) compare outcomes among different groups of clients over time, and (d) evaluate the effectiveness of different clinical interventions or various mixes of services.

Scales have been developed in the following areas:

• The ADL Hierarchy Scale measures functionality using selected ADL items to represent stages of the disablement process. Early loss ADLs (e.g., dressing) are assigned lower scores than late loss ADLs (e.g., eating). The scale ranges from 0 (no impairment) to 6 (total dependence; Morris, Fries, & Morris, 1999).
  
• The ADL Index is a summary measure of ADLs that combines scores for bed mobility, toileting, transferring, and eating. Scores range from 4 (independent or supervision in all 4 areas) to 15 (severe impairment in all 4 areas) and track well with estimated formal and informal time spent caring for the client (Björkgren, Fries, & Shugarman, 2000; Williams, Fries, Foley, Schneider, & Gavazzi, 1994).
  
• The Changes in Health, End-Stage Disease and Signs and Symptoms (CHESS) Scale identifies individuals at risk for serious decline. It can serve as an outcome indicator where the objective of care and services is to minimize problems related to declines in function. CHESS, originally developed for use in the nursing home, has been adapted for use with the RAI-HC. It uses six items to create a 6-point scale. In both the nursing home and home care populations, there is clear differentiation of all six levels of CHESS scores, and higher levels are predictive of adverse outcomes like mortality, hospitalization, pain, caregiver stress, and poor self-rated health (Hirdes, Frijters, & Teare, 2003).
  
• The Cognitive Performance Scale (CPS) combines information on memory impairment, level of consciousness, and decision making, with scores ranging from 0 (intact) to 6 (very severe impairment). A variety of validation studies have demonstrated that the CPS is highly correlated with the MMSE (Morris et al., 1994).
  
• The Depression Rating Scale, composed of seven RAI-HC mood items, can be used as a clinical screen for depression. Validation studies compared this scale with the Hamilton Depression Rating Scale and the Cornell Scale for Depression. Compared to Diagnostic and Statistical Manual of Mental Disorders (4th edition) major or minor depression diagnoses, the Depression Rating Scale was 91% sensitive and 69% specific at a cut-point score of 3 out of 7 (Burrows, Morris, Simon, Hirdes, & Phillips, 2000).
  
• The Instrumental Activities of Daily Living (IADL) Involvement Scale is based upon a sum of seven items: meal preparation, ordinary housework, managing finances, medications, phone use, shopping, and transportation. The codes for individual items are summed to produce a scale that ranges from 0 to 42, with higher scores indicating greater difficulty in performing IADLs (Morris, Carpenter, Berg, & Jones, 2000).
  
• The Pain Scale uses two RAI-HC items to create a score from 0 to 3. It has been shown to be highly predictive of pain on the Visual Analogue Scale in nursing home residents in the United States, although it has not yet been validated with a home care population (Fries, Simon, Morris, Flodstrom, & Bookstein, 2001).
  

Quality Indicators
Researchers have also developed and tested a set of quality indicators—the Home Care Quality Indicators (HCQIs)—that can be used in a quality-monitoring system.

Researchers derived the initial 22 HCQIs and associated risk adjusters using data from Canada, the United States, and Italy (Dalby, Hirdes, & Fries, 2005; Hirdes et al., 2004). The development parallels the work done with the nursing home RAI for CMS's Nursing Home Compare program (Morris, Murphy, Mor, Berg, & Moore, 2002).

One obvious advantage of the HCQIs is that they require only RAI-HC data, without additional data collection or reporting. Furthermore, the quality indicators include measures of both process (e.g., inadequate meals) and outcomes (e.g., functional decline). Each HCQI excludes clients for whom computations are inappropriate (such as terminally ill clients for evaluating weight loss), and certain outcomes have client-level risk adjusters. Also, HCQI agency adjusters can address selection and ascertainment bias, protecting, for example, an agency that might be penalized because it is more adept at recognizing mood disturbances (Dalby et al., 2005).

Case-Mix Classification
Case-mix systems classify clients into distinct service-use intensity categories that can help to allocate resources more fairly among clients with different needs. The RAI-HC data support the Resource Utilization Groups for Home Care (RUG-III/HC) algorithms and provide a measure of the relative resources (i.e., wage-weighted formal and informal care time) that clients are likely to use. RUG-III/HC also enables comparisons with resources that a client would have used if he or she had been in a nursing home (Fries et al., 1994). RUG-III/HC uses 74 RAI-HC variables to create 21 homogeneous resource-use categories and explains about one third (33.7%) of the variance in estimated resource use (Björkgren et al., 2000). Case-mix data also have important applications for analyzing populations served for policy research (Shugarman, Fries, & James, 1999) or adjusting quality indicators (Dalby et al., 2005).

Screening
Selected RAI-HC items can also be used in screening systems intended to identify appropriate care pathways. This helps in the difficult tasks of prioritizing target populations and allocating increasingly scarce public resources in an equitable manner. The data contained in the RAI-HC can support systematic, standardized methods for such screening. One example, used in Michigan, is the MI Choice system, which uses 32 RAI-HC items to classify individuals into one of five levels: nursing home, home care, intermittent personal care, homemaker, and information and referral services. During the assessment process, MI Choice can also serve as a complement to the assessor's clinical insights and the individual's preferences about the most appropriate care setting (Fries, Shugarman, Morris, Simon, & James, 2002). A separate telephone screening instrument using selected items could potentially be used to identify individuals who are unlikely to meet health, cognitive, and functional criteria for community-based or nursing home care and to target expensive in-person assessments for individuals more likely to qualify as medically eligible for assistance (Fries, James, Hammer, Shugarman, & Morris, 2004)

As part of the Millennium Act, the DVA adopted an adapted MI Choice screen—the Geriatrics and Extended Care Referral tool—as the standardized referral tool. DVA Direction 2004-059 mandated its use for identifying "the care needs of veterans and establishing a level of care prior to placement in all ... long-term care programs [paid, provided or coordinated by the DVA]" (U.S. Department of Veterans Affairs, 2004).

Other screeners can be derived from the RAI-HC. For Example, MAPLe classifies clients into five priority levels based on their risk for adverse outcomes (Hirdes, Poss, Curtin-Telegdi, & Chase, 2002). Michigan has recently adopted a system that makes selected light-care RUG-III/HC groups ineligible for a nursing home level of care. New Jersey and Louisiana have created state-specific medical eligibility screens from the RAI-HC, and similar efforts are underway in Georgia and Arkansas.

In addition to the above-mentioned uses, numerous studies have demonstrated the RAI-HC's applicability in identifying risk factors, describing the prevalence and outcomes of various health conditions, and identifying health service use patterns among home care clients (Fletcher & Hirdes, 2002; Landi, Cesari et al., 2004; Landi, Onder, Carpenter, Garms-Homolova, & Bernabei, 2005; Landi, Onder et al., 2004; Paddock & Hirdes, 2003; Sorbye, Finne-Soveri, Ljunggren, Topinkova, & Bernabei, 2005; Vik, Maxwell, & Hanley, 2005). Similarly, the Aged in Home Care study, funded by the 5^th Framework project of the European Union, compared models of home care for elders across 11 European countries along with the clinical and functional characteristics of their clients. This led to the creation of a cross-national population-based data set containing information on organizational characteristics of the home care systems and baseline and 1-year follow-up data on 4,010 home care clients. The study found differences in clients and care needs among the participating countries and generated several papers that used common data elements and definitions to assess elder abuse, compliance with medication, pain, vaccination, and inappropriate medication prescribing patterns (Carpenter et al., 2004).

	Challenges to Using the RAI-HC: Lessons for the DVA From Other Settings

TOP Abstract Rationale for Standardized... Content of the RAI-HC Testing the RAI-HC for... Potential DVA Applications of... Challenges to Using the... Conclusions References

 
The RAI-HC system is currently being used in Canada, Europe (Italy, Switzerland, France, Finland, Iceland, Denmark, the United Kingdom, Czechoslovakia, Germany), Asia (Hong Kong, Japan), Australia, and New Zealand (Black, Mitchell, Finlayson, & Peterson, no date; Coyte, Baranek, & Daly, 2000; Lincoln Centre for Ageing and Community Care Research, 2004; Martin & Martin, 2003; see also Milbank Memorial Fund, 2003). In the United States, it is being used in 11 states for a variety of purposes (see Table 3). These implementation experiences, as well as nearly a decade of RAI operation in U.S. nursing homes, provide some potentially valuable lessons for the DVA, as well as for new states, Canadian provinces, and other countries as they move to introduce the RAI-HC (Hawes, Vladeck, Morris, Phillips, & Fredeking, 2003; Milbank Memorial Fund, 2003). These include lessons from the comparative appraisals of a variety of standardized tools that have been conducted in several countries and findings such as the acceptability of the RAI instruments by potential users, the value users place on the availability of training materials and software support (Lincoln Centre for Ageing and Community Care Research, 2004), and the value stakeholders have found in the extensive development and testing process of the RAI-HC and its comprehensive nature (Black et al., no date). We next address implementation lessons with relevance to the issues of how to increase the accuracy of the data and their clinical utility, as well as how to create a useful data system.

Staff Turnover and Inadequate Training
In U.S. nursing homes, staff turnover and limited access to training and information have constituted major challenges to the accuracy of RAI assessment data. For instance, some facilities purchased a single copy of the RAI User's Manual and kept it under lock and key in an administrator's office (Harrington, Summers, Curtis, & Maynard, 1996). DVA administration can send a strong message about the importance of data quality by making RAI-HC training a priority. Ideally, this would include providing both initial and ongoing training to both clinical and administrative staff, developing a cadre of trained trainers, and ensuring that everyone who performs assessments has a user's manual. Other investments might include using computer-directed learning technology to facilitate flexible training for initial and ongoing training, establishing a Web-based "help desk" to respond to clinical questions, posting Frequently Asked Questions and updating them frequently, and ensuring that DVA contract agencies have parallel training requirements and access to the help desk.

RAI-HC as "Paperwork"
Some assessors will treat completion of the RAI-HC form strictly as paperwork, especially if they are not invested in the clinical utility of the data. The DVA can ameliorate this tendency by engaging staff in applying the data to everyday practice. Activities to evaluate intervention outcomes, to examine the costs of care among case-mix groups, or to measure caseload size and complexity may encourage staff to own the data and invest in good data quality. In addition, multiple applications provide countervailing incentives to over- or underreport a client's characteristics.

Burden From Duplicate Forms
Often the RAI-HC is implemented in a system in which other assessment tools are already in use. There is a tendency for staff to continue using these familiar tools. For example, staff may persist in employing the MMSE, even though the RAI-HC will yield a comparable cognitive status measurement. This practice increases staff workload, and continued use of duplicate instruments can lead to miscoding of items on one or both, which may affect measurement accuracy. Familiarizing staff with the validity of RAI-HC scales, making scale scores readily available to the assessor through software, and creating a process to identify and simplify duplicate assessment processes can help address this problem.

RAI Nurse
In U.S. nursing homes, many facilities assigned an MDS nurse to complete all assessments. This person often didnot work on the floor and thus lacked first-hand knowledge of the person being assessed. The MDS nurse was often not involved in care planning or other use of the assessment data. As a result, data accuracy was harmed. Engaging the same individuals to conduct the assessment and be part of the team developing the care plan is preferable and supports the collection of higher quality data.

Assessment in Another Care Environment
Some of the components of the RAI-HC require evaluation of the individual in his or her usual living environment. This can be a problem when assessment is part of hospital discharge planning or while the person resides in an assisted living facility. Program managers can encourage staff to identify such situations when the standard responses may not quite apply, always relying on the RAI-HC User's Manual. Ideally, the DVA will systematically compile any questions and develop uniform responses and interpretations at the national level, such as through the Office of Geriatrics and Extended Care, and ensure through trainings that they are known and applied in all sites.

Minimizing the Individual's Status
When a person's clinical condition triggers one of the CAPs, this should guide the care-planning team to consider an intervention. However, assessors may have incentives to overlook or not report a problem and thereby avoid expending additional fiscal or clinical resources on it or generating a negative quality indicator score (Schnelle, Bates-Jensen, Chu, & Simmons, 2004). This is an important and difficult problem to address. DVA training and continuing education can promote discussion of such incentives, including whether resource constraints are real and how else they might be addressed. Audit processes can identify underreporting and possibly minimize its occurrence.

Safeguarding the Clinical Utility of the RAI-HC
Implementation experiences in other venues have also identified common challenges to the clinical utility of RAI-HC data. We discuss possible strategies to address these below.

Missing the Forest for the Trees
When learning the RAI-HC system, most clinicians will necessarily focus on individual items and coding conventions. This may result in a preoccupation with the form itself rather than the links to service planning that represent the clinical power of the system. This tendency is directly addressed when users employ software that automatically calculates the CAP triggers. Training exercises in which staff assess clients regarded as difficult cases may demonstrate how CAP triggers and the additional assessment information called for in the guidelines can produce new clinical insights. Ensuring that care plans as well as assessment forms are included in the agency's audit and quality assurance activities may also help.

Cookie Cutter Service Plans
In the nursing home arena, the growth of RAIsoftware has been accompanied by the emergence of computer-generated standardized care plans. Such software features defeat the instrument's focus on individualized plans. The software selected or developed by the DVA should be designed to support rather than supplant clinical decision making, thus encouraging individualized service plans for each client.

Timely Access to Needed Data
Although software designers have created many innovative ways to organize and display RAI-HC data, these features may not be readily available to aid the staff who would most benefit from them because of factors such as insufficient access to computer hardware, low computer literacy among staff, budget woes, or inadequate system designs that provide such information only in management reports. Involving clinical staff and administrators alike in key software decisions and the plan for information flow within the organization is essential. Setting aside adequate funds for system rollout, including funds to create additional outputs identified by users as the system matures, will also help build broad ownership of the data.

Bolstering the RAI-HC Data System
It is also useful to consider possible issues regarding the data system at the policy level.

Lack of a Business Intelligence Plan
For a variety of reasons, nearly 7 years after nursing facilities began to submit RAI/MDS data, most states still lacked the capacity to analyze their own RAI data (Reinhard, Hendrickson, & Bemis, 2005), and little technical assistance has been forthcoming from CMS. The DVA can avoid this pitfall by investing in a thoughtful analysis of the desired uses of RAI-HC data. The national system can be designed to provide needed information to regional and local stakeholders, including contract providers, in a timely and coherent manner. Business information technology plans can address the differing needs of clinicians, administrators, and policy staff through identification of features and reports tailored to each group's needs.

Shifting Priorities at the Leadership Level
Creating a national data system is a multi-year endeavor with inevitable unforeseen problems. Unfortunately, the budget and staffing for such initiatives are often "low-balled" at the outset, so that by the time a crisis occurs that necessitates more resources, leadership interest and belief in the project has waned. The RAI-HC data system rollout will compete with other departmental priorities for leadership attention and funds, and a strategy is needed to maintain leadership support over the long run. This can include developing and ensuring an adequate multi-year budget when enthusiasm for the project is fresh; identifying issues that the leadership is particularly keen to know more about; creating reports that can keep key stakeholders interested; and widening support for the effort by educating other audiences, such as consumer advocates or legislators, about the "fruits" of the new system.

Investing in Future Analysis
Business intelligence planning should consider not only how data will be collected, but also who will analyze it. Although an initial set of standard reports is likely to be specified, decision makers will eventually need ad hoc reports, new standard reports, or studies requiring linked data sets, such as clinical assessment and cost data, as the system matures. Successful planning strategies include the following: identifying relevant staff who will need to be knowledgeable about the system's architecture and including them in the design and testing of the initial system, exposing such staff to the day-to-day clinical use of RAI-HC data so that they develop an appreciation for what the data mean, thinking through "make or buy" decisions regarding the extent to which internal versus contracted personnel will be used to carry out more complex analytic or research efforts, and updating data dictionaries frequently and making these accessible to internal and external users.

	Conclusions

TOP Abstract Rationale for Standardized... Content of the RAI-HC Testing the RAI-HC for... Potential DVA Applications of... Challenges to Using the... Conclusions References

 
The use of the RAI-HC within the DVA system offers many opportunities to ensure high-quality assessment and care planning for veterans who receive DVA-provided and -contracted home- and community-based services. Beyond its clinical uses, the RAI-HC provides data for a range of other purposes, including case-mix payment or resource allocation, quality assessment and monitoring, determination of eligibility for various services, and research efforts that can inform future policy decisions. By implementing a system in common with federal- and state-regulated programs, the DVA has taken an important step that will enable it to compare the quality of its long-term-care programs and services with other public sector efforts.

Although this article has focused on aspects of the RAI-HC, the same concepts apply to other interRAI assessment systems, including the assisted living tool (RAI-AL). The commonality of items and measurement approaches permits individuals to be followed across sectors of the long-term-care system over time. As an integrated health care system, the DVA provides an ideal environment in which to examine such integration.

	Footnotes

 
The research reported here was supported in part by the U.S. Department of Veterans Affairs, Veterans Health Administration, Health Services Research and Development Service. The views expressed in this article are our own and do not necessarily represent the position or policies of the Department of Veterans Affairs.

¹ Department of Health Policy and Management, Texas A&M University System Health Science Center, College Station.

² Institute of Gerontology, University of Michigan, Ann Arbor.

³ Ann Arbor Veterans Affairs Medical Center Geriatric Research, Education, and Clinical Center, MI.

⁴ Center for Management of Complex Chronic Care (CMC³), Edward Hines, Jr., VA Hospital, Hines, IL.

⁵ Institute for Health Services Research and Policy Studies, Northwestern University, Chicago, IL.

Decision Editor: William D. Spector, PhD

Received for publication May 13, 2005. Accepted for publication February 6, 2007.

	References

TOP Abstract Rationale for Standardized... Content of the RAI-HC Testing the RAI-HC for... Potential DVA Applications of... Challenges to Using the... Conclusions References

 

• Applegate, W. B., Akins, D., Vanderzwagg, R., Thoni, K., & Baker, M. G., (1983). A geriatric rehabilitation and assessment unit in a community hospital. Journal of the American Geriatrics Society, 31 206-210.[Medline]
• Björkgren, M. A., Fries, B. E., & Shugarman, L. R., (2000). Testing a RUG-III based case-mix system for home care. Canadian Journal on Aging, 19 (Suppl. 2), 106-125.
• Black, C., Mitchell, L., Finlayson, M., & Peterson, S., (n.d.). Enhancing capacity to study and evaluate home care in Manitoba. Retrieved January 23, 2007, from http://www.umanitoba.ca/centres/mchp/reports/pdfs/htf_english.pdf.
• Burrows, A., Morris, J. N., Simon, S., Hirdes, J. P., & Phillips, C., (2000). Development of a Minimum Data Set-based depression rating scale for use in nursing homes. Age and Ageing, 29 (2), 165-172.[Abstract/Free Full Text]
• Carpenter, I., Gambassi, G., Topinkova, E., Schroll, M., Finne-Soveri, H., & Henrard, J. C., et al (2004). Community care in Europe: The Aged in Home Care project (AdHOC). Aging Clinical and Experimental Research, 16 259-269.[Medline]
• Centers for Medicare and Medicaid Services. (2006). OASIS background. Retrieved January 25, 2007, from www.cms.hhs.gov/OASIS/02_Background.asp.
• Coyte, P. C., Baranek, R. M., & Daly, T., (2000, December). Identifying outcome indicators for evaluating services provided by community care access centres (CCACs). Retrieved January 23, 2007, from http://hcerc.utoronto.ca/PDF/short_outcomeindxccac2000.pdf.
• Crooks, V. C., Schnelle, J. F., Ouslander, J. P., & McNees, M. P., (1995). Use of the Minimum Data Set to rate incontinence severity. Journal of the American Geriatrics Society, 43 1363-1369.[Medline]
• Dalby, D. M., Hirdes, J. P., & Fries, B. E., (2005). Risk adjustment methods for home care quality indicators (HCQIs) based on the Minimum Data Set for Home Care. BMC Health Services Research, 5 (1), 7.[Medline]
• Fletcher, P. C., & Hirdes, J. P., (2002). Risk factors for falling among community-based seniors using home care services. Journal of Gerontology: Medical Sciences, 57A M504-M510.[Abstract/Free Full Text]
• Fries, B. E., James, M., Hammer, S. S., Shugarman, L. R., & Morris, J. N., (2004). Is telephone screening feasible? Accuracy and cost-effectiveness of identifying people medically eligible for home- and community-based services. The Gerontologist, 44 680-688.[Abstract/Free Full Text]
• Fries, B. E., Shugarman, L. S., Morris, J. N., Simon, S., & James, M., (2002). A screening system for Michigan's home and community-based long-term care programs. The Gerontologist, 42 462-474.[Abstract/Free Full Text]
• Fries, B. E., Simon, S. E., Morris, J. N., Flodstrom, C., & Bookstein, F. L., (2001). Pain in U.S. nursing homes: Validating a pain scale for the Minimum Data Set. The Gerontologist, 41 173-179.[Abstract/Free Full Text]
• Fries, B. E., Schneider DP, Foley WJ, Gavazzi M, Burke R, & Cornelius E., (1994). Refining a case-mix measure for nursing homes: Resource Utilization Groups (RUG-III). Medical Care, 32 (7): 668-685.[Medline]
• Harrington, C., Summers, P. R., Curtis, M., & Maynard, R., (1996). Study of the Resident Assessment Information System. San Francisco: University of California, San Francisco Department of Social and Behavioral Sciences.
• Hawes, C., Morris, J. N., Phillips, C. D., Mor, V., & Fries, B., (1995). Reliability estimates for the Minimum Data Set for nursing home resident assessment and care screening (MDS). The Gerontologist, 35 172-178.[Abstract]
• Hawes, C., Phillips, C. D., Morris, J., Mor, V., Fries, B. E., & Steele-Friedlob, E., et al (1997). The impact of OBRA-87 and the RAI on indicators of process quality in nursing homes. Journal of the American Geriatrics Society, 45 977-985.[Medline]
• Hawes, C., Vladeck, B. C., Morris, J. N., Phillips, C. D., & Fredeking, H., (2003). The RAI and the politics of long-term care: The convergence of science and politics in U.S. nursing home policy. Retrieved January 4, 2007, from http://www.milbank.org/reports/interRAI/030222interRAI.html.
• Hendricksen, C., Lund, E., & Stromgard, E., (1984). Consequences of assessment and intervention among elderly people: A three-year randomized controlled trial. British Medical Journal, 289 1522-1524.[Abstract/Free Full Text]
• Hirdes, J. P., Fries, B. E., Morris, J. N., Ikegami, N., Zimmerman, D., & Dalby, D., et al (2004). Home care quality indicators (HCQIs) based on the MDS-HC. The Gerontologist, 44 665-679.[Abstract/Free Full Text]
• Hirdes, J. P., Frijters, D., & Teare, G., (2003). The MDS CHESS scale: A new measure to predict mortality in the institutionalized elderly. Journal of the American Geriatrics Society, 51 96-100.[Medline]
• Hirdes, J. P., Poss, J., Curtin-Telegdi, N., & Chase, M., (2002, October). Method for Assigning Priority Levels (MAPLe) for community and institutional services. Presentation at the Annual Scientific and Educational Meeting of the Canadian Association on Gerontology, Montreal, Quebec, Canada.
• Hittle, D. F., Shaughnessy, P. W., Crisler, K. S., Powell, M. C., Richard, A. A., & Conway, K. S., et al (2003). A study of reliability and burden of home health assessment using OASIS. Home Health Care Services Quarterly, 22 (4), 43-63.[Medline]
• Institute of Medicine. (1986). Improving the quality of care in nursing homes. Washington, DC: National Academy Press.
• Kohn, L. T., Corrigan, J. M., & Donaldson, M. S., (Eds) (2000). To err is human: Building a safer health system. Washington, DC: National Academy Press.
• Kosecoff, J., Kahn, K. L., Rogers, W. H., Reinisch, E. J., Sherwood, M. J., & Rubenstein, L. V., et al (1990). Prospective payment system and impairment at discharge. The ‘quicker-and-sicker’ story revisited. Journal of the American Medical Association, 264 1980-1983.[Abstract/Free Full Text]
• Kwan, C. W., Chi, I., Lam, T. P., Lam, K. F., & Chou, K. L., (2000). Validation of the Minimum Data Set for Home Care assessment instrument for Hong Kong Chinese elders. Clinical Gerontology, 21 35-48.
• Landi, F., Cesari, M., Onder, G., Lattanzio, F., Gravina, E. M., & Bernabei, R., (2004). Physical activity and mortality in frail, community-living elderly patients. Journal of Gerontology: Medical Sciences, 59A 833-837.[Medline]
• Landi, F., Onder, G., Carpenter, I., Garms-Homolova, V., & Bernabei, R., (2005). Prevalence and predictors of influenza vaccination among frail, community-living elderly patients: An international observational study. Vaccine, 23 3896-3901.[Medline]
• Landi, F., Onder, G., Cesari, M., Barillaro, C., Lattanzio, F., & Carbonin, P. U., et al (2004). Comorbidity and social factors predicted hospitalization in frail elderly patients. Journal of Clinical Epidemiology, 57 832-836.[Medline]
• Landi, F., Onder, G., Tua, E., Carrara, B., Zuccalà, G., & Gambassi, G., et al (2001). Impact of a new assessment system, the MDS-HC, on function and hospitalization of homebound older people: A controlled clinical trial. Journal of the American Geriatrics Society, 49 1288-1293.[Medline]
• Landi, F., Tua, E., Onder, G., Carrara, B., Sgardari, A., & Rinaldi, C., et al (2000). Minimum Data Set for Home Care: A valid instrument to assess frail older people living in the community. Medical Care, 38 1184-1190.[Medline]
• Leung, A. C., Liu, C. P., Tsui, L. L., Li, S. Y., Tang, G. W., & Yau, D. C., et al (2001, Fall). The use of the Minimum Data Set Home Care in a case management project in Hong Kong. Care Management Journal, 3 8-13.
• Lincoln Centre for Ageing and Community Care Research. (2004, October). A review and identification of an existing validated, comprehensive assessment tool: Final report. Retrieved January 23, 2007, from http://www.health.vic.gov.au/subacute/assess.pdf.
• Martin, J. O., & Martin, I. R., (2003). Assessment of community dwelling older people in New Zealand: A review of the tools. Retrieved December 18, 2006, from http://www.nzgg.org.nz/guidelines/0030/Final_Report_tools_review.pdf.
• Milbank Memorial Fund. (2003). Implementing the Resident Assessment Instrument: Case studies of policymaking for long-term care in eight countries. Retrieved January 12, 2007, from http://www.milbank.org/reports/interRAI/030222interRAI.html.
• Mor, M., Intrator, O., Hiris, J., Fries, B. E., Phillips, C. D., & Hawes, C., et al (1997). Impact of the MDS on changes in nursing home discharge rates and destinations. Journal of the American Geriatrics Society, 45 1002-1010.[Medline]
• Morris, J., Fries, B. E., Steel, K., Ikegami, N., Bernabei, R., & Carpenter, I., et al (1997). Comprehensive clinical assessment in community setting: Applicability of the MDS-HC. Journal of the American Geriatrics Society, 45 1017-1024.[Medline]
• Morris, J., Murphy, K., Mor, V., Berg, K., & Moore, T., (2002). Validation of long term and post acute care quality indicators. Report to the Centers for Medicare & Medicaid Services. Roslindale, MA: Hebrew Rehabilitation Center for Aged, Research and Training Institute.
• Morris, J. N., (1997). RAI-Home Care clinical assessment protocols report. New York: United Hospital Fund.
• Morris, J. N., Carpenter, I., Berg, K., & Jones, R. N., (2000). Outcome measures for use with home care clients. Canadian Journal of Aging, 19 87-105.
• Morris, J. N., Fries, B. E., Bernabei, R., Steel, K., Ikegami, N., & Carpenter, I., et al (2000). RAI-Home Care (RAI-HC) assessment manual for version 2.0. Marblehead, MA: Opus Communications.
• Morris, J. N., Fries, B. E., Mehr, D. R., Hawes, C., Philips, C., & Mor, V., et al (1994). MDS Cognitive Performance scale. Journal of Gerontology: Medical Sciences, 49 M174-M182.
• Morris, J. N., Fries, B. E., & Morris, S. A., (1999). Scaling ADLs within the MDS. Journal of Gerontology: Medical Sciences, 54A M546-M553.[Abstract]
• Morris, J. N., Hawes, C., Fries, B. E., Phillips, C. D., Mor, V., & Katz, S., (1990). Designing the national Resident Assessment Instrument for nursing homes. The Gerontologist, 30 293-307.[Abstract]
• Paddock, K., & Hirdes, J. P., (2003). Acute health care service use among elderly home care clients. Home Health Care Services Quarterly, 22 (1), 75-85.[Medline]
• Phillips, C. D., Morris, J. N., Hawes, C., Mor, V., Fries, B. E., & Murphy, K., et al (1997). The impact of the RAI on ADLs, continence, communication, cognition, and psychosocial well-being. Journal of the American Geriatrics Society, 45 986-993.[Medline]
• Reinhard, S., Hendrickson, L., & Bemis, A., (2005, February). Using the Minimum Data Set (MDS) to facilitate nursing home transition. Community Living Exchange: Issue Brief. New Brunswick, NJ: Rutgers Center for State Health Policy. Retrieved April 24, 2007 from http://www.cshp.rutgers.edu/cle/products/MDS_Final_WEB.pdf.
• Rubenstein, L. Z., Rhee, L., & Kane, R. L., (1982). The role of geriatric assessment units in caring for the elderly: An analytic review. Journal of Gerontology, 37 513-521.[Abstract/Free Full Text]
• Schnelle, J. F., Bates-Jensen, B. M., Chu, L., & Simmons, S. F., (2004). Accuracy of nursing home medical record information about care-process delivery: Implications for staff management and improvement. Journal of the American Geriatrics Society, 52 1378-1383.[Medline]
• Shugarman, L. R., Fries, B. E., & James, M. L., (1999). A comparison of home care clients and nursing home residents: Can community based care keep the elderly and disabled at home? Home Health Care Services Quarterly, 18 (1), 25-45.[Medline]
• Sorbye, L. W., Finne-Soveri, H., Ljunggren, G., Topinkova, E., & Bernabei, R., (2005). Indwelling catheter use in home care: Elderly, aged 65+, in 11 different countries in Europe. Age and Ageing, 34 (4), 377-381.[Abstract/Free Full Text]
• Tonner, M. C., LeBlanc, A. J., & Harrington, C., (2001). State long-term care screening and assessment programs. Home Health Care Services Quarterly, 19 (3), 57-85.[Medline]
• Tulloch, A. J., & Moore, V., (1979). A randomized controlled trial of geriatric screening and surveillance in general practice. Journal of the Royal College of General Practitioners, 29 733-742.[Medline]
• U.S. Department of Veterans Affairs (2004, October 21). Geriatrics and Extended Care (GEC) Referral: VHA Directive 2004-059. Department of Veterans Affairs, Veterans Health Administration, Office of Geriatrics and Extended Care. Washington, DC. Retrieved April 26, 2007 from http://www1.va.gov/geriatricsshg/docs/VHADirective2004-059.pdf.
• Vik, S. A., Maxwell, C. J., & Hanley, D. A., (2005). Treatment of osteoporosis in an older home care population. BMC Musculoskeletal Disorders, Feb 11; 6: 7.
• Williams, B. C., Fries, B. E., Foley, W. J., Schneider, D., & Gavazzi, M., (1994). Activities of daily living and costs in nursing homes. Health Care Financing Review, 15 (4), 117-135.[Medline]

This article has been cited by other articles:

				D. Wieland and L. Ferrucci Multidimensional Geriatric Assessment: Back to the Future J. Gerontol. A Biol. Sci. Med. Sci., March 1, 2008; 63(3): 272 - 274. [Full Text] [PDF]

This Article Abstract Full Text (PDF) Services Download to citation manager Citing Articles Citing Articles via HighWire PubMed PubMed Citation