Research in Gerontological Nursing

The articles prior to January 2010 are part of the back file collection and are not available with a current paid subscription. To access the article, you may purchase it or purchase the complete back file collection here

Secondary Analyses and Qualitative Interviews 

The Impact of Policy on Nursing and Allied Health Services: Lessons from the Medicare Home Health Benefit

Joan K. Davitt, PhD, MSS, MLSP; Sunha Choi, PhD, MSW

Abstract

This article analyzes changes in Medicare home health staffing and service delivery patterns across three different reimbursement methods: cost based (1996), interim payment system (IPS) (1999), and the prospective payment system (PPS) (2002). This study combined secondary analysis of existing data (Provider of Services File and Statistical Supplement) with qualitative interviews of 22 home health agency directors to understand agency responses to policy changes created under the Balanced Budget Act of 1997. Cuts in staff and visits were greater under the IPS than they were under the PPS. Agencies cut staff and visits more dramatically for nonskilled services across both time periods. As a proportion of total services and visits, nursing and therapy services increased the most. Directors used various strategies to sustain the agency financially during these dramatic cuts in reimbursements, including eliminating staff, shifting staff roles, training staff on reimbursement methods, increasing use of telephone monitoring, increasing patient and family education and self-care, and cutting services to patients. Directors expressed concerns about staff stress related to the changes and the need to increase productivity without increasing staff. However, directors also believed the agency’s position would improve under the PPS. Additional research is needed to determine whether increased staff stress, work demands, and fewer resources for patients will affect the quality of care delivered and, thus, patient outcomes under the PPS.

Abstract

This article analyzes changes in Medicare home health staffing and service delivery patterns across three different reimbursement methods: cost based (1996), interim payment system (IPS) (1999), and the prospective payment system (PPS) (2002). This study combined secondary analysis of existing data (Provider of Services File and Statistical Supplement) with qualitative interviews of 22 home health agency directors to understand agency responses to policy changes created under the Balanced Budget Act of 1997. Cuts in staff and visits were greater under the IPS than they were under the PPS. Agencies cut staff and visits more dramatically for nonskilled services across both time periods. As a proportion of total services and visits, nursing and therapy services increased the most. Directors used various strategies to sustain the agency financially during these dramatic cuts in reimbursements, including eliminating staff, shifting staff roles, training staff on reimbursement methods, increasing use of telephone monitoring, increasing patient and family education and self-care, and cutting services to patients. Directors expressed concerns about staff stress related to the changes and the need to increase productivity without increasing staff. However, directors also believed the agency’s position would improve under the PPS. Additional research is needed to determine whether increased staff stress, work demands, and fewer resources for patients will affect the quality of care delivered and, thus, patient outcomes under the PPS.

This article uses several data sources to explicate the influence of policy changes on access and service delivery patterns in the Medicare home health care benefit. Focusing on the changes made to the Balanced Budget Act (BBA) of 1997, we demonstrate the direct effects of policy on nursing and allied health care staff, as well as on patients. In particular, we highlight how agencies respond to policy changes by modifying their own practices and procedures to sustain the organization.

The BBA of 1997 altered home health reimbursements for beneficiaries in the fee-for-service Medicare program via two phases. The first phase, referred to as the interim payment system (IPS), was implemented between 1997 and 2000. The two main goals behind the creation of the IPS were that it would immediately begin to limit expenditures within the home health program by controlling volume per person, and that it would allow time for the Centers for Medicare & Medicaid Services (CMS) to develop the prospective payment system (PPS) on the basis of results from a national demonstration on prospective payment methods.

The IPS established new reimbursement limits for feefor-service beneficiaries served by Medicare-certified home health agencies. Under the IPS, these agencies would be reimbursed based on “the lowest of: (1) actual costs; (2) reduced per visit limits; or (3) a blended, agency-specific per beneficiary annual limit” (BBA, 1997, pp. 87–88). The per-beneficiary cap, the lowest limit for most agencies, was calculated using a blended formula incorporating 75% of the agency’s average per-beneficiary payment and 25% of the regional average from 1993 (Berke, 1998; Forster, 1998). The per-beneficiary limit was increased slightly after 1998 for agencies that had been established before 1998 (Davitt & Choi, in press).

Several challenges were created by the IPS. Agency reimbursements were reduced to pre-1994 levels, representing a dramatic reduction. A new per-beneficiary limit introduced capitation in the traditional home care benefit for the first time, requiring drastic changes in how agencies delivered services. The per-beneficiary limit did not include casemix adjustment, which would allow agencies to serve more costly patients (Berke, 1998; Forster, 1998; Komisar & Feder, 1997; Lewin Group, 1998; Lin et al., 2005; McCall, Petersons, Moore, & Korb, 2003). In addition, agencies were given their per-beneficiary limits in March 1998 (Leon, Davitt, & Marainen, 2002) and thus operated for several months without clear guidelines on spending limits for each patient. Furthermore, because the per-beneficiary limits were based on a blended agency/region formula, agencies with higher reimbursements in 1994 received a higher reimbursement under the IPS (Lewin Group, 1998).

The base rate for the second phase of reimbursement changes, the PPS, was established using expenditure figures from the drastically reduced IPS. Therefore, concerns about the effect on patients, agencies, and staff continued under the PPS. The PPS, implemented in October 2000, continued the prospective payment arrangement but added case-mix adjustment. Under the PPS, agencies receive a fixed payment for a 60-day episode of care for each patient on the basis of their acuity. Acuity, measured via 80 home health resource groups, is established through a comprehensive assessment (Outcome and Assessment Information Set [OASIS]) conducted by the home care agency, which measures a patient’s clinical severity, functional status, and service needs. Therefore, agencies are paid based on the expected service needs for different categories of patients, rather than on actual cost to deliver the service (pre-BBA) or on an arbitrarily derived, per-beneficiary limit (IPS). However, under the PPS, the agency shoulders the financial risk of serving the patient.

Literature Review

Use of the home health care benefit decreased considerably after implementation of the IPS (Davitt, 2003; McCall, Korb, Petersons, & Moore, 2002; Spector, Cohen, & Pesis-Katz, 2004). One study found that overall expenditures for home health services decreased by 52%, the number of users per 1,000 enrollees declined by 21%, and visits per user dropped by 41% (McCall, Komisar, Petersons, & Moore, 2001). In addition, the percentage of all users receiving more than 150 visits decreased significantly (Davitt, 2003; Komisar, 2002). Other studies found similar decreases in use (Davitt & Marcus, in press; Liu, Long, & Dowling, 2003; McKnight, 2006). Researchers also found decreases in length of stay between 1996 and 1998 (Han & Remsburg, 2005; Murkofsky, Phillips, McCarthy, Davis, & Hamel, 2003). In fact, utilization decreased much more than was originally projected by the Congressional Budget Office (Leon et al., 2002).

Some of these studies also demonstrated changes in patient mix. McCall et al. (2003) found higher than average decreases in home health care use and the likelihood of any use during the IPS period for beneficiaries older than age 85, as well as a greater decrease in visits for patients older than 85 with diabetes, heart failure, cerebrovascular disease, and skin ulcers. Liu et al. (2003) found that users who did not have a hospital stay experienced a greater decrease in the likelihood of receiving any home health care and in the number of visits received between 1996 and 1999, compared with those with a hospital stay. While controlling for some predisposing and enabling characteristics, they also showed that users with greater functional impairment experienced a greater decrease in visits than did those with fewer impairments (Liu et al., 2003). Similarly, Davitt and Marcus (in press) found that users with poorer health and greater functional and cognitive impairments received fewer home care visits in 1998 than in 1996. McKnight (2006) found reductions in use of the benefit, especially for beneficiaries who were less healthy. All of these studies focused on use during the IPS period compared with use pre-BBA. In a study of home health use during the IPS, Adams, Michel, DeFrates, and Corbett (2001) found that rural patients had much poorer health than did their urban counterparts, which increased direct care time from home health staff.

Initial studies surveying hospital discharge planners and home care agency staff found changes in service delivery procedures after implementation of the IPS. Agencies altered their admissions practices to limit the number of high-cost patients admitted, provided fewer services to individual patients than they previously had, and established stricter discharge procedures, especially for patients perceived to be high cost (Markham-Smith, Maloy, & Hawkins, 1999; Medicare Payment Advisory Commission [MedPAC], 1999). Agencies also eliminated special staff positions and units (Markham-Smith et al., 1999) and reported providing less home health aid visits and more skilled nursing visits (McCall et al., 2001; MedPAC, 1999).

More recent studies using Medicare claims data found similar changes in service mix, with noticeable decreases in aide and medical social worker visits (Komisar, 2002; McCall et al., 2003). Other studies showed a slight decline in the likelihood of home health use after the PPS was implemented (Murtaugh, McCall, Moore, & Meadow, 2003), but this was much less than under the IPS (FitzGerald et al., 2006).

New Contribution

This article examines the use of the benefit and specific services in a policy context. Policy incentives influence agency reactions, which can alter the intended impact of the program. On the basis of early studies of agency responses to the BBA changes (Markham-Smith et al., 1999; MedPAC, 1999), it became clear to us that agency directors and staff were altering their service delivery patterns to manage the cuts in reimbursement and to keep the agency in operation. This article presents findings from a larger national study (Davitt, 2003) on those service delivery alterations by highlighting changes in the mix of services provided, including nursing, home health aid, medical social work, and therapy between 1996 and 2002. In particular, we expand on earlier work by comparing service delivery changes across three different reimbursement methods. In addition, we present agency directors’ reactions to the policy and their adjustments to the benefit, demonstrating its effects on patients, services, and staff. Using existing data, we describe the broader picture of changes in the home health benefit. The qualitative interview data allow us to understand, in depth, how the agencies responded and the specific strategies used. No other studies have combined an analysis of broad patterns in service mix with in-depth interviews on the specific strategies used by agencies for all three reimbursement methods.

Method

We combined a secondary analysis of Medicare Provider of Services (POS) and CMS statistical supplement data for 1996, 1999, and 2002, with qualitative interviews of home health agency directors in Pennsylvania. POS data focus on staffing patterns, and the CMS statistical supplement data show the changes in service delivery between 1996 and 2002.

Existing Data

Collected by CMS, POS data offer detailed information on all kinds of health care providers, their characteristics, and staffing patterns. The POS is a cumulative file that consists of all agencies ever certified by Medicare to provide health services. For this analysis, our sample consists of only active home health agencies for each respective year (1996: N = 9,917; 1999: N = 7,924; 2002: N = 7,011).).

In the larger study, Davitt (2003) assumed that use of the home care benefit in general would dramatically decrease after the IPS, as found by several researchers (Davitt, 2003; Liu et al., 2003; McCall et al., 2003; McKnight, 2006). However, it was more difficult to predict the kind of services agencies would cut the most. It was hypothesized that nonskilled services (i.e., home health aid, medical social work) would be more likely to be cut than would skilled services (i.e., nursing, therapy) because of the incentives in the policy. In this article, we present the POS data to demonstrate the effects of these changes on specific kinds of services in terms of the number and proportion of staff employed by the agency before and after these major policy changes. The statistical supplement data provide information on the number of visits in each kind of visit, except medical social work. These data were used to describe changes in staff and visits from before the BBA (1996) through the IPS (1999) and the PPS (2002). Because these data files include all active agencies in each year, inferential statistics are not reported.

Qualitative Interviews

The secondary data analysis was supplemented by key informant interviews with representatives from home health agencies in Pennsylvania. We were concerned the secondary data would not allow for a depth of understanding regarding actual everyday practice and service delivery. In particular, the existing data offered no information on agencies’ rationale for altering service delivery patterns, agency level challenges, and strategies to overcome those challenges. A semistructured interview guide was used, allowing respondents to describe what was happening in their agency in these various areas. This permitted the details of practice to arise from the data, rather than fitting researcher-defined categories for response.

All interviews were conducted with agency directors either by telephone or in person. In one or two cases, additional staff participated with the director. It was important to speak to the administrators because they would have a better understanding of the myriad areas of concern and the changes made by the agency. Responses were recorded with computer assistance.

Respondents were identified using a theoretical sampling framework to ensure all agency types were included in the interviews. The sampling frame consisted of the Pennsylvania Department of Health’s list of licensed and Medicarecertified home health agencies in the state. Pennsylvania was chosen because of its overall fiscally conservative approach to home health care use prior to the BBA. In addition, we had ample contacts throughout the state to ensure an adequate response rate. Pennsylvania experienced patterns of service changes similar to those of the rest of the country during this time frame. For example, between 1997 and 1999, Medicare home health revenues dropped by 43% nationally, and the number of beneficiaries served declined by 24% (CMS, 1999). Across Pennsylvania, revenues dropped by 45%, and beneficiaries declined by 21%, indicating similar service use changes as the rest of the country. However, fewer agencies in Pennsylvania (3%) closed during this period than did nationally (36%) (Leon et al., 2002).

An initial quota matrix was established to guide the selection of agencies from the list. Agencies were selected based on the following quota categories:

  • Agency tax status (e.g., not for profit/public, for profit).
  • Agency auspice (e.g., freestanding, hospital based).
  • Region of the state (northeast, southeast, north central/northwest, south central, southwest).
  • Geographical location (i.e., urban, rural, mixed).

From this list, we sampled 30 agencies that met the matrix criteria. A total of 22 agencies completed the interview process, for a 73% response rate (Davitt, 2003).

Each sampled agency was contacted by telephone to identify the appropriate staff member to be interviewed and to request participation in the study. All participants were asked to review and sign a consent form that explained the purpose and voluntary nature of the study. This study conformed to all human participant protocols as approved by the institutional review board.

The qualitative component of this study had two main limitations. First, due to resource constraints, the qualitative component was limited to agencies in the state of Pennsylvania. It is possible that the information gathered in these interviews may not reflect patterns experienced in other states. However, Pennsylvania did experience patterns of service changes similar to those of the rest of the country during the IPS. Second, only representatives of home health care agencies were interviewed; therefore, somewhat biased opinions about these policy changes may have been generated. However, the goal of this study’s qualitative component was to enhance the depth of understanding regarding the concerns of providers and how they responded to the policy. This allowed us to be sensitive to the practices of home health agencies post-BBA and to investigate provider practices that could not be understood from the existing data.

The qualitative analysis of interview data used a grounded theory framework (Glaser & Strauss, 1967). This framework relies on a constant comparative method whereby data is coded and analyzed simultaneously. Throughout this process, we constantly compared interview results, paying special attention to the similarities and differences in responses.

Quantitative Results

Policy in Context: Results from the Existing Data

The POS data on the average number of staff in home health agencies (Table 1) suggest that changes were more dramatic during the IPS than during the PPS. Except for speech therapy and occupational therapy, all other kinds of staff declined between 1996 and 1999. The most dramatic declines were for home health aides, physical therapists, medical social workers, and licensed practical nurses (LPNs). As a proportion of staff, greater decreases occurred for home health aides, medical social workers, and LPNs (Table 2). However, skilled services, especially therapy, increased as a proportion.

Average Number of Staff and Rate of Change by Year

Table 1: Average Number of Staff and Rate of Change by Year

Proportion of Staff and Rate of Change by Year

Table 2: Proportion of Staff and Rate of Change by Year

Agencies rebounded under the PPS in terms of staffing. Except for home health aid and medical social work, all other services experienced an increase in the number of staff, on average, between 1999 and 2002. The largest increases were in the therapy services (e.g., occupational therapy, physical therapy) and RNs. Therapists increased dramatically as a proportion of all staff, on average, between 1999 and 2002, whereas nonskilled staff continued to decrease as a proportion of all staff.

Agencies also altered the number of visits provided for specific services, some more dramatically than others (Tables 3 and 4). Between 1996 and 1999, the total number of home health aide visits decreased by 69%. In 1996, home health aide visits comprised almost half of total visits, but in 1999, they had decreased to 34.3%. The total number of nursing visits decreased by 49%. However, as a proportion of total visits, nursing visits actually increased from 41.1% to 48.4%. Other services, including medical social work, decreased by 56%. Therapy services also decreased between 1996 and 1999 but not as dramatically as did the other services. In fact, all therapy services increased as a proportion of total services.

Visit Totals by Staff and Rate of Change by Year

Table 3: Visit Totals by Staff and Rate of Change by Year

Proportion of Total Visits by Staff and Rate of Change by Year

Table 4: Proportion of Total Visits by Staff and Rate of Change by Year

Visits per home health care user were analyzed, and Table 5 shows that decreases occurred between 1996 and 1999 for all kinds of services. The biggest decrease was for home health aid (48%), followed by other services (33%) and nursing (31%).

Visits Per User by Staff and Rate of Change by Year

Table 5: Visits Per User by Staff and Rate of Change by Year

Under the PPS, further decreases in total visits occurred for home health aid, other services, nursing, and speech therapy visits. Both occupational therapy and physical therapy increased in visits. All therapy and nursing services increased as a proportion of total visits during the PPS. Only nonskilled visits decreased as a proportion during the PPS. Continued decreases in visits per user occurred for home health aid, nursing, speech therapy, and occupational therapy during the PPS.

Qualitative Results

In the qualitative interviews, agency directors had a chance to express their concerns regarding BBA changes and to describe their responses. The changes affected the agencies’ financial stability, service delivery, and staffing patterns. No part of the agencies seemed untouched by the reimbursement cuts, including non-Medicare-funded programs. The changes took their toll on staff and patients as well.

The Policy Response: Lessons from the Agency Level

Agency directors indicated that after the BBA, they were preoccupied with fiscal sustainability. Some agencies relied on endowments to support unreimbursed care. Others operated in debt or closed. The following quotations from directors demonstrate this concern with fiscal issues.

  • As a small company, we had about 30 [Medicare] clients, mostly wound care and diabetics needing a lot of daily care; most wiped out their [per-beneficiary] limit within 1 month, but we still had to provide services until they healed; we were $70,000 in debt.
  • Even if we held visit costs the same, we had to cut somewhere because of the loss of inflation increases. You cut cost by increasing productivity, without increasing number of staff.

Strategies to maintain fiscal stability included reducing the number of visits to each patient, increasing patient education and caregiver support, using clinical pathways and multidisciplinary teams, eliminating services, eliminating staff, and shifting staff duties. A majority of administrators reported reducing staff either through layoffs or attrition (n = 16). Agencies more often reported eliminating administrative staff (n = 12) than clinical staff (n = 6). Respondents stated:

  • Did not know until 9 months into the first year if what we were doing was correct. At first we undercut visits, wanted to go from 45 to 33 per patient. But we actually went to 19.
  • We did not hire to replace those who did leave voluntarily. We laid off one third [of] nurses, half of aides. We only have RNs now, no longer have LPNs; needed nurses who could do everything and could switch back and forth between roles.
  • Therapists tend to assess patients as more severe under activities of daily living than the nurses [do]; nurses use patient self-report where therapists observe behavior, therefore nurses will need longer to assess, and we will have therapists train nurses on this.

The directors also reported reducing staff salaries or benefits (n = 10); this was done either as direct reductions or by not providing cost-of-living or merit increases. Agencies also reported cutting mileage reimbursements (n = 9) and reducing or eliminating staff education and inservice programs (n = 8). They also reported retraining staff to be more aware of the reimbursement issues. One director said:

We never made them [staff] stewards of the money before, this is a change; we need them to think financially. Altruistic nurses who just wanted to care find it a conflict, this stresses them. You have to see the financial stability of the agency as an important goal as well.

When asked to discuss how these changes might be affecting their staff, the directors showed genuine concern for the staff and their situations. Almost half of the directors reported increased stress and anxiety among staff due to implementation of the IPS. Such stress stemmed from fear about losing one’s job, in some agencies, to concern for clients and getting clients the help they need within a system with such severe cutbacks. Staff also worried about the stress being placed on informal supports and what might happen to their clients after discharge. One director said:

It has affected staff morale—they lost raises and increases in mileage reimbursement and are being asked to do a lot more for less. There is an overall impending doom feeling. It has put a damper on the atmosphere where it used to be upbeat.

There was also tremendous pressure on directors and nurse managers. The directors described:

  • I cried a lot and haven’t slept. [It is] The worst thing that has ever happened, and I have 20 years with this agency. It drained us emotionally and physically.
  • There’s an extraordinary amount of care that CMS doesn’t address. You can’t expect someone to just walk in the house and do a blood pressure or dressing change. There’s [many] problems in that household and, therefore, it is very difficult for nurses.

One director seemed to capture the sentiment of most of the interviewees in her statement:

[In doing this interview] It will be so good to tell somebody just what I think of this whole mess. I have been a nurse for 30 years, and this is the first time that I wish I had another skill. I feel sorry for future generations of nurses; I used to be quite proud of telling nurses you provide good care, and we will worry later about payment. I can’t do this anymore; we are reducing the person to a number.

Reduction in Access to Nursing Services: Effects on Vulnerable Patients

Most of the directors would not admit they were refusing admission to patients with particular conditions. Some were willing to indicate that they based admission decisions on the same criteria as before IPS; that is, whether they have the necessary resources (e.g., adequate numbers of trained staff) to take on a patient’s care. But when asked to discuss their concerns about patient access to care within the overall system, rather than in their specific agency, most directors were willing to openly discuss their concerns related to access. One director stated:

In this agency, there was no change in access, but I have seen it with other agencies; patients are getting less visits, pushed out sooner, get teaching in quicker. It’s more of a crunch.

Key patient characteristics for which directors had concerns regarding access to care included the need for care of chronic conditions (n = 15), living in a rural area (n = 11), not having a caregiver (n = 8), having high acuity or comorbidity (n = 6), and needing some kind of high-tech or complex care (n = 6). The directors said:

  • Patients who are rural depend on us more; they can’t get to the doctor, there’s no transportation system. So they do without, or the home care agency does what it can to make up for it.
  • Families have changed, living at great distance, with more and more elderly living alone, and if they can’t pay privately, or aren’t indigent, they won’t be eligible for home care. Not sure where they will go, or how they will manage.

One administrator described the following example of a patient the agency could no longer serve under the new rules:

Bilateral, above-knee amputee with Parkinson’s living alone and can’t see to fill his syringes [insulin], what would you suggest for this patient? He can inject if they are filled but cannot fill them himself. So should a nursing home be a better choice for him?

The directors also indicated the diagnostic characteristics they believed might negatively affect access to care, including wound care needs (n = 19), congestive heart failure (CHF) (n = 7), diabetes/daily insulin needs (n = 6), chronic respiratory needs (n= 6 ), and post-stroke care needs (n = 3). The directors said:

  • Where are those patients going to go? Who is going to take care of wound care, insulin, or CHF, these chronic care patients, revolving-door patients?
  • The very ill will get services, chronically ill are cut out of the loop; those who cannot improve and are most in need of help will have the least access.

The agency directors were clearly concerned with patient care and, in many cases, were worried about the capability of family members to provide the necessary assistance and hands-on medical care that was required. They were also concerned and expressed their fears that many patients would not be served or would be underserved, resulting in further deterioration in health status and potentially in more costly care through inpatient or emergency department services. The comments of one director summed up the overall reaction to these changes:

Congress needs to understand what nurses do. You can’t put a system in place for 30 years and now that it is going bankrupt then accuse it of fraud, etc. and change it overnight. They went too far.

Reactions to the Prospective Payment System

The directors were equally concerned about how patients with those same conditions would fare under the PPS. Patients of concern included those needing daily or more frequent care (e.g., wound care, daily insulin), those with chronic care needs (e.g., chronic obstructive pulmonary disease [COPD], CHF, post-stroke), those requiring costly high-tech care, those in rural areas, and those lacking caregivers. The directors stated:

  • Patients in rural areas who have lost access to care will continue [without care] under PPS. A good amount of time will have to go by before agencies will expand back into these areas, putting great strain and burden on families who are already burdened.
  • Chronic care patients like [those with] COPD [and] CHF are at risk under PPS because it is difficult to document ongoing need to see these patients. They are at risk for a crisis event and hospitalization. If they are rural, it costs even more.

Most directors were optimistic that they would do better under PPS compared with IPS. For some agencies, this might have been too little, too late, but many directors considered the PPS a savior, allowing their agencies to stay in business. More important, many believed the PPS was a fairer system for all agencies because it was based on patients’ severity of illness, rather than on a preset, agency-specific limit. One director said:

The good news—our revenue is higher under PPS than IPS. We were also able to relax some cuts and provide increases in salaries and benefits, but will have to slow down on some of these due to cash flow issues.

The directors believed the PPS would continue to require greater emphasis on patient and caregiver education and training. However, they also noted they had more adequate time to provide such training under the PPS than they did under the IPS. The following quotation summed up directors’ thinking on this issue:

Under PPS, we can extend time for teaching, thereby gradually weaning patients over a longer period and giving them greater time to get comfortable with the procedure.

The directors also speculated on the kinds of strategies agencies might use to maintain their financial viability under the PPS. More directors were planning to implement tools such as clinical pathways, telemedicine, and wound formularies to make their agencies more efficient. For example, increasing the number of telephone contacts with patients would save time, thus enhancing productivity, as well as reducing costs for travel. The agencies might also increase the total number of clients, rather than the number of episodes per client, to increase their revenue base. Their marketing plans would continue to focus on attracting clients who are less ill, because most of the directors did not expect the outlier payment would be adequate to cover the additional costs of caring for patients who are very ill. The directors mentioned the need for more intensive staff training to enhance their understanding of the new reimbursement system and the pivotal nature of adequate documentation on the OASIS assessment. The directors said:

  • Telephone contacts are likely to be an important strategy for agencies. We will have to wait and see if too much is done by phone.
  • Under the PPS, the key is volume, and that keeps the mix stable. Smaller organizations have less ability to do that; closures under PPS will most likely be smaller agencies.

Discussion

Incentives in the policy changes encouraged certain gaming patterns across the variety of agencies. For example, agencies shifted the service mix toward skilled services. This pattern was dramatic under the IPS but continued, albeit less so, under the PPS. Therapy services fared best under these policy changes, suggesting the additional payment added to the Home Health Resource Group under the need for therapy has increased agency provision of these services. In addition, under both the IPS and PPS, agencies cut more visits per user than admissions. This approach clearly matched the incentive to control volume per patient. In fact, under the PPS, the incentives encourage agencies to admit more patients but provide less visits to each of them. However, this does not necessarily reduce costs in the overall system, nor does it encourage reduction in inappropriate use of the benefit. Rather, it generated incentives for agencies to cut the amount of service provided to the most potentially high-cost patients, regardless of the legitimacy of their claim and their need for home health care. This trend was certainly reflected in the directors’ concerns related to access to care for higher cost patients.

These data demonstrate the interrelationship between social policy, agency practices, and access to care. Agencies reacted to the policy change by changing a variety of service delivery procedures, staffing patterns, and administrative strategies to ensure their financial security and sustainability during the IPS and PPS. Agencies revised service delivery patterns to be more efficient and productive. In some cases, this affected staff financially through layoffs or pay cuts. Other staff were affected emotionally through increased job-related stress, higher demands on productivity, and less support for job performance.

These data also show that certain patients (e.g., those who are healthier, those who do not require care for chronic conditions, those who do not live in rural areas) were better positioned to weather policy changes than were others. Agency directors identified patients with high-cost needs, such as those living in rural areas, those needing multiple daily visits, and those with complex care needs or without adequate informal support, as the most affected (Davitt, 2003; McCall, 2003). Several studies have confirmed the directors’ concerns related to access for vulnerable patient subgroups (Davitt & Marcus, in press; Liu et al., 2003; McCall, 2003; McKnight, 2006; Peng, Navaie-Waliser, & Feldman, 2003). Therefore, policy changes, by encouraging certain practices on the part of agencies, indirectly affected access to care, and more so for certain patients.

The effects of such policy changes on staff has not been adequately discussed in the literature. The increased stress and push for higher productivity, along with decreasing staff levels, may negatively affect the quality of services being provided and the ability to attract qualified personnel to home health care, particularly in nursing. In fact, one agency director noted the difficulty in their region of the state in finding quality staff and how policy changes would only exacerbate the problem. Similarly, stressed staff are more prone to burnout, which may increase staff turnover and, thus, staff training costs for agencies. Clearly, further research is needed to assess whether such cuts have affected staff morale, turnover rates under the PPS, and indirectly, quality of care.

It is difficult to state whether service delivery changes or access patterns result in lower quality of care and thus poorer health outcomes. Initial studies examining outcomes of home health care post-BBA focus mainly on the IPS period, and the results are mixed. For example, McCall et al. (2002) found that mortality and skilled nursing facility use increased significantly for home health users post-BBA, whereas hospitalizations decreased. Another study showed overall lower hospitalization or emergency care when comparing the PPS to IPS periods (Schlenker, Powell, & Goodrich, 2005). This study also found less improvement on wounds, incontinence, and cognitive outcomes. However, McKnight (2006) found no increases in institutional or other medical care nor decreases in health status outcomes.

Conclusion

The effects of policy on the care delivery system and patients requires continued study. Research is needed on the ongoing effects of prospective payment on staffing patterns, staff stress and morale, access to care, quality of care, and patient outcomes, especially for vulnerable patient subgroups.

The results of this study show additional effects of the BBA changes that were not uncovered in earlier studies. This information is critical to understanding the unintended effects of policy changes. Using a significant federal policy change as a laboratory, we highlight potential problems that might occur in future policy practice in Medicare or other large national programs by describing the responses of providers within the system. Finally, this study informs the policy design process in general, providing a prime example of how policy design and implementation are equally important in generating positive results.

References

  • Adams, CE, Michel, Y, DeFrates, D & Corbett, CF2001. Effect of locale on health status and direct care time of rural versus urban home health patients. Journal of Nursing Administration, 31, 244–251. doi:10.1097/00005110-200105000-00004 [CrossRef]
  • Balanced Budget Act of 1997, Pub. L. No . 105–33, 42 U.S.C. § 1395 (1997).
  • Berke, D1998. The Balanced Budget Act of 1997—What it means for home care providers and beneficiaries. Journal of Long Term Home Health Care, 73, 2–9.
  • Centers for Medicare & Medicaid Services. (1996, 1999, 2002). Non identifiable data files: Provider of Services file. Retrieved January 2007, from http://www.cms.hhs.gov/NonIdentifiableDataFiles/04_ProviderofServicesFile.asp#TopOfPage
  • Centers for Medicare & Medicaid Services. (1998, 2001, 2004). Medicare & Medicaid Statistical supplement. Retrieved January 2007, from http://www.cms.hhs.gov/MedicareMedicaidStatSupp/
  • Centers for Medicare & Medicaid Services. 1999. Medicare bulletin 422. Washington, DC: U.S. Department of Health and Human Services.
  • Davitt, JK2003. Access to Medicare home health care in the wake of the Balanced Budget Act. Final report. Bryn Mawr, PA: Bryn Mawr College.
  • Davitt, J & Choi, S (in press). Tracing the history of Medicare home health care: The impact of policy on benefit use. Journal of Sociology and Social Welfare
  • Davitt, J & Marcus, S (in press). The differential impact of Medicare home health care policy on impaired beneficiaries. Journal of Policy Practice
  • FitzGerald, JD, Mangione, CM, Boscardin, J, Kominski, G, Hahn, B & Ettner, SL2006. Impact of changes in Medicare home health care reimbursement on month-to-month home health utilization between 1996 and 2001 for a national sample of patients undergoing orthopedic procedures. Medical Care, 44, 870–878. doi:10.1097/01.mlr.0000220687.92922.64 [CrossRef]
  • Forster, TM1998. Home care, the Balanced Budget Act of 1997, and IPS. Caring Magazine, 72, 8–13.
  • Glaser, BG & Strauss, AL1967. The discovery of grounded theory: Strategies for qualitative research. Hawthorne, NY: Aldine de Gruyter.
  • Han, B & Remsburg, RE2005. Impact of the Medicare interim payment system on length of use in home care among patients with Medicare-only payment source. Home Health Care Services Quarterly, 244, 65–79. doi:10.1300/J027v24n04_04 [CrossRef]
  • Komisar, H & Feder, J1997. The Balanced Budget Act of 1997: Effects on Medicare’s home health benefit and beneficiaries who need long-term care. Washington, DC: Institute for Health Care Research and Policy, Georgetown University. Retrieved November 19, 2007, from The Commonwealth Fund Web site: http://www.commonwealthfund.org/publications/publications_show.htm?doc_id=221406
  • Komisar, HL2002. Rolling back Medicare home health. Health Care Financing Review, 242, 33–55.
  • Leon, J, Davitt, J & Marainen, J2002. A comparison of the effects of new Medicare reimbursement methodologies on home health. Horsham, PA: Polisher Research Institute.
  • Lewin Group. 1998. Implications of the Medicare home health care interim payment system of the Balanced Budget Act. Washington, DC: National Association for Home Care.
  • Lin, CJ, Davitt, JK, Meit, M, Schwartz, M, Marainen, J & Davis, L2005. The financial impact of interim and prospective payment systems on home health providers in rural communities. Journal of Health Care for the Poor and Underserved, 16, 576–587. doi:10.1353/hpu.2005.0056 [CrossRef]
  • Liu, K, Long, SK & Dowling, K2003. Medicare interim payment system’s impact on Medicare home health utilization. Health Care Financing Review, 251, 81–97.
  • Markham-Smith, B, Maloy, KA & Hawkins, DJ1999. An examination of Medicare home health services: A descriptive study of the effects of the Balanced Budget Act interim payment system on access to and quality of care. Washington, DC: Center for Health Services Research and Policy, George Washington University.
  • McCall, N, Komisar, HL, Petersons, A & Moore, S2001. Medicare home health before and after the BBA. Health Affairs (Project Hope), 20, 189–198. doi:10.1377/hlthaff.20.3.189 [CrossRef]
  • McCall, N, Korb, J, Petersons, A & Moore, S2002. Constraining Medicare home health reimbursement: What are the outcomes?Health Care Financing Review, 242, 57–76.
  • McCall, N, Petersons, A, Moore, S & Korb, J2003. Utilization of home health services before and after the Balanced Budget Act of 1997: What were the initial effects?Health Services Research, 381 Part 1, 85–106. doi:10.1111/1475-6773.00106 [CrossRef]
  • McKnight, R2006. Home care reimbursement, long-term care utilization, and health outcomes. Journal of Public Economics, 90, 293–323. doi:10.1016/j.jpubeco.2004.11.006 [CrossRef]
  • Medicare Payment Advisory Commission. 1999. Report to the Congress: Selected Medicare issues. Washington, DC: Author.
  • Murkofsky, RL, Phillips, RS, McCarthy, EP, Davis, RB & Hamel, MB2003. Length of stay in home care before and after the 1997 Balanced Budget Act. Journal of the American Medical Association, 289, 2841–2848. doi:10.1001/jama.289.21.2841 [CrossRef]
  • Murtaugh, CM, McCall, N, Moore, S & Meadow, A2003. Trends in Medicare home health care use: 1997–2001. Health Affairs (Project Hope), 22, 146–156. doi:10.1377/hlthaff.22.5.146 [CrossRef]
  • Peng, TR, Navaie-Waliser, M & Feldman, PH2003. Social support, home health service use, and outcomes among four racialethnic groups. The Gerontologist, 43, 503–513.
  • Schlenker, RE, Powell, MC & Goodrich, GK2005. Initial home health outcomes under prospective payment. Health Services Research, 401, 177–193. doi:10.1111/j.1475-6773.2005.00348.x [CrossRef]
  • Spector, WD, Cohen, JW & Pesis-Katz, I2004. Home care before and after the Balanced Budget Act of 1997: Shifts in financing and services. The Gerontologist, 44, 39–47.

Average Number of Staff and Rate of Change by Year

Average Number
Rate of Change (%)
Staff 1996 (N = 9,917) 1999 (N = 7,924) 2002 (N = 7,011) 1996–1999 1999–2002 1996–2002
Home health aides 14.37 10.53 8.27 −26.72 −21.46 −42.45
Medical social workers 0.75 0.64 0.64 −14.67 0 −14.67
Licensed practical nurses (LPNs) 3.68 3.19 3.53 −13.32 +10.66 −4.08
Registered nurses (RNs) 14.72 12.84 15.03 −12.77 +17.06 +2.11
LPNs and RNs 18.4 16.03 18.56 −12.88 +15.78 +0.87
Physical therapists 2.36 1.76 2.09 −25.42 +18.75 −11.44
Speech therapists 0.21 0.23 0.26 +9.52 +13.04 +23.81
Occupational therapists 0.36 0.45 0.56 +25 +24.44 +55.56
All staff 43.33 35.12 36.45 −18.95 +3.79 −15.88

Proportion of Staff and Rate of Change by Year

Proportion (%)
Rate of Change (%)
Staff 1996 (N = 9,917) 1999 (N = 7,924) 2002 (N = 7,011) 1996–1999 1999–2002 1996–2002
Home health aides 32.16 29.64 24.23 −7.84 −18.25 −24.66
Medical social workers 2.27 2.24 2.2 −1.32 −1.79 −3.08
Licensed practical nurses (LPNs) 8.62 8.6 10.25 −0.23 +19.19 +18.91
Registered nurses (RNs) 40.86 41.08 41.72 +0.54 +1.56 +2.10
LPNs and RNs 49.48 49.68 51.97 +0.40 +4.61 +5.03
Physical therapists 3.26 4.16 6.5 +27.61 +56.25 +99.39
Speech therapists 0.73 0.79 0.96 +8.22 +21.52 +31.51
Occupational therapists 0.99 1.23 1.85 +24.24 +50.41 +86.87

Visit Totals by Staff and Rate of Change by Year

Total Visits
Rate of Change (%)
Staff 1996 1999 2002 1996–1999 1999–2002 1996–2002
Home health aides 129,502,000 38,949,000 18,479,000 −69 −52 −86
Nurses (LPNs and RNs) 108,839,000 54,914,000 39,092,000 −49 −29 −64
Physical therapists 19,320,000 14,865,000 16,384,000 −23 +10 −12
Speech therapists 1,292,000 792,000 594,000 −39 −25 −54
Occupational therapists 3,142,000 2,731,000 2,921,000 −13 +7 −7
Othera 2,704,000 1,188,000 723,000 −56 −39 −73

Proportion of Total Visits by Staff and Rate of Change by Year

Proportion of Total Visits (%)
Rate of Change (%)
Staff 1996 1999 2002 1996–1999 1999–2002 1996–2002
Home health aides 48.9 34.3 23.6 −30 −31 −52
Nurses (LPNs and RNs) 41.1 48.4 50 +18 +3.3 +22
Physical therapists 7.3 13.1 21 +79 +60 +188
Speech therapists 0.5 0.7 0.8 +40 +14 +60
Occupational therapists 1.2 2.4 3.7 +100 +54 +208
Othera 1 1 0.9 0 −10 −10

Visits Per User by Staff and Rate of Change by Year

Visits Per User
Rate of Change (%)
Staff 1996 1999 2002 1996–1999 1999–2002 1996–2002
Home health aides 73 38 25 −48 −34 −66
Nurses 32 22 17 −31 −23 −47
Physical therapists 14 11 11 −21 0 −21
Speech therapists 11 8 7 −27 −12 −36
Occupational therapists 8 7 6 −12 −14 −25
Othera 3 2 2 −33 0 −33
Authors

Dr. Davitt is Assistant Professor and Hartford Geriatric Social Work Faculty Scholar, School of Social Policy and Practice, University of Pennsylvania, Philadelphia, Pennsylvania, and Dr. Choi is Assistant Professor, Department of Social Work, College of Community and Public Affairs, Binghamton University, Binghamton, New York.

The authors thank the following for their support of this research: The AARP Andrus Foundation, The Centers for Medicare and Medicaid Services, and The John A. Hartford Foundation.

Address correspondence to Joan K. Davitt, PhD, MSS, MLSP, Assistant Professor and Hartford Geriatric Social Work Faculty Scholar, School of Social Policy and Practice, University of Pennsylvania, 3701 Locust Walk, Philadelphia, PA 19104; e-mail: jdavitt@sp2.upenn.edu.

10.3928/19404921-20080101-03

Sign up to receive

Journal E-contents