Effects of Policy, Reimbursement, and Regulation on Home Health Care
Peter A. Boling
Human factors in home health care involve many stakeholders: those receiving care, families and friends, paid workers of various kinds and skill levels, employers, and communities in a broad sense, because tens of millions of individuals rely on home care each day. Despite excellent workers and advanced technology, people experience inconsistent quality, fragmented care, and poorly aligned payments. Home care remains a stepchild while health policy centers on hospitals and other facilities, physicians, pharmacies, and insurers. Despite a holistic culture, a large, dedicated workforce, and care valued by recipients, home care remains poorly understood, sometimes mistrusted, and abstract for many decision makers, and as currently organized it has limited proven ability to control overall costs. This all can change.
THE LANDSCAPE OF HOME HEALTH CARE
Societal Context, Costs, and Care Silos
There is a growing sense of crisis. Medicare cost about $500 billion in 2009 and the Medicare Hospital Insurance fund balance will be zeroed in seven years. Without a plan to “bend the cost curve,” corrective options include reduced provider payments, reduced coverage, higher copayments, or taxes to raise revenues equal to about 4 percent of gross domestic product 10 years hence. The larger frame is over $2 trillion annually spent on health care, or one-sixth of the economy, impacting the nation’s ability to compete internationally. Nearing the brink, people debate about the num-
bers of uninsured (45 million) or underinsured (80 million), public and private insurance options, regional cost variation, poor public health metrics, and financing for potential solutions while spending 30 percent more per capita than any other industrialized nation on health care.
At $57 billion in 2007 including all payers, home care is only 3 percent of U.S. health care (National Association for Homecare & Hospice, 2009) (see Figure 12-1). And like health care generally, home care is organized into separately funded categories, called silos: home health agencies, hospices, medical equipment, home health aides, pharmacy managers, medical providers, and thousands of private bureaucracies. Two key referral sources, hospitals and nursing homes, are in fiscal and regulatory silos of their own. Thus, financing promotes discontinuous care. Rules that govern the care of people with heavy chronic illness burdens result in an uncoordinated, overly costly, and hazardous delivery system that is far from patient-centered and is known for its failings (Schoen et al., 2008) (see Figure 12-2).
Some financing models, including Medicare Part C, offer opportunities to break down silos using financial risk. However, only 11.4 million people, or about a quarter of Medicare beneficiaries, are in such payment models (Henry J. Kaiser Family Foundation, 2009, 2010). And there are
other concerns. Importantly, managed care enrollees are disproportionately healthy, and most high-cost seniors rely on standard Medicare benefits. Even in risk-bearing entities, integrated at-home care with a strong medical component is rare. Reasons for this lack of emphasis include the origins of managed care models in conventional settings and finance systems, competing corporate interests for the insurer, and lack of senior management familiarity with what advanced home care can provide.
However, portable technologies and experiences with caring for sick patients in the community are now lighting new paths in which fully developed and integrated interdisciplinary home care teams, which include physicians and other medical providers, work together with other providers of care and community resources to produce better care at lower cost.
Population Need for Home Care
Individual needs for community-based care vary. Although it is simplified, and acknowledging that some patients fall into more than one of these categories, for the purposes of framing, I divide the general population into subsets based on typical care needs (see Figure 12-3). Their needs vary by
group, as do the human factors impacts of policy, regulation, and available care options.
Group A contains healthy, predominantly younger persons who are functional and independent but might be considered home care users if one includes public health measures deployed at home to promote healthy living, such as reminder systems to lose weight or stop smoking and web-based self-help tools for episodic illness (flu, ankle sprain, rash).
Group B members are mobile and can readily visit physician offices but have chronic diseases, such as high blood pressure, diabetes, asthma, or sleep apnea. Home modification, diet, exercise, and monitoring are routinely used at home, and in-home care improves outcomes (Ogedegbe and Schoenthaler, 2006; Shelledy et al., 2009). Examples are blood pressure monitors (if accurate) (Akpolat et al., 2009), glucose monitors, peak flow meters, asthma action plans, machines for maintaining continuous positive airway pressure, air filters, and removal of mold-bearing items. Individuals use the Internet to send data to physicians from home or to receive medical advice. Broadly construed, they are home care users.
Group C members either have conditions from birth, like cerebral palsy or other chronic conditions—which are often neurological or mus-
culoskeletal and severely limit function but are not progressive—or disabling permanent injuries acquired in childhood or adult life. An example is paraplegia. With assistive technologies, many such individuals function independently, are infrequently ill, and are rarely hospitalized. Group C members use home care for support of activities of daily living (ADLs) and instrumental activities of daily living (IADLs), often in the form of medical devices plus personal assistance with bathing, dressing, and mobility. Group C members fall through cracks in insurance coverage or are harmed by bureaucracy used to control cost and fraud. For example, a 50-year-old quadriplegic man, free of wounds for decades, now has a deep, dangerous Stage 4 pressure wound that will take months to heal because of a 2-week administrative delay in replacing a damaged support surface.
Group D members are generally older, frail, and functionally impaired but medically stable. Many have dementia or inborn cognitive impairment that requires supervision and standby help to prevent injury or to remain in the community. Also prevalent are advanced stages of disabling conditions, like arthritis or prior stroke, that mandate physical support each day from equipment and other persons. However, because these illnesses do not cause frequent acute medical crises, Group D individuals use limited skilled home care. Care plans do not change from month to month. Group D members have difficulty accessing office-based medical care because of immobility. However, in Group D money cannot be readily saved through case management, other than in a small subset whose eventual nursing home placement can be deferred by socially oriented caregiver support. Group D needs long-term home care. Third-party coverage is largely limited to those in poverty (Medicaid) or those who are affluent (long-term care insurance).
Group E members of all ages are in postacute care, recovering from an acute episode like major surgery, acute illness, or trauma. Return to independence is expected, and baseline health is fairly good. Some may receive intravenous therapy at home or at work for a curable infection. Technically advanced home care may be extensive, but its use is temporary. Current Medicare and private insurance payment designs best suit the episodic needs of Group E.
Group F includes most of the need for what typically comes to mind when thinking about home health care. Users vary in age from infants to very old adults, carry heavy chronic illness burden, often with end-stage organ system failure (heart, lung, liver, kidney, brain), have multiple concurrent conditions, and have significant persistent functional deficits, often three or more, and in many cases five or six, ADL deficits.
In Group F, infants might be premature with resultant home ventilator dependence or a condition requiring artificial feeding. Imagine the mother, caring for three other children and a household, trying to manage these daily responsibilities and also bring such a technology-dependent child to
the doctor’s office. Younger adults in Group F might have complications of childhood diabetes or cystic fibrosis.
Older adults, however, make up the majority of Group F. They have a mix of diseases, with no one predominant. This mix is shaped by conditions that are most prevalent: stroke, vascular disease, emphysema, complicated diabetes, and more. The variety and complexity of conditions challenge clinicians and require diverse skills. Illness acuity and stability fluctuate. Individuals rapidly get sick enough to be hospitalized. Group F members move frequently between health care silos and have multiple changes in condition that require revised care plans and active, longitudinal coordination among providers. Errors are made in as many as half of the transitions between settings. Most of the 2 to 3 million Group F members deteriorate with time and die, with an average life expectancy of 3 to 4 years.
Functional Dependency, Chronic Illness, and Service Use
Activity limitation due to chronic conditions occurs in 7 percent of children and 42 percent of people ages 75 and over. ADL limitations (walking, transferring, bathing, dressing, eating, or toileting) increase with age, and severe deficits (three or more ADLs) occur in 1.9 million community-dwelling U.S. adults (see Table 12-1) (Centers for Disease Control and Prevention, 2009).
Among community dwellers ages 65-74, the National Health Interview Survey revealed that 912,000 out of 17.4 million people (5 percent) cannot sit for 2 hours (are bedfast) (Centers for Disease Control and Prevention, 2009), and about 15 to 20 percent need help to walk. Such people often have difficulty leaving home for health care. These functional deficits are on the final common path, where the combination of severe illness and incapacity drive health care costs.
Sorting the Medicare population by cost, the top 5 percent used 43 percent of resources, with average costs of $63,000 in 2002 (Holtz-Eakin, 2005).
Some were catastrophic cases, some fully recovered, and some died. Those who lived 5 years (the majority) were intermittently high-cost users from month to month. In 2005, the top 10 percent of Medicare beneficiaries used 63 percent of resources, with average annual costs in the top 10 percent bracket of $44,000 (Potetz and Cubanski, 2009). A similar pattern is seen in younger (Medicaid) and employer-based insured groups.
Impaired physical function itself predicts high costs. In 2004, simply having three or more ADL deficits conveyed average costs of $32,000 per person (Medicare Beneficiary Survey) compared with $15,000 with one or two ADL deficits and $7,000 with no ADL deficits.
Along with function, the presence of multimorbid conditions drives cost. People with five or more chronic conditions use two-thirds of the resources, receive far more prescriptions, see a multitude of doctors, and have exponentially more hospital stays, including many that are preventable (Goldfield et al., 2008). Cost-containment strategies based on a single disease or on a nurse care coordinator who is not a core health care team member are unlikely to succeed, as most recently shown in the failed Medicare Health Support demonstration (Linden and Adler-Milstein, 2008) and the Coordinated Care Demonstration (Brown et al., 2008).
No published national analysis directly links ADL deficits, home health use, and overall costs at the individual level. However, because of the close link with hospital use, home care users probably do include most of the costliest beneficiaries. In Medicare Part A, for example, each year about 3 million people meet the homebound and skilled care criteria and receive skilled home health care at some point, totaling about 7 million discrete episodes each year.
Summing up, there are 3-4 million relatively immobile people in U.S. communities who span all ages, are sick with multiple concurrent illnesses, and have intermittently high health care costs. Many are not terminally ill (i.e., will not die in 6 months or less) and do not qualify for hospice; these people regularly fall through the cracks of the formal care system.
Finally, every discussion of home care must include unpaid family caregivers, who do most of the work. Caregiving falls to people who are not trained as clinicians yet are required to do technical clinical work along with basic daily care. Learned skills include intravenous therapy, tracheostomy and ventilator support, wound care, catheter changes, and medication oversight, work that in other care settings is restricted to licensed nurses. Most caregivers are women, their average duration exceeds 4 years, and they average 25 hours per week (Giovannetti et al., 2009). In this role, they lose time from work, lose employment, and develop health problems, with adverse economic consequences for worker and employer (Coe and Van Houten, 2009). Employers’ cost for full-time employees with caregiving roles is $17.1 billion ($2,441 per employee), and total cost to employers
for full-time, employed caregivers is $33.6 billion (Metlife, 2006). This is a significant part of the human factors equation. Chapter 7 of this volume discusses informal caregivers in detail.
FIVE SERIOUS PROBLEMS OF LONG-TERM HOME HEALTH CARE
Human ADL Support, Medicaid, and Long-Term Care Coverage
The lack of a consistent national policy on long-term care or of a systems approach to care at home is a major overarching problem for home care that daily affects paid providers, care recipients, and their families, who are trying to solve these problems.
In patient Groups C and D, most paid home care is long-term care, financed through Medicaid or out of pocket. Medicare does not cover long-term care, and most U.S. citizens cannot afford or have not purchased long-term care insurance. This leaves no options for millions of older persons or for younger functionally impaired persons without property or income. It leaves professional care providers daily scrambling to help find resources from a polyglot of local charities and other programs that have inadequate scope, wax and wane, and are difficult to find.
Care recipients who qualify for Medicaid and their families depend heavily on paid, unskilled caregivers who may be in the home from 4 to 16 hours per day for ADL support and standby assistance. Even when they can find work, low-income care recipients and family members avoid seeking employment (income) because they risk losing coverage for needed equipment, medication, and personal assistance. People even divorce to secure Medicaid coverage. Long-term care is not covered by most employer-based insurance. Moving across state lines to be near other family members who can help with care can be problematic, since state Medicaid benefits vary greatly. In addition, states cannot carry deficits and may abruptly reduce services when fiscally stressed, whereas federal programs like Medicare are more protected.
Along with impacts on care recipients and families who need the services, health care providers and case managers spend many frustrating hours each week on securing coverage. Medicaid is often the key. Medicaid requires a comprehensive means test, with a threshold near the federal poverty level that varies from state to state.
Medicaid approval and access to services depends on review of detailed information about personal finances and often takes months. Once coverage is approved, weeks more may pass before services start—which can be a long time during a sociomedical crisis. Coverage often lapses due to failure to receive or complete annual update forms, wrong addresses, changes in
regulation, or small changes in household finances. All these policy and regulatory gaps affect both families and home care workers in terms of burden and care outcomes.
Once coverage is in place, medical providers who have little training in or knowledge of long-term care are deluged with volumes of documents to review and sign for supplies, equipment, and services—such as identical forms every 6 months for diapers, disposable pads, or feedings. The paperwork, with the invisible rules behind it, is a de facto throttle on the system. The work, uncompensated and tedious, is an inefficient, ineffective way to manage resource use. Overburdened providers may sign papers to “clear the stack” without scrutiny or fail to sign important forms. The administrative burden derives from the limited financing of long-term care, which generates regulatory demands to document need and use. The lack of incentives for clinicians to engage in oversight has led to a dearth of qualified clinical managers who know what services are needed and what is superfluous or fraudulent.
Because Medicaid long-term care financing is a key theme, it is important to consider the impact of state payment policy on care. States vary fivefold in Medicaid funding for community-based care, which averages 2.3 percent of gross state product (GSP): in 2004, the range was from less than 1 percent in the Dakotas and Nebraska to 4.1 percent in Massachusetts and 4.8 percent in New York. By contrast, nursing home care averaged 7.4 percent of GSP, ranging from 1.9 percent in Alaska and 3.2 percent in Nevada to 12.2 percent in Connecticut. New York and Massachusetts are also above average in nursing home spending, consistent with relatively generous Medicaid programs in those states (see http://www.kff.org/). Some argue that targeted and well-managed Medicaid community-based services meet needs more effectively (Kemper et al., 2008) and reduce state nursing home bills by more than the home care cost. However, state policy has evolved slowly, despite much expert concurrence with such concepts.
The quality of home health aide and personal care also is a recognized national issue (Stone and Newcomer, 2009). There is minimal required training, and caregivers are paid near minimum wage. Commonly reported problems include rapid turnover and aides failing to appear, being rude, stealing, or doing poor work. Of the typical hourly cost of $18, half goes to agency overhead. Conversely, in many cases aides are trusted like family members; donate their own time, money and possessions to the household of the care recipient; and are invaluable. A new survey called Consumer Assessment of Healthcare Providers and Systems (CAHPS) for home and community-based care services has been created to systematically measure quality. The CAHPS program is a family of standardized surveys that ask consumers and patients to evaluate health care experiences, covering issues like provider communication skills and service accessibility.
Individual home care agreements are made by private individuals with other private individuals. With no agency overhead, these workers are paid more. Such arrangements are treasured by those who find reliable help and are difficult to count. Initiatives such as the Cash and Counseling Demonstration and Evaluation (Foster, Dale, and Brown, 2007), in which care recipients or families manage home care using public funds, have succeeded in improving quality and afford insight into the human factors dimensions of long-term care in the community (Foster, Dale, and Brown, 2007; Kemper, 2007; Wiener, 2007).
Between CAHPS and creative new models like Cash and Counseling, plus the increasing affluence of baby boomers, the personal care component of home health care is moving forward, but it will have to contend with the increasing ratio of older persons needing care to younger persons available to deliver it and will increasingly depend on an international workforce.
Assistive Technology and ADL Support
A second core feature of long-term care at home is assistive technology, including durable medical equipment (DME). Financing is less problematic in this area than for personal care, since devices are funded by both Medicaid and Medicare Part B (with a 20 percent copayment for some). Equipment includes basic items, like wheelchairs, walkers, commodes, and hospital beds, and advanced items, like pneumatic patient lifts and powered personal vehicles (Brummel-Smith and Dangiolo, 2009). Key human factors principles include recognizing clinical need, selecting the right device for the person and care setting, training the user, and ensuring maintenance. Although a new era is on the horizon with technical capacity for “smart houses,” the nation is far from ready to disseminate this costly technology to the broad population in need.
Financing approaches include rental (for high-cost devices), monthly fees, rent-to-own, and direct purchase. A Medicare issue arises under rent-to-own: when rental ends in ownership, service support often stops. Items wear out or break. Some repairs are covered; other items are replaced. Too often, bureaucratic delays have serious human clinical consequences.
Historically, profits from high-end items have fed durable medical equipment providers’ bottom lines. These profits paid for customer services (trained support staff who deliver equipment and teach users) and charity care (forgiveness of unattainable copayments). With reduced Medicare payments, support services are shrinking. In most locations, competition and regulatory oversight still ensure new or newly refurbished equipment, but the quality of training and support is uncertain. Further Medicare cuts could weaken support because of the cost of expert respiratory and occupational therapists. Other human factors impacts include fear and risks associated
with the equipment: some devices frighten users and families. Risk of injury or death comes with improper use, so training and support are important. Mandatory oversight of quality is provided by external accrediting bodies and Medicare program surveyors, but it is limited.
Unfettered direct-to-consumer marketing for high-end durable medical equipment, like scooters and lift chairs, plus low-cost items, like moist heating pads that lack proven benefit but feel good, is another problem. Individuals are told “all you need is an order from your doctor” without knowledge of whether they meet coverage criteria. This puts physicians in an awkward position. Informing consumers may relieve unmet needs, yet marketing creates both overuse and inherent resistance among physicians to approve even the legitimate services that are lost in the background of overuse.
Oversight of durable medical equipment is complex. Some is managed under contracts through Medicare Part A organizations, like hospices and nursing homes. This secondary arrangement was devised to prevent duplication and target services, but it complicates physicians’ and discharge coordinators’ roles and alters patient choice. Ordering durable medical equipment requires specialized knowledge. Regulations are opaque; a physician cannot see what is covered when doing paperwork. Overall, misdirected use of durable medical equipment is modest compared with excessive and redundant diagnostic tests and procedures performed by medical specialists and readily accepted by patients and payers.
As technology advances, it will be increasingly vital to educate consumers about selecting the right device and about safe use. I recall recently talking with a young woman now caring for her ventilator-dependent 50-year-old mother at home. The daughter is bright and has learned a lot about highly technical clinical care in a short time. Yet she cannot cope with this challenge without substantial ongoing professional support.
Inadequate Postacute Care Model for Complex Patients
The transition from hospital to home and the immediate posthospital interval are vital. Unacceptable rates of serious errors are well documented, affecting perhaps half of cases: wrong medicines, no medicines, no follow-up plans, and little or no clinical information for postacute providers are common rather than exceptional (Boling, 2009). Postacute home health care is delivered by teams of nurses, therapists, aides, and social workers employed by home health agencies. The care plan is reconstructed at home with limited data. Physician involvement is minimal, making home care the only setting in which seriously ill individuals are cared for almost entirely without active physician input.
Medicare Part A has the largest influence on this home care silo. Cost-based financing in the 1980s and 1990s was associated with exponential
growth and evolution into chronic care rather than the postacute care specified by the 1965 Medicare legislation. The 1997 Balanced Budget Act abruptly transformed the industry (Murkofsky and Alston, 2009). Although gradual modulation was reportedly intended, a regressive Interim Payment System (1997), followed 3 years later by at-risk prospective payment (2000), caused a 50 percent drop in service within 2 years. This affected care recipients and families and caused major losses in the industry talent pool. Agencies rebounded under the Prospective Payment System (PPS) and its defined payments for 60-day episodes. Payment is adjusted for case mix and ranges fivefold in about 80 categories, averaging about $2,400. Visits per 60-day interval dropped sharply (from 50 to 20) with renewed emphasis on post-acute care and efficiency. After initial comprehensive assessment, care focuses on defined goals. Episodes are shorter and families feel pressure from the first day to assume responsibility for care, including wound care and other newly learned technical skills.
Unlike Medicare, the private insurance sector has thousands of payers and no central data source to help evaluate quality or impact. Generally, home care is a small budget item, paid by the visit and approved a few visits at a time. Home care can be used effectively as an integrated feature in some large health systems (Suter et al., 2008), and technologies like telemedicine have been used to augment home care. However, these exemplars are notable for their scarcity.
The quality of Medicare home health agency postacute care is measured nationally using OASIS data. Risk-adjusted outcome data, (Home Care Compare), publicly reported by the Centers for Medicare & Medicaid Services (CMS), are used to improve quality (Schade et al., 2009), modulate payment, and compete. Most agencies are accredited by either the Joint Commission or the Community Health Accreditation Program (CHAP). All are subject to Medicare audits. These rigorous evaluations focus on regulatory compliance and the care process within the silo, not on outcomes across time and settings. Despite instances of excessive or inappropriate care, most fraud and abuse were weeded out in the 1990s by the Office of the Inspector General.
Current home health care models perform well for individuals on a trajectory back to full recovery after an acute illness (Group E), but there are problems in Group F when individuals have advanced chronic illness that waxes and wanes. Hospitalization was frequent under cost-based reimbursement and has increased under PPS: the national risk-adjusted rate in a 60-day Medicare episode is 29 percent (National Association for Home Care and Hospice, 2009). Notably, in 2004 the best 1,750 Medicare agencies averaged 17 percent, while the worst 1,750 averaged 47 percent, showing both high acuity (the best hit only 17 percent) and opportunity for improvement in others. Similarly, in the final 2 years of life, marked
geographic variation in home care spending and a positive correlation between home health care cost and hospitalization argue against the case for effective substitution (Wennberg et al., 2008). Finally, 1 in 7 Medicare home health episodes ends in hospitalization within 2 weeks; this highlights the need to engage immediately and actively.
This overall mediocre record relates to operating in silos, poor communication, lack of aggressive medication reconciliation, absence of medical care and skilled case management in the early weeks, and thus inability to respond to changes in condition. In 35 percent of cases, there is no medical encounter during the home health episode, showing limited physician participation (Wolff et al., in press). Even when there are encounters, they are not timely and not provided to individuals who need them most. However, the exemplars show what is possible when there is a systematic approach to home-based care as a defined focus.
The value of integrating active medical management with home health agency work is supported by transitional care research. In rigorous clinical trials, advanced practice nurses were assigned to selected, high-risk cases in the hospital, and the cases were followed closely at home for a month (Naylor et al., 1999), producing significant cost savings (50 percent). Issues remain of “scalability” and who will care for graduates of a new “transitional care” silo when they can’t return to a clinic. Savings in 2004 were about $5,000 per case.
Less intensive but perhaps more scalable are “guided care” (Leff et al., 2009) or transition coach strategies (Coleman et al., 2006), which rely more on the health record and the care recipient or caregiver. In this model, busy office-based physicians with many competing responsibilities are expected to address medical management needs. It is unclear whether this approach is robust enough to alter outcomes for complex, unstable patients. Savings are about $500 to $1,000 per patient.
No Longitudinal Care in Complex Cases with High Comorbidity
Ultimately, home care for Group F members with high comorbidity suffers greatly from intermittency and lack of medical continuity. Frail individuals move into and out of skilled care and lack consistent physician contact over time and during critical intervals; one-third of the time there is no billed physician service during a home health care episode (Wolff et al., in press). This burdens the care recipients, families, and the nonphysician home care professionals who need physician guidance on the complex medical issues that are prevalent.
When home health agency care starts, the agency team performs extensive data gathering and then provides care. If care recipients don’t die, they
improve or they go to a hospital or nursing home and are discharged from home health care. Many different physicians are involved only tangentially, and there is little continuity. At home, individuals and their families lack access to trusted medical advice when having trouble and receive little in-person medical care until they are sick enough to visit an emergency department. Functional deficits hinder physician office visits. If they make the trip, they need advance notice and expensive transportation, while worsening health may afford little warning time to schedule appointments. Family members must take time off work to attend physician encounters. Born of financing mechanisms, this is an expensive care model that is ill suited to population needs. The care model adversely affects care recipients, their families, the home health care staff, and the physicians involved.
Physicians started fading from home care 40 years ago, as they became busy with other roles and house call fees failed to keep pace with costs. When the 1992 Resource-Based Relative Value Scale (RBRVS) sought to level the playing field between procedural and cognitive services in Medicare payment, house calls were rare and were never addressed. House calls were formally valued in 1998 through advocacy by the American Academy of Home Care Physicians and secondarily supported by family physicians and geriatricians. In 2001, visits to group homes and assisted living facilities were also upgraded so that margins were thin but adequate for those with a passion for this work to form new practices. In 2007, proposed payment cuts of 20 percent on house call codes again threatened new mobile practices. Urgent advocacy reduced the cuts, reinforcing need for constant vigilance and the human impact of public policy on care providers.
Unsolved problems also remain for house calls in the Medicare fee-for-service arena. The practice expense component (about half of the total service code value) omits travel time, and cost inputs are based on an office scenario. Furthermore, since 1997 state Medicaid programs are not obliged to pay Medicare Part B copayments; most pay below Medicare’s 80th percentile, forcing a 20 percent loss on those who serve low-income populations. Case management, which occupies about 30 percent of work time in home medical care, is underrepresented in the fee schedule. New codes for reviewing and signing home health agency care plans and overseeing complex agency cases were added in 1995 and 2001. Although well meant, the oversight codes require too much fine-grained documentation of minutes spent and apply only to select patients. Overall costs, including travel, phone calls, and paperwork, exceed Medicare revenue, requiring home-based medical programs to seek other revenue sources to create a sustainable business plan, rather than concentrating on those with the greatest needs for home care and medical attention.
With few exceptions, private payers have expressed little interest in home medical care, preferring to concentrate on other issues. This dis-
advantages younger adults and low-income (Medicaid) individuals, who again lack access unless their physicians choose to accept financial burdens by seeing them for payment that is below cost, if they are paid at all.
Finally, regulatory oversight of house calls by some Medicare carriers has burdened providers. Audits for all billed claims, which cause payment delays and restricted cash flow, have bankrupted high-quality house call programs, leaving the people they served with no recourse. Audits are a side-effect of mistrust and automated fraud prevention systems. House call providers exhibit high-frequency use of house call codes that are used rarely by office-based peers, thus flagging house call providers’ profiles as outliers. Those with the stamina and the cash to fight find most or all denials overturned on appeal, but many have given up and sought other lines of work. Audit abuse has not been addressed by CMS management, perhaps due to competing priorities. Senior CMS administrators deny intent to discourage house calls as a matter of national policy.
Payment and compliance regulation are often linked with quality measurement. In home medical care, formal evaluation of quality is offered by the Joint Commission. Only large health systems can afford the considerable cost and the loss of efficiency associated with data gathering, and there is no economic value from accreditation. Other attempts have been made to develop measures that are suited to the role, but these are not in general use. An expert panel identified appropriate quality benchmarks for home care using Assessing Care of Vulnerable Elders (ACOVE) indicators as a platform (Smith, Soriano, and Boal, 2007). There also is a house call provider exam offered by the American Academy of Home Care Physicians. Most standard quality benchmarks, like diabetic control and cancer screening rates, are meaningless in the care of people in the final years of life and advanced states of ill health (Boyd et al., 2005). One thing does stand out in virtually every home care study: care recipients prefer it.
On balance, facing significant disincentives and lower income potential has caused all but the most committed champions to avoid home medical practice. This fact stands in contrast to the growing evidence of economic and clinical benefit from a home medical care model that includes transitional care.
The Hidden Costs of Privatizing Public Benefits
Budgeting only 2.5 percent for administration, in contrast to the lowest cost private payers at 7 percent, Medicare runs lean and almost all Medicare funds pay for actual care. Critics argue that more administrative controls would reduce overuse and fraud. Providers complain about low payments compared with the private sector. Economists note that lack of competition leads to inefficiency and poor quality. Still, it is instructive to consider
two decades of experience with private-sector diversification of Medicare, including explosive growth as Part D and the 2003 Medicare Modernization Act have resulted in 4,000 additional private-sector plans. This experience should teach caution when considering human factor implications.
Among new options, prescription drug plans (PDPs) offer only drug coverage, whereas Medicare Advantage plans offer one-stop shopping for all health care. Two positive results are notable: consumers are more aware of drug costs and willing to accept generic alternatives, and half of the 6 million “Medicare poor” have newly attained drug coverage, along with 20 million others who had less need of the expensive new government benefit (Zhang et al., 2009).
Problems with human factors implications counterbalance the positive results. The new plans have created much unfunded work and frustration for physicians, pharmacists, and consumers, as hours are spent every week navigating a forest of new formularies and sluggish preauthorization processes for drug or service coverage, which can take an hour for a single item and regularly cause delays in care. Many beneficiaries do not know what plan they have or what it covers. Health plans sell policies to unwitting individuals with low health literacy or dementia. The plans restrict care and disrupt established relationships without recipients’ being aware of the impacts. Marketers create favorable selection bias (outside Walmart, not in senior centers, where frail people congregate). Some plans employ care management for enrollees with higher comorbidity, but most focus on maximizing enrollment. MedPAC now advises reduced incentives for private insurers. Finally, patients often change plans, further complicating matters.
When debating administrative efficiency, some consider the private insurance industry better suited than government. However, recall that the main drags on current public payment systems relate to defining eligibility (means testing) and providing care with limited resources (de facto rationing), not to administrative inefficiency once services are adequately funded. Experience with services actively regulated by private payers shows those processes to be more burdensome and expensive than government. Experts (Kronick, 2009; Zarabozo and Harrison, 2009) have found that cost shifting in Medicare managed care has increased overall cost by 13 percent, with fewer services rendered and reduced consumer perception of care quality among sicker beneficiaries who most need care (Keenan et al., 2009).
EVIDENCE OF VALUE FROM MEDICALLY LED HOME CARE TEAMS
Evidence is mounting that medically managed home care can produce outcomes far superior to 1-2 percent improvements in major endpoints (e.g., hospitalization) that are routinely accepted in drug trials and other clinical research.
One source of data is the Department of Veterans Affairs (VA) health system, in which systematic implementation of standardized home-based primary care (HBPC) produced a 24 percent decrease in overall costs for care of these ailing veterans (now over 11,000). These programs now adhere to national standards for patient selection, care process, and quality, producing greater homogeneity and a more comprehensive approach than the HBPC model tested at 16 sites in the 1990s. At that time, there was improved satisfaction, but variable model adherence and no cost savings. Current national VA data show that although home care costs increase with HBPC, much lower hospital costs (–62 percent) and nursing home costs generate overall savings. While the data are limited to the VA, the results are impressive and the sample is large (Beales and Edes, 2009).
The Program of All-Inclusive Care for the Elderly (PACE) model, a permanent federal benefit since 1997, also offers insight. PACE teams serve frail low-income older persons with Medicare and Medicaid coverage who live in the community but meet criteria for nursing home care. The care model is organized around a community day health center. Financing is a global risk agreement, funded by monthly capitation from Medicare based on projected costs (using Hierarchical Condition Category scoring) plus Medicaid costs. PACE now has 65 sites and 17,000 enrollees. The program has grown slowly, in part due to stringent enrollment requirements, substantial capital and escrow needs (now around $4 million per site), and low capacity (150-200 per site). Analyses show that change in care financing and delivery can reduce dependence on inpatient settings for sick, disabled populations (Kane et al., 2006). PACE participants and providers consider this an excellent model.
Another option for frail Medicare beneficiaries is the Special Needs Program (SNP) offered since the late 1990s, now with about 775 programs and 1 million enrollees. This is a Medicare managed care risk contract for persons with multimorbidity and higher costs. Success has varied. Core issues are patient selection and recruitment, plus clinical team makeup and fiscal management. Most of the programs arise in traditional care systems and conventional provider groups that may not be ready for this work.
Hospital at Home, which provides acute home care for patients who meet criteria for inpatient hospital care, has also shown improved satisfaction and fewer in-hospital complications, though as yet no definite cost savings (Cheng, Montalto, and Leff, 2009).
Office Without Walls (OWW) is a special program developed in one Nevada geriatric managed care organization. Unpublished data compare its enrollees (n = 437) with 875 comparably old (age 86) and ill persons managed by office practices. OWW members received 3,365 home visits but no office visits and had much less hospital use (33 percent less emergency department use and 50 percent fewer bed days), saving $1.5 million per
year ($1,500 per person) if adjusted to an enrolled population of 1,000 individuals.
A recent randomized clinical trial of an at-home geriatric assessment and consultation program for low-income seniors in Indianapolis called GRACE showed marked improvement in clinical care. There was also a 40 percent reduction in hospital use during the second year of operation (Counsell et al., 2007) among higher risk participants, even though this was not a case management or primary care model and did not include urgent response capacity.
In sum, comparing current usual home health care to office, hospital, or even nursing home settings, one finds a model that is largely missing one key ingredient: an active medical presence. Yet substantial evidence and common sense both suggest that combining medical care with other home health services will improve outcomes, will save money, and is preferred. Financial incentives are a key reason for this striking incongruity. As a matter of health policy, for the benefit of both users and providers, a new, integrated home care model is needed, free of the cost of bricks and mortar and empowered by portable technology.
FUTURE POLICY DIRECTIONS
In considering people with functional loss and advanced chronic illness (Groups C through F), there are two main policy priorities. One is a national policy for financing long-term care in support of activities of daily living that incorporates a broad vision with the home as the preferred locus and center for care. To achieve this, a systems approach must be employed, with careful consideration of the many contributing elements: living space design, technology applications of a variety of types, workforce development and training, financing mechanisms with incentives aligned to attain objectives, and community support services, like delivered food and other necessary amenities.
The other priority is redesign of medical care for people who are frequent users of costly services, again using a broad conceptualization of what home care can deliver. This requires changes in both delivery systems and regulation and financing, with incentives aligned to produce desired results and the use of setting-specific quality measures, like those discussed earlier. Payment reform alone is inadequate, as managed care experiments have shown. Because home care has been a stepchild in health care policy, it is necessary to put new home care models into the spotlight—models that are patient-centered and focus on population needs—rather than starting with the needs of the health care system and its funding sources.
The serious health problems of community-dwelling elderly persons cannot be addressed in silos or by managing one condition at a time.
Certain systemic conditions (like vascular disease, either alone or with diabetes) underpin much of the most costly care but produce multiple ailing body systems: a person with coronary disease, peripheral vascular disease, ischemic stroke, chronic kidney disease, congestive heart failure, atrial fibrillation, peripheral neuropathy, visual impairment, and diabetes has many conditions. Most of these diagnoses involve vascular pathology, but the key to optimal management is a core team of advanced generalists who can concurrently treat all of the conditions, with help as needed from organ-specific specialists, rather than having each of several different specialized teams manage one condition, or “care by committee.” Going to the care recipient’s home is also the only way to really know what is going on in many such cases. Focusing on these costly, immobile individuals with enhanced, medically led, interdisciplinary care delivered in the home can reduce resource use by up to 50 percent while improving safety and satisfaction.
Payment models must be addressed simultaneously. Seasoned administrators and noted analysts write about strategies to control runaway Medicare and Medicaid costs (Hsiao, 2007; Rosenthal, 2009; Wilensky, 2009). Options are few: fee-for-service, per-episode (short-term risk-bearing arrangement), capitation (longer-term risk-bearing arrangement that may include global or partial risk), salary with defined responsibilities (VA or staff model health maintenance organization), and gain-sharing based on predicted costs. Ultimately providers must be either placed at risk or rewarded differentially on the basis of desired outcomes. Quality must be part of the final formula, linking results to payment. Pay for performance is a way of modifying payment based on specific process measures. Gainsharing is a results-based form of pay for performance.
For physicians, fee-for-service payment incentives foster high-volume service in efficient practice settings; they maximize encounters and procedures rather than care coordination and focused management of a few high-cost individuals. And fees for home visits and for care coordination are inadequate to generate sustained interest and growth. However, studies show that alternate modes of financing can attract physicians (Devlin, 2008).
Risk-bearing strategies require capital reserves and large organizations. To date, many risk-bearing mechanisms were subverted by recruiting low-cost individuals. Risk-bearing arrangements for complex patients have either enrolled small numbers (PACE) or failed due to their care model. In any organization, unless senior management is committed to innovation, advanced home care is unlikely to evolve. Pay for performance in conventional financing models offers little incentive for the arduous case management work needed to improve patterns of use. Gain-sharing in conjunction with restructured care delivery systems remains largely untested.
One new idea is Independence At Home (IAH), a gain-sharing strategy introduced in Congress in 2009 (H.R. 2560 and S. 1131) and included in the final health reform bill as Section 3024. It is based on using a technology-enhanced, medically led mobile team to care for small numbers (150 minimum) of high-risk Medicare beneficiaries. Leaving Medicare Parts A and B intact, IAH teams would share in savings from prevented hospitalizations, creating an incentive to work hard and to be responsive. Savings would be calculated based on predicted costs of the population served. IAH teams would follow vulnerable, medically complex, function-limited persons for years, probably for life in most cases, providing prompt access to holistic, comprehensive care, and a one-stop shop. IAH teams would gradually expand, “building out” the problems in the current system.
PACE has shown that CMS can estimate cost at the individual level well enough for its sites to succeed despite small size and wide variance around mean per-person costs. With protections against inadequate service that include measures of satisfaction, easy voluntary withdrawal, and selected evidence-based chronic care process measures that are meaningful to patients with multimorbidity, IAH offers improved care plus cost savings. It is designed for low-cost, local start-up to speed implementation, without having large risk contracts or physical plants. Electronic health records are essential and expected to improve quality and safety and enhance efficiency, but they are also costly. IAH teams will invest in health information technology from the margins created by gain-sharing.
It is difficult to precisely calculate potential cost savings from wide-scale adoption of an IAH model or others like it, which focus care on the people at greatest risk, aligning incentives while changing primary care delivery, but it can be estimated. Ranking Medicare beneficiaries by cost, in the top 10 percent (4 million people) the average 2009 cost is about $50,000. Reducing average cost 25 percent would save $50 billion per year. And the 4 million is also the number of persons in the center of the target using functional status as the primary selection criterion.
A low-end estimate could be based on home health agency Part A rehospitalizations. Using the 29 percent national rate and 7 million episodes yields 2 million hospital admissions. A 25 percent improvement would avert 500,000 hospitalizations; if these averaged $15,000, Medicare savings would be $7.5 billion. This is a low estimate, as some at-risk persons are either not referred or are in and out of Part A care.
The VA Home-Based Primary Care Program is another frame. In 2002 dollars, annual home care cost per person was $11,000 more after HBPC than before it. This was offset by reduced hospital costs ($11,000) and nursing home costs ($10,000). Total costs were lowered from $38,000 to $29,000 (24 percent). Adjusting for inflation at 5 percent and applying this math to 4 million Medicare beneficiaries who cost $50,000 each in 2009,
home care would cost $62 billion more but hospital and nursing home use would be $116 billion lower, saving $54 billion overall.
To satisfy critics, it is important to address concerns about workforce and the impact of an IAH “carve-out” on other stakeholders and other new health care designs.
The baby boom will change the shape of health care. If the current path continues, hospitals can anticipate more Medicare volume and the prospect of providing those services at a discounted rate relative to current payments, which are often at or below cost. Thus, hospitals should support IAH in order to focus on what they do best: advanced surgery, intensive care, and complex inpatient management using lots of high technology.
Typical primary care physicians already coordinate care with 117 other practices (Pham et al., 2009), and they also have more than enough clients. Furthermore, the frail IAH population is not ideal for office practice, since these persons require more time during encounters and much more time between encounters. High-volume physician offices will always struggle to meet the needs of this complex and immobile population. Although a large enough office practice can structure a support system that will accommodate the care management requirements of this population, the physician owners or business managers must intentionally choose to set aside time to care for the immobile patients and the payment system must support that, or these patients will always be marginalized as they are now. None of the other health care reform options will achieve this, except for the most advanced medical homes, after many years of substantial investment.
IAH can be viewed as a specialized patient-centered medical home (PCMH) without bricks and mortar. Taking a frail, sick older person out of bed and forcing them to travel to the physician’s office, often requiring a family member to take time off work to make sure key issues are addressed, does not seem patient-centered. An office-based PCMH strategy of using telemedicine and delegating direct care to professionals with less training, in lieu of physician care, would create a lower standard of care for some of the sickest persons in society. Finally, the PCMH financial model is inadequate to cover the extra care costs of this population.
Specialty physicians likewise will have more than enough Medicare business and would not miss seeing the IAH population subset who require more time and bring more risk-benefit issues and more complex decisions to procedural care. Specialists face the same risk as hospitals—of working harder for lower rates unless ways can be found to target resources and lower overall Medicare costs.
The accountable care organization (ACO) model is designed to compensate large health systems for care of a population that includes frail, complex cases. An IAH program would help an accountable care organization deliver care and control costs, just as it would for a Medicaid Advan-
tage plan that serves large numbers of high-cost elders or a special needs program. An IAH program matches the financial needs of accountable care organizations and special needs programs. The question remains whether most such large entities will ever develop home-based care models. Historically, most have not done so.
It is likely that varied approaches will emerge for managing patients with advanced chronic illness. Some will use more telemedicine, more smart technology, more specialized consultation by pharmacists, more fully developed interdisciplinary teams. Some will purchase additional social work, personal care, community care managers, and other support services.
PROFESSIONAL WORKFORCE DEVELOPMENT AND OTHER POSSIBLE BARRIERS
Critics of the home care model for advanced chronic illness care have focused on the lack of sufficient experienced medical providers (physicians, nurse practitioners, other advance practice nurses, physician assistants). Although rarely noticed, a few thousand medical providers nationally have focused their efforts on mobile medical care using Part B payments and contract roles. Care of 300 very sick older persons at home might require a team with two nurse practitioners, a physician, and a small staff. To reach 4 million patients under these assumptions would require 13,000 physicians and 26,000 nurse practitioners supported by a team. Much of the team exists in home care, except for the medical component. The question is: How long would it take to grow such a workforce?
Consider hospital medicine, in which a “hospitalist” employed by the hospital assumes the responsibilities of attending physician and case manager for the duration of an in-patient’s stay. This rapidly growing field with 30,000 workers arose in a decade from only 500 in 1997 due to a reliable payment source (the hospital) and a defined clinical role that is appealing to young physicians. Hospitalists work hard, with defined schedules and strong infrastructure supports. One can imagine medical home care teams emerging quickly given suitable incentives and supports, particularly within the established framework of home health agencies. Young physicians are inherently altruistic. Many would welcome the challenge and personal rewards of this work, for which care recipients and families are very grateful. The IAH teams in turn will attract more talented nurses and other workers to the field.
After 10 years, a successful gain-sharing model might contend with insufficient numbers of persons in usual care against which to calculate savings as well as statistical regression to the mean, which would impact estimated cost savings. Recalibration and possibly even a revised payment
model would be needed. By then, the market and provider workforce would be ready for risk-based care for frail elders living in the community.
The biggest barriers to effective change in home care will come from lack of vision, entrenched systems, and interests that seek to maintain current revenue streams, to their own detriment. Change must come, and early adopters will prosper. Horizontal and vertical integration must be expected, much as was predicted during the “failed” national experiment with for-profit managed care, which has worked best for people who are not ill.
In order to meet the needs of the sickest individuals in the home, those delivering care need flexible and efficient designs while promoting high-quality care and proper rewards and allowing patients to move between settings, as suggested in Figure 12-4.
Importantly, incentives are needed for members of an interdisciplinary home-based team to engage in the difficult aspects of this work for the members of Group F:
Selecting appropriate persons for intensive team care; suitable candidates are much sicker than the average person seen by office physicians or home health.
Making and implementing complex medical decisions for ongoing care of individuals with multiple active problems, often several problems that are unstable or meta-stable, and for which treating one impacts several others.
Timely, effective response to urgent situations in complex patients.
Interacting with hospitals and specialty physicians in complicated cases.
Care coordination involving multiple services and multiple sources of support.
Addressing unrealistic expectations of the care recipient and the family for what health care can deliver and compassionately making difficult decisions.
Disentangling social from medical issues; a line must be drawn when people seek to use medical resources to address social needs that are sometimes limitless.
Finding or creating community resources to solve complex problems.
The complex scenarios in which advanced medical home care providers, care recipients, and their families find themselves remind one of surgery, played out in slow motion over years. Care recipients depend heavily on a trustworthy physician-led interdisciplinary team who will see them through their entire course of care. Serious consequences attach to ill-advised decisions. Care providers must know how to prepare, what to cut, what not to cut, and how to manage the recuperative period. It takes considerable knowledge, training, skill, and experience to do this well. And each case is a deeply personal human drama that leaves indelible lifelong memories in all who are involved.
In summary, home care has an extensive workforce at varying levels of skill and has developed a much enhanced range of technical capability that is needed by people of all ages. By having a broad vision that has its foundation in population needs and a willingness to restructure health care so that it is intentionally organized around the home and in the community and is thus truly patient-centered, people can achieve greater user satisfaction, better outcomes, and lower cost. Care and support should be targeted and proportionate to need. Robust quality measures have been developed that are appropriate to home care models and can be used to ensure value-based purchasing. Much can be learned from other nations that have invested in community-based care as society moves forward. A systems-based approach with prominent consideration of human factors
will be vital to the proposed and much-needed redesign of payment, regulation, and care delivery in the home.
ABOUT THE AUTHOR
Peter A. Boling is professor of medicine at Virginia Commonwealth University, where he heads geriatrics and is interim chair of general internal medicine, which includes geriatrics, ambulatory care, hospitalists, and women’s health, medical education, and research. He has a longstanding interest in health policy and has published and spoken widely on home care.
Akpolat, T., Dilek, M., Aydogdu, T., Adibeli, Z., Erdem, D.G., and Erdem, E. (2009). Home sphygmomanometers: Validation vs. accuracy. Blood Pressure Monitoring, 14, 26-31.
Beales, J.L., Edes, T. (2009). Veteran’s Affairs home based primary care. Clinics in Geriatric Medicine, 25(1), 149-154.
Boling, P.A. (2009). Care transitions and home health care. Clinics in Geriatric Medicine, 25(1), 135-148.
Boyd, C.M., Darer, J., Boult, C., Fried, L.P., Boult, L., and Wu, A.W. (2005). Clinical practice guidelines and quality of care for older adults with multiple co-morbid diseases: implications for pay-for-performance. Journal of the American Medical Association, 294(6), 716-724.
Brown, R., Peikes, D., Chen, A., and Schore, J. (2008). 15-site randomized trial of coordinated care in Medicare FFS. Health Care Financing Review, 30(1), 1-25.
Brummel-Smith, K., and Dangiolo, M. (2009). Assistive technologies in the home. Clinics in Geriatric Medicine, 25(1), 61-79.
Centers for Disease Control and Prevention. (2008). Vital and health statistics series 10, Number 242: Summary health statistics for U.S. adults. National Health Interview Survey, 2008. Available: http://www.cdc.gov/nchs/data/series/sr_10/sr10_242.pdf [accessed September 2009].
Centers for Disease Control and Prevention. (2009). National Health Interview Survey, 2003-2007. Table 1: Frequency Distributions of Numbers of Limitations in Activities of ADL and IADL from 2003 to 2007. Available: http://www.cdc.gov/nchs/health_policy/ADL_IADL_tables.pdf [accessed September 2009].
Cheng, J., Montalto, M., and Leff, B. (2009). Hospital at Home. Clinics in Geriatric Medicine, 25(1), 79-82.
Coe, N.B., and Van Houten, C.H. (2009). Caring for mom and neglecting yourself? The health effects of caring for an elderly parent. Health Economics, 18, 991-1,010.
Coleman, E.A., Parry, C., Chalmers, S., and Min, S.J. (2006). The care transitions intervention: results of a randomized controlled trial. Archives of Internal Medicine, 166, 1,822-1,828.
Counsell, S.R., Callahan, C.M., Clark, D.O., Tu, W., Buttar, A.B., Stump, T.E., and Ricketts, G.D. (2007). Geriatric care management for low-income seniors: A randomized controlled trial. Journal of the American Medical Association, 298(22), 2,623-2,633.
Devlin, R.A. (2008). Do physician remuneration schemes matter? The case of Canadian family physicians. Journal of Health Economics, 27, 1,168-1,181.
Foster, L., Dale, S.B., and Brown, R. (2007). How caregivers and workers fared in Cash and Counseling. Health Services Research, 42(1, Part II), 510-532.
Giovannetti, E.R., Wolff, J.L., Frick, K.D., and Boult, C. (2009). Construct validity of the Work Productivity and Activity Impairment Questionnaire across informal caregivers of chronically ill older patients. Value in Health, 12(6), 1011-1017.
Goldfield, N.I., McCullough, E.C., Hughes, J.S., Tang, A.M., Eastman, B., Rawlins, L.K., and Averill, R.F. (2008). Identifying potentially preventable readmissions. Health Care Financing Review, 30(1), 75-91.
Henry J. Kaiser Family Foundation. (2009). Medicare Advantage. Website publication #2052-12. Available: http://www.kff.org/Medicare/upload/2052-12.pdf [accessed September 2009].
Holtz-Eakin, D. (2005). High-cost Medicare beneficiaries. Congressional Budget Office report to Congress. May. Available: http://www.cbo.gov/ftpdocs/63xx/doc 6322/05-03 [accessed September 2009].
Hsiao, W.C. (2007). Why is a systemic view of health financing necessary? More money is a necessary but insufficient condition for better health. Health Affairs, 26(4), 950-961.
Institute for Health Policy and Clinical Practice. (2008). The Dartmouth atlas of health care. Available: http://www.dartmouthatlas.org/atlases/2008_Chronic_Care_Atlas.pdf [accessed September 2009].
Kane, R.L., Homyak, P., Bershadsky, B., Flood, S. (2006). Journal of Gerontology, 61A(7), 689-693.
Keenan, P.S., Elliott, M.N., Cleary, P.D., Zavlavsky, A.M., and Landon, B.E. (2009). Quality assessments by sick and healthy beneficiaries in traditional Medicare and Medicare managed care. Medical Care, 47, 882-888.
Kemper, P. (2007). Commentary: Social experimentation at its best: The Cash and Counseling demonstration and its implications. Health Services Research, 42(1, Part II), 577-586.
Kemper, P., Weaver, F., Short, P.F., Shea, D., and Kang, H. (2008). Meeting the need for personal care among the elderly: Does Medicaid home care spending matter. Health Services Research, 44(1, Part II), 344-362.
Kronick, R. (2009). Medicare and HMOs—the search for accountability. New England Journal of Medicine, 360(20), 2,048-2,050.
Leff, B., Reider, L., Frick, K.D., Scharfstein, D.O., Boyd, C.M., Frey, K., Karm, L., and Boult, C. (2009). Guided care and the cost of complex healthcare: A preliminary report. American Journal of Managed Care, 15(8), 555-559.
Linden, A., and Adler-Milstein, J. (2008). Medicare disease management in policy context. Health Care Financing Review, 29(3), 1-11.
Murkofsky, R.L., and Alston, K. (2009). The past, present and future of skilled home health agency care. Clinics in Geriatric Medicine, 25(1), 1-18.
National Association for Home Care & Hospice. (2009). Home care and hospice facts and stats. Available: http://www.nahc.org [accessed September 2009].
Naylor, M.D., Brooten, D., Campbell, R., Jacobsen, B.S., Mezey, M.D., Pauly, M.V., and Schwartz, J.S. (1999). Comprehensive discharge planning and home care follow-up of hospitalized elders. Journal of the American Medical Association, 281(7), 613-620.
Ogedegbe, G., and Schoenthaler, A. (2006). A systematic review of the effects of home blood pressure monitoring on medication adherence. Journal of Clinical Hypertension, 8(3), 174-180.
Pham, H.H., O’Malley, A.S., Back, P.B., Salontz-Martinez, C., and Schrag, D. (2009). Primary care physicians’ links to other physicians through Medicare patients: The scope of care coordination. Annals of Internal Medicine, 150, 236-242.
Potetz, L., and Cubanski, J. (2009). A primer on Medicare financing (pages 1-22). Henry J Kaiser Family Foundation. July. Available: http://www.kff.org/medicare/7731.cfm [accessed September 2009].
Rosenthal, M.B. (2009). Beyond pay for performance—emerging models of provider-payment reform. New England Journal of Medicine, 359(12), 1,197-1,200.
Schade, C.P., Esslinger, E., Anderson, D., Sun, Y., and Knowles, B. (2009). Impact of a national campaign on hospital readmissions in home care patients. International Journal for Quality in Health Care, 21(3), 176-182.
Schoen, C., Osborn, R., How, S.K.H., Doty, M.M., and Peugh, J. (2008). In chronic condition: Experiences of patients with complex health care needs, in eight countries. Health Affairs, 28(1), w1-w16.
Shelledy, C.D., Legrand, T.S., Gardner, D.D., and Peters, J.I. (2009). A randomized, controlled study to evaluate the role of an in-home disease management program provided by respiratory therapists in improving outcomes and reducing the cost of care. Journal of Asthma, 46, 194-201.
Smith, K.L., Soriano, T.A., and Boal, J. (2007). Brief communication: National quality of care standards in home-based primary care. Annals of Internal Medicine, 146, 188-192.
Stone, R., and Newcomer, R. (2009). Advances and issue in personal care. Clinics in Geriatric Medicine, 25(1), 35-60.
Suter, P., Hennessey, B., Harrison, G., Fagan, M., Norman, B., and Suter, W.N. (2008). Home-based chronic care: An expanded integrative model for home care professionals. Home Healthcare Nurse, 26(4), 222-229.
Wennberg, J.E., Fisher, E.S., Goodman, D.C., and Skinner, J.S. (2008). Tracking the care of patients with severe chronic illness: The Dartmouth atlas of health care. pp. 180. Available: http://www.dartmouthatlas.org/downloads/atlases/2008_Chronic_Care_Atlas.pdf [accessed May 2010].
Wiener, J.M. (2007). Commentary: Cash and Counseling in international context. Health Services Research, 42(1, Part II), 510-532.
Wilensky, G.R. (2009). Reforming Medicare’s physician payment system. New England Journal of Medicine, 360(7), 653-655.
Wolff, J.L., Meadow, A., Boyd, C.M., Weiss, C.O., and Leff, B. (in press). Physician evaluation and management of Medicare home health patients. Submitted to Medical Care.
Zarabozo, C., and Harrison, S. (2009). Payment policy and the growth of Medicare Advantage. Health Affairs, 28(1), 55-67.
Zhang, Y., Donohue, J.M., Lave, J.R., O’Donnell, G., and Newhouse, J.P. (2009). The effect of Medicare Part D on drug and medical spending. New England Journal of Medicine, 361, 52-61.