MMH074.qxd 4/8/09 5:09 PM Page 238 Adam Darkins is the Chief Consultant for Care Coordination at the US Department of Veterans Affairs. He has a track record of developing the clinical, technology and business processes necessary to implement, manage and sustain innovative new healthcare programmes that involve the use of health informatics, telehealth and disease management. His experience includes working in both government and nongovernmental sectors of healthcare in both the USA and Europe and a clinical background in neurosurgery. Jay H. Sanders is CEO of The Global Telemedicine Group, Adjunct Professor of Medicine at Johns Hopkins School of Medicine and a Founding Board Member and President Emeritus of the American Telemedicine Association. He developed the first statewide telemedicine system in the USA, the first correctional telemedicine programme, and the first telehomecare technology. He is a graduate of Harvard Medical School, magna cum laude, was a member of AOA, and did his residency training at the Massachusetts General Hospital. Remote patient monitoring in home healthcare: Lessons learned from advanced users Adam Darkins and Jay H. Sanders Received (in revised form): 6th April 2009 Abstract Most healthcare organisations will face the challenge of having to reengineer their legacy care delivery systems to accept health information technologies (HIT) that will improve their efficiency and effectiveness. Current facility-based healthcare provision that was designed to cope with acute exacerbations of disease and to undertake clinical procedures will need to adapt and restructure to cope with the logistic and economic burden of caring for aging populations with a preponderance of chronic healthcare needs related to conditions such as diabetes, heart failure, chronic obstructive pulmonary disease and high blood pressure. HIT is revolutionising the way in which healthcare is being provided and promises to make the home into the preferred place of care. The advantages of this new paradigm are high levels of patient satisfaction, early intervention for disease progression, support for care-givers, and economic benefits associated with reduced hospitalisation rates. This paper reviews the underlying drivers to adopt home telehealth, an advanced HIT application, and the related clinical, technological and business challenges this presents. It explains why this is an essential strategy that forward-thinking healthcare providers must adopt. Given the associated social and cultural changes the adoption of home telehealth will bring, a vision is outlined of how the routine monitoring of health indices will promote health and not simply stave off disease. Keywords: home telehealth, home care, outcomes, integrated care, chronic disease management Introduction Jay H. Sanders 1317 Vincent Place McLean Virginia USA Tel/Fax: 1 703 448 9640 E-mail: jsanders@tgtg.com 238 The evaluation of patients at home is not a new undertaking for healthcare delivery systems. It dates back as far as the provision of healthcare itself. In 1930, for example, 40 per cent of all physician/patient interactions were 'house calls'. In the second half of the 20th century, the introduction of ever-more sophisticated technologies to comprehensively diagnose and treat medical conditions prompted the migration of the examination room away from the patient's home and into doctor's offices and hospital settings. As a result of this migration, by 1980, 'house calls' constituted less than 1 per cent of all physician/patient encounters in most Western industrialised countries. HENRY STEWART PUBLICATIONS 1753-3031. Journal of Management & Marketing in Healthcare. VOL. 2 NO. 3. PP 238-252. JULY 2009 MMH074.qxd 4/8/09 5:09 PM Page 239 Remote patient monitoring in home healthcare Now, in the first half of the 21st century, the pendulum is swinging back and care is set to centre on the patient's home once again. The main driver behind this change is a need to undertake frequent health status assessments to manage the progression of chronic diseases at an early stage. An aging population that has an overwhelming burden of chronic conditions, and for whom the cost, inconvenience and delay of travelling to healthcare facilities is a deterrent to care, makes the restructuring of the healthcare environment of particular importance. The 'information age' is set to bring a ubiquitous uptake of telecommunications and information technologies to support entertainment and work in the home environment and as it does it offers a platform upon which there is a resurgence of interest in making the patient's home into the preferred site for care delivery. Applying information and telecommunications technologies to deliver care in situations where patients and practitioners are geographically distant is known as telehealth. Remote monitoring of patients by practitioners with the aim of intervening to prevent a subsequent deterioration in their condition is commonly known as home telehealth. In the literature, the term 'home telehealth' is somewhat confusing; another term, 'telecare', is also used. For the purposes of this paper, remote patient monitoring, home telehealth and telecare are synonymous. The resultant change in healthcare practice and the benefits that accrue from this innovation are of profound importance to all healthcare systems. Home telehealth offers a way to focus on population health1 and in doing so emphasises prevention and patient self-management. This paper examines how current technologies make remote patient monitoring possible, reviews telehealth deployments, and discusses the implications from the lessons learned. Technologies supporting the remote monitoring of patients in the home It was clear in the late 1990s that advances in microprocessor technology coupled with parallel changes in telecommunications infrastructure were set to revolutionise healthcare and open up access to the home.2 Instead of requiring the physical co-location of patient and practitioner, these emerging technologies enabled virtual patient-practitioner interactions and, where appropriate, ongoing rather than intermittent patient evaluation. Today these technologies are tried and tested. They are widely available as commercial-off-the-shelf products. Their costs have decreased and functionality increased in the intervening period as the market for them has grown and their component parts commoditised.3 Further adoption of these technologies will continue to be affected by legal, regulatory, privacy, quality, cost and standardisation issues. In terms of this current review, however, it is sufficient to say that the necessary technologies now exist to routinely and reliably monitor the health status of patients in their home for a range of chronic conditions. Table 1 provides an overview of these technologies and their capabilities. HENRY STEWART PUBLICATIONS 1753-3031. Journal of Management & Marketing in Healthcare. VOL. 2 NO. 3. PP 238-252. JULY 2009 239 MMH074.qxd 4/8/09 5:09 PM Page 240 Darkins and Sanders Table 1: Overview of remote patient monitoring technology Area of remote monitoring Aspect monitored Comments Routinely available but issues with lack of standardisation Digital image capture Real-time video into the home Text messaging Pulse, rhythm, respiration, weight, temperature, blood pressure, pO2 Blood glucose, pulse oximetry, INR Clinical images eg wound care Telephone supported and internet protocol (IP) supported SMS, e-mail, secure messaging The 'smart home' Activities of daily living (ADLS) Vital sign data Other clinical data Routinely available but issues with lack of standardisation JPEG, TIFF no DiCom Standard Concerns with privacy and with technology management Concerns with privacy and litigation Uses sophisticated systems to monitor appliances and utilities in the home There is often a certain naivet that exists in relation to adopting new technologies and this is particularly so with respect to health information technologies (HIT). This circumstance is epitomised by the philosophy of 'provide it and they will come' with respect to HIT. It does not follow that the mere availability of HIT and the associated applications, such as described in Table 1, means that they will enter routine practice. The time required for the emergence of a new healthcare technology, developing an evidence base to support it and its ultimate inclusion in routine practice averages 15 years. A key factor influencing whether a particular healthcare technology is widely implemented is whether the requisite patient need exists to justify its adoption. Although this fundamental requirement is often ignored by the technophile advocates of a technology, it should be self-evident, given that the more patients that use a technology, the greater will be the associated revenue to defray the overhead costs associated with adoption. Organisational, cultural and business issues are usually of equal if not greater importance than the elegance of any technology solution under consideration. Successful resolution of these issues follows logically if the underlying patient need that supports a technology's implementation is addressed up front. Which patient care needs support applications involving remote monitoring? Remote patient monitoring technologies are particularly suited to address a range of major healthcare needs. Some of these are explored below. Care of chronic conditions Changes in both human fertility and lifespan are increasing the world's population, particularly those aged over 65.4 Remote monitoring would therefore be useful for caring for people with chronic health-related conditions. In the USA, for example, it is projected that the proportion of 240 HENRY STEWART PUBLICATIONS 1753-3031. Journal of Management & Marketing in Healthcare. VOL. 2 NO. 3. PP 238-252. JULY 2009 MMH074.qxd 4/8/09 5:09 PM Page 241 Remote patient monitoring in home healthcare the population aged 65 years will increase from 12.4 per cent in 2000 to 19.6 per cent in 2030.5 This translates into an expected increase in those aged 65 years from approximately 35 million in 2000 to an estimated 71 million in 2030, with the number of people aged 80 years increasing from 9.3 million in 2000 to 19.5 million in 2030. This increase will place significant burdens on the health system, especially as many of these people will be affected by chronic diseases such as diabetes, chronic obstructive pulmonary disease, chronic heart failure, hypertension, depression and Alzheimer's disease chronic health conditions that presently account for three-quarters of all medical care spending in the USA. In addition to concerns for the elder population, it is vital to consider that the majority (60 per cent) of people with such chronic conditions in the USA are of working age (approximately 65 million people). Unless lifestyle changes occur (healthy diet, exercise, smoking cessation, etc) the incidence and prevalence of these chronic conditions will escalate, translating into ever-increasing costs to the healthcare system and the diminution of available resources. Given the absence, at present, of effective cures for these medical conditions, remote monitoring of patients is a means of limiting disease progression, reducing the frequency of associated complications and lowering the costs of providing care. It is not only the remote monitoring of patients per se that can provide the benefits of home telehealth for populations of patients with chronic disease. Home telehealth can enhance patient-practitioner interactions and enable health promotion as well as disease reduction. Health practitioner shortages The changing population demographics that are resulting in the increased numbers of patients with chronic conditions are also resulting in relatively fewer numbers of practitioners and care workers to provide care. Home telehealth enables practitioners, usually nurses or social workers, to manage a population of patients and provide care/case management more efficiently. More timely assessment of the patient's condition is important with respect to treatment, but a key aspect in implementing these home telehealth technologies is in the opportunity they provide for patient education to promote self-management and lifestyle modification. Table 2 provides examples of the clinical conditions for which home telehealth applications are applicable. Pervasive computing When this technology was originally introduced in the early 1990s, the predominant communication infrastructures available were plain-old telephone lines and coaxial cable which predominantly limited the application to the home environment. However, the growth and availability of wireless communications and information technologies, as simple and ubiquitous as the cell-phone, provide the ability to HENRY STEWART PUBLICATIONS 1753-3031. Journal of Management & Marketing in Healthcare. VOL. 2 NO. 3. PP 238-252. JULY 2009 241 MMH074.qxd 4/8/09 5:09 PM Page 242 Darkins and Sanders Table 2: Patient care needs involving remote patient monitoring technology Condition Aspect monitored Chronic heart failure hypertension Pulse, weight, temperature, blood pressure Diabetes Blood glucose, pulse, weight, temperature, blood pressure Pulse oximetry, pulse, weight, temperature, blood pressure Digital imaging, /- pulse oximetry, pulse, weight, temperature, blood pressure, /- blood glucose Real-time video, the smart home Chronic obstructive respiratory disease Wound care Alzheimer's Depression Real-time video Public health emergencies Digital imaging, /- pulse oximetry, pulse, weight, temperature, blood pressure, /- blood glucose, real-time video Smart home Frail elderly Messaging and e-health information resources Disease management protocols, information on condition and self-management Disease management protocols, information on condition and self-management Disease management protocols, information on condition and self-management Disease management protocols, information on condition and self-management Disease management protocols, information on condition and self-management Disease management protocols, information on condition and self-management Disease management protocols, information on condition and self-management Monitoring of ADLs, emergency response monitor and evaluate a patient wherever they may be. Some examples of the settings in which this expanded capability are being evaluated include: the worksite; retail clinics; nursing homes; intensive care units; long-term care facilities. The ability to monitor patients continuously, not just at home, is known as pervasive computing. The conditions indicated in Table 2 are as suited to management in a pervasive computing environment as with home telehealth. Public health emergencies The ability to provide patient healthcare where and when it is needed is now realisable. The hospital and practitioner's office no longer have to be the self-evident place of care, and given the reduction in both the number of hospital beds and the number of primary care physicians, this is of singular importance. For example, during the past 20 years the number of hospital beds in the USA has diminished, reflecting the movement towards ambulatory care.6 Nonfederal acute care hospital bed numbers fell from 5,784 in 1985 to 4,602 in 2005, with the total bed numbers falling from 1 million in 1985 to less than 767,000 in 2005.7 242 HENRY STEWART PUBLICATIONS 1753-3031. Journal of Management & Marketing in Healthcare. VOL. 2 NO. 3. PP 238-252. JULY 2009 MMH074.qxd 4/8/09 5:09 PM Page 243 Remote patient monitoring in home healthcare In the past, hospital beds could be switched from providing elective care to emergency care in the event of a major public health emergency. It is now unclear whether there would be sufficient hospital beds to deal with the consequent influx of patients associated with public health emergencies such as pandemic flu or bioterrorism. The availability of hospital beds that can be reconfigured to deal with acute changes in healthcare need, such as a public health emergency, is known as 'surge capacity'. In its definition of a surge-capacity benchmark, the US Health Services Resources Agency (HRSA) recommends 500 hospital beds per million of the population. It is clear, therefore, that the above figures suggest that many communities would find it difficult to meet HRSA's benchmark for hospital beds;8 this is reinforced by the routine overcrowding of emergency departments.9 Home telehealth/telecare provides a means to monitor, diagnose and manage patients in other settings and can therefore offer an alternative way to create surge capacity. It has the added benefit of protecting healthcare practitioners who may be in relatively short supply during public health emergencies from exposure to harmful agents/micro-organisms, as well as functionally 'quarantining' individuals who can infect others, without compromising their care. Care of frail elderly In addition to the monitoring of patients' vital signs for healthcare purposes, there are very large numbers of frail elderly who live on their own and face the challenge of continuing to live independently in their own homes.10,11 Monitoring such people is not necessarily a question of managing conditions such as diabetes and heart failure. The predominant issue in this population is whether they can cope with the regular activities of daily living such as washing, cooking, bathing etc. With appropriately placed sensors, this remote patient monitoring technology may be used to determine the regularity with which patients bathe, the length of time they are in bed, whether they are using the bathroom, whether they have fallen, etc. Homes equipped with these technologies are known as 'smart homes'. For these individuals, the smart home provides a means whereby social services, and/or assisted living facilities, can monitor the activities of normal daily living and detect problems that may occur as soon as possible and prevent avoidable deteriorations. In some countries this technology has been introduced as part of social care programmes to help individuals remain living independently in their own homes. This technology is also of great importance for family members who are living geographically distant from elderly relatives. These family members are prepared to invest in such monitoring systems to help ensure their relatives are well and to detect problems when they occur, with the intention of using the systems to facilitate care in the event of a problem being detected. Case histories Two case histories, the first involving social care in the UK and the second in a healthcare setting in the USA, show why organisations have HENRY STEWART PUBLICATIONS 1753-3031. Journal of Management & Marketing in Healthcare. VOL. 2 NO. 3. PP 238-252. JULY 2009 243 MMH074.qxd 4/8/09 5:09 PM Page 244 Darkins and Sanders adopted home telehealth into care delivery and what their strategies are for further expansion from which important underlying principles can be derived. Case study 1: Kent County Council, UK Kent County Council (KCC) in the UK began piloting home telehealth in 2005. The goal was to bridge between health and social care services and help maintain people with complex care needs in the home and community settings. Health and social care services are provided by different bodies in the UK and their services are variably integrated. Deterioration in a person's health can affect their ability to cope and result in avoidable admissions to care. This means there are benefits to be realised from monitoring the health and ability of people with complex care needs using telehealth technologies. The aim of doing so is to maintain their independence and reduce avoidable outpatient and institutional care. The results of the KCC pilot involving the daily monitoring of 250 people show significant benefits.12 Use of telehealth technologies prevented unplanned hospital admissions and enabled patients with complex comorbidities to remain at home and prevented avoidable hospital admissions. Costs associated with acute care were reduced by more than 60 per cent in some patient groups. Interestingly, the initial model that involved primary care physicians and nurses was modified into one that utilised community nurses. Following the success of this pilot, it has been expanded into a wider demonstration project involving other councils around the UK, with a view towards national implementation of this approach to care. Case study 2: The Center for Connected Health, Boston, USA Partners HealthCare (PHC) is an integrated healthcare system in Boston, Massachusetts. PHC was founded in 1994 by Brigham and Women's Hospital and Massachusetts General Hospital. This nonprofit organisation includes primary care and specialty physicians, community hospitals, the two founding academic medical centres, specialty facilities, community health centres and other health-related entities. The population's healthcare needs that PHC serves, the clinical spectrum of services the organisation provides, and the geographic area it covers, led the organisation to outline an innovative future for service development. PHC's vision includes the use of new health information technologies to re-engineer the healthcare services it provides to the population it serves. In implementing its vision for services, PHC established its Center for Connected Health (CCH). Programmes developed by CCH use a combination of remote monitoring, online communications and intelligence, and technology applications.13,14 Pilot studies of this approach have shown that it may lead to improved patient knowledge, engagement and accountability, as well as improved patient-provider communication to improve patient adherence, engagement and clinical outcomes. To substantiate these findings and generate scientific evidence for their wider adoption, CHC has initiated a randomised clinical trial 244 HENRY STEWART PUBLICATIONS 1753-3031. Journal of Management & Marketing in Healthcare. VOL. 2 NO. 3. PP 238-252. JULY 2009 MMH074.qxd 4/8/09 5:09 PM Page 245 Remote patient monitoring in home healthcare involving 200 patients from six primary care practices affiliated to either Massachusetts General Hospital or Brigham and Women's Hospital. PHC is demonstrating an enlightened vision of how there are synergies between elements of health information technology that benefit patient care. It is exploring how the reductions in hospital admissions and efficiencies in the management of patients brought about by home telehealth justify its implementation in circumstances where there is no direct reimbursement for these services. The importance of the studies PHC is undertaking is to provide the evidence to bring about future reimbursement strategies for home telehealth. Brief commentary on case studies These two vignettes cover service provision in both governmental and non-governmental sectors in the UK and USA. These implementations of home telehealth demonstrate some key lessons: an enterprise undertaking to adopt home telehealth is essential; services must be based upon defined patient need; piloting is an essential part of developing and refining a basic model; outcomes information is a critical component to substantiate the model; home telehealth plays a vital role in the integration of care. In summary, a range of remote monitoring technologies are available to monitor patients as part of home telehealth or telecare services that collect and interpret data from either the patient themselves or from the home environment. These technologies can address major societal healthcare needs. Given this environment, what is the current status in terms of the adoption of these technologies? Current adoption of remote patient monitoring applications Following the early introduction of this technology in the form of 'the electronic house call' at the Medical College of Georgia in 1993 as a 'proof of concept' by one of the authors (JHS), and the results of a USbased study undertaken within Kaiser Permanente,15 the possibility of bringing the examination room to the patient was clearly established. Since then, home telehealth has transitioned from the research environment into direct clinical care.16 Currently it has been variably introduced in multiple countries and healthcare settings. Table 3 provides a synopsis of the current state of the implementations as derived from a 2008 European Union analysis. A review of Table 3 shows that the implementation of remote patient monitoring for home health is not as pervasive as the patient need would dictate. As stated previously, this is probably a reflection of the maturing process that most new technologies require for their incorporation into accepted practice. For completeness, there are other applications that are not included in Table 3, such as remote monitoring for sleep studies, remote pacemaker monitoring and remote monitoring for epilepsy. As the HENRY STEWART PUBLICATIONS 1753-3031. Journal of Management & Marketing in Healthcare. VOL. 2 NO. 3. PP 238-252. JULY 2009 245 MMH074.qxd 4/8/09 5:09 PM Page 246 Darkins and Sanders Table 3: International status of remote patient monitoring applications, 2008 Country Current state of implementation Comments England* There were approximately 150,000 new telecare users in England in 2006/07, and a further 161,000 in 2007/08. This may amount to about 3 per cent of the population aged 65 years. Telecare is now embedded in government health and social care policy but it has yet to be fully embedded in mainstream services. Germany A number of health insurances offer large-scale home telehealth service programmes as part of integrated healthcare programmes, while others are involved in a trial phase with a rather low number of participants. There are smart home applications on the market but wider usage is not yet in evidence. Single smart home components are being offered on the market, but networked house solutions and integrated systems have so far not succeeded on the market. A number of pilot implementations have been funded by the Ministry of Health and Welfare, eg a large-scale model project (the so-called Telemedicine Promotion Model Project), addressing municipalities with a view to encouraging mainstreaming of a set of 20 telecare services directed towards older people living in the community. Previously, the ministry had issued a document entitled 'Guidelines for Implementing Information Technology in the Areas of Health, Medical Care and Welfare'. However, some local mainstreaming of passive alarm sensors, ie sensors that do not need to be actively triggered by the service users, has been reported. Japan has a long industrial tradition in smart home technology. From the information available so far, it seems however that these efforts have not been directed towards the support of older people in particular. Some mainstreaming of smart homes with relevance to older people seems to have occurred along with the emergence of a number of pilot and trial projects which have been reported. However, no evidence is available on the level of uptake in terms of end users involved. Smoke detectors have been obligatory in newly-built houses for a number of years already, but these smoke detectors are not integrated with the Preventative Technology Grant funding is given to councils in England with the expectation that they will work with partners in the NHS, housing and district authorities, voluntary and independent sectors and service users and carers in developing services. Some local authorities/ primary care trusts have recently claimed to be providing mainstream telecare services. While home telehealth has not yet reached wide diffusion in Germany, there are many activities and the area is developing quickly. Actual uptake of networked houses or dwellings for older people is largely confined to experimental settings and demonstrators. Japan Netherlands In relation to telecare, ie services and systems that go beyond simple pushbutton alarms, mainstreaming still seems to have to occur, although some piloting and trial activities seem to have happened over recent years. Despite such efforts and the existence of a strong industrial basis when it comes to equipment manufacturing, uptake of more advanced telecare applications has fallen below expectations as of today. Telecare is mainly provided in pilot and trial activities. Fall detectors are hardly used in the country, as (Continued) 246 HENRY STEWART PUBLICATIONS 1753-3031. Journal of Management & Marketing in Healthcare. VOL. 2 NO. 3. PP 238-252. JULY 2009 MMH074.qxd 4/8/09 5:09 PM Page 247 Remote patient monitoring in home healthcare Table 3: Country USA International status of remote patient monitoring applications, 2008 (Continued) Current state of implementation social alarm system. Gas and flood detectors are only used to a limited extent. There are some ongoing trials with add-on sensors, but there is certainly no mainstreamed incorporation of telecare in social care for older people as far as these sensors are concerned. Smart home systems are available across the whole country. In 2006 and 2007 together about 350 projects were granted financial support for smart home and related telecare technology. Another 250 were expected for 2008. On average, each project involves 90 homes, which means that by the end of 2007 some 40,000 homes had been equipped with smart home and telecare technology. It is estimated that this number will have increased by 22,500 homes by the end of 2008, which means that 3 per cent of the population aged 65 lives in houses with smart home or telecare technology. 'Telecare' services are provided by a range of providers including medical/clinical practice sites, hospitals and social service providers, both public and private. The availability of services varies from state to state, with little or no coherence in application or utilisation. There are no data available on the extent of takeup, although it seems to vary a lot across the country. To date, the Veterans Administration healthcare system seems to be the main provider of telecare services with an independent living focus, even though the main focus of its remote support/monitoring has so far been on telehealth. There appear to be numerous trials of smart homes technologies occurring nationwide, many of which are driven by industry funding provided to research institutions. Developments are driven by engineering R&D efforts in collaboration with social service organisations, medical rehabilitation practitioners and consumer supply and demand. There seems to be no major mainstreaming to date, so very low take-up in practice. Comments the ones that are commercially available are generally not considered sufficiently easy to use. There has been an overall increase in interest in telecare, with the emphasis/focus apparently more on healthcare than social care in a wider sense. Source: European Commission ICT & Ageing European Study on Users, Markets and Technologies. * Available at: http://www.ict-ageing.eu/?page_id 397 (accessed 2nd January 2008) Available at: http://www.ict-ageing.eu/?page_id 403 accessed 2nd January 2008) Available at: http://www.ict-ageing.eu/?page_id 399 (accessed 2nd January 2008). Available at: http://www.ict-ageing.eu/?page_id 270 (accessed 2nd January 2008). Available at: http://www.ict-ageing.eu/?page_id 401 (accessed 2nd January 2008). HENRY STEWART PUBLICATIONS 1753-3031. Journal of Management & Marketing in Healthcare. VOL. 2 NO. 3. PP 238-252. JULY 2009 247 MMH074.qxd 4/8/09 5:09 PM Page 248 Darkins and Sanders remote monitoring of pacemakers technically entails monitoring a device rather than a patient, it is not further considered here, although the convergence of remote pacemaker/defibrillator monitoring will be considered in the discussion below. Table 3 also indicates the implementation of home telehealth into the routine care environment of the US Veterans Health Administration (VHA). VHA is a large integrated healthcare system that provides care for over 5.5 million veteran patients annually. The rising burden of chronic disease within this population that is rapidly aging addresses the undertaking of the organisation to provide care for its patients with longterm care needs in the least restrictive possible setting. VHA has systematised the clinical, technological and business aspects necessary to develop a sustainable programme that demonstrably shows positive outcomes from the use of this technology.17 The results of providing care to over 17,000 patients show reductions in hospital admission and bed days of care with high levels of patient satisfaction. VHA now has implemented remote patient monitoring into its routine services and will probably serve as a model (as its hospital and ambulatory care HIT implementations have done) for other healthcare delivery systems to follow in implementing home telehealth. Factors affecting implementation of remote patient monitoring in the home Given the major areas of health and social care needs that remote monitoring of patients can address and the relative maturity of both technology and applications to support this, the implementation thus far, as previously noted, has been limited. There has been no systematic study involving the implementation of remote monitoring technologies on a national scale that provides detailed data on the factors that promote or inhibit the uptake of this technology. Table 4 describes the opinions of the authors with respect to these questions. In addition to the specific issues highlighted in Table 4, there are currently more general barriers to the implementation of IT within healthcare. Many of these are not insurmountable but rather require systematic approaches and solutions as follows: 248 Many programmes tend to focus more on the technology and not the patient requirements and clinical processes that drive the need for the technology. This results in an unbalanced focus. The solution is an organisational one in which IT is seen as a tool, not a goal.18 A general issue within healthcare is that the patient is not the direct customer. This makes it challenging to ensure there are the necessary incentives for patients to engage and functionalities incorporated into the technology to meet the needs of the patients, not just the providers. Technologies that enhance the ability of the patient to selfmanage their condition will significantly resolve many of the problems related to non-compliance and disease progression that are faced today. HENRY STEWART PUBLICATIONS 1753-3031. Journal of Management & Marketing in Healthcare. VOL. 2 NO. 3. PP 238-252. JULY 2009 MMH074.qxd 4/8/09 5:09 PM Page 249 Remote patient monitoring in home healthcare Table 4: Factors influencing remote patient monitoring technology Area of remote monitoring Facilitators General issues applicable to all applications Senior leadership commitment Integrated healthcare system Culture that accepts change Defined population for care Collaborative inter-professional working arrangements Quality assurance arrangements Clinical risk management capabilities Suitable clinical champions Computerised patient record system (CRS) Standards for upload of data onto CRS Adequate and supportive information technology division Adequate technology platform with backup and redundancy Good clinical contracting department Reimbursement/allocation system of funding to support programmes Patient privacy* issues addressed Technology non-obtrusive and intuitive to use Connectivity into the home using multiple telecommunications modalities The availability of decision-support technology for initial assessment and analysis of data streams so as to be able to direct the information to the appropriate provider Data standards Incorporation into routine clinical processes Clear model of care Suitable peripheral devices in the home Data standards Incorporation into routine clinical processes Clear model of care Suitable peripheral devices in the home Clinical standards Incorporation into routine clinical processes Clear model of care Suitable image capture devices in the home Adequate connectivity (preferably internet protocol) Clinical workload capture system for video visits Good usability for patients and staff Clear model for clinical care Privacy and confidentiality issues covered Clear business model Worked through sensitivity and specificity requirements of alerts No patient privacy concerns Affordable technology and minimal retrofitting Vital sign data Other clinical data Digital image capture Real-time video into the home Text messaging The 'smart home' *Kaplan, B. and Litewka, S. (2008) 'Ethical challenges of telemedicine and telehealth', Cambridge Quarterly of Healthcare Ethics, Vol. 17, No. 4, pp. 401-416. Hensel, B., Demiris, G. and Courtney, K. (2006) 'Defining obtrusiveness in home telehealth technologies: A conceptual framework', J Am Med Inform Assoc, Vol. 13, No. 4, pp. 428-431. The way in which home telehealth programmes are marketed is frequently inadequate. Marketing strategies often ignore the patient experience and focus on the practitioner. The solution is to adopt a patient-centric mission for a programme, aligning a clear marketing strategy to this and then ensuring adequate personnel and resources for marketing are available. At the outset, pilot programmes often rely on informal relationships to develop the necessary clinical, technology and business support to HENRY STEWART PUBLICATIONS 1753-3031. Journal of Management & Marketing in Healthcare. VOL. 2 NO. 3. PP 238-252. JULY 2009 249 MMH074.qxd 4/8/09 5:09 PM Page 250 Darkins and Sanders get the programme up and running. Typically, the associated problemsolving strategy to deal with issues that arise is to enable individual clinicians to develop bespoke solutions as part of the scope-ofpractice. Although this is of practical advantage, in a pilot it creates unacceptable practice variations. The solution is that a definitive programme must have explicit processes and procedures to support them, and the requisite organisational buy-in. Programmes often run into trouble because they fund the information technology infrastructure without linking its growth and evolution to revenue generation and/or future revenue potential for the programme. They thereby create an overhead cost that is not clearly linked to the underlying business case for the programme. Consequently, the technology infrastructure is not compatible with the programme needs. The solution is to ensure that associated clinical data systems provide data sets that link outcomes to the relevant technology infrastructure and to align incentives for practitioners to adoption of the programme.19 Delivery of acute services with home telehealth solutions poses medico-legal problems and logistic issues related to how emergency care should be delivered (eg care linked with cardiac pacemaker data). The solution is to focus on chronic care and either avoid acute care, or limit acute services to those emergency services which can be readily provided into the home. However, it is clear that home telehealth has the capability to identify at its earliest stage an exacerbation of a chronic disease process, and thus avoid the natural progression to an acute state. Implications and conclusions The authors believe that, in parallel with other service industries (eg banking, commerce and entertainment) that have successfully used telecommunications and information technologies to conveniently bring their services to the consumer when they need it, the healthcare service industry can now do the same. But the implications are much more important than simply offering convenience, reduced travel time, decreased costs or enhanced productivity. It is clear that the examination room needs to be where the patient lives and works, not where the practitioner does. Taking an individual's blood pressure or their peak expiratory flow rate in 'their' exam room will give a much better physiological assessment than in their doctor's office. Secondly, being able to evaluate an individual's vital signs and other parameters on a more continuous basis will provide a personal profile of what is normal/abnormal for that individual, thus precluding the potential inaccuracies and misdiagnoses created by measuring that individual versus a population norm. This offers the opportunity to identify the presence of an abnormal state at its earliest and most treatable stage. A woman who has a routine blood pressure of 95/65, who over a period of 3-5 years is noted to have her blood pressure slowly increase to 115/80, cannot still be told that she has a normal blood pressure, even though for the population as a whole it would be considered normal. 250 HENRY STEWART PUBLICATIONS 1753-3031. Journal of Management & Marketing in Healthcare. VOL. 2 NO. 3. PP 238-252. JULY 2009 MMH074.qxd 4/8/09 5:09 PM Page 251 Remote patient monitoring in home healthcare This paper has described many practical issues that require resolution. In the authors' opinion, the major issues, or opportunities, that need to be addressed are: creating an incentive for both the patient and practitioner to adapt and accept a re-engineered healthcare system; causing the patient, with real-time information regarding their health status, to assume much greater responsibility for their own health; assuring the availability of a broadband communication infrastructure to all sectors of society; and developing accurate medical decision support algorithms to process and analyse the vast amount of information flow associated with this emerging area of care delivery. The current situation with remote patient monitoring for home health is reminiscent of the immediate period when early adopters in the office used the word processor, spreadsheet and database. It was clear that everything was going to change; in retrospect the only issue was when the 'tipping point' would occur. So it seems now with remote monitoring of patients in home health. Note The contribution of Adam Darkins does not reflect the position or views of the US Department of Veterans Affairs. References 1. Department of State and the Department of Health and Human Services, National Institute on Aging, National Institutes of Health (2007) Why Population Aging Matters: A Global Perspective, US Government Printing Office, Washington, DC. 2. Winters, J. and Herman, W. (1999) Proceedings of The Workshop on Home Care Technologies for the 21st Century, Rockville, MD, 7-9th April, Catholic University of America, Washington, DC. 3. Park, H., Murray, P. and Delaney, C. (2006) 'Consumer-centered computer-supported care for healthy people', paper presented at the 9th International Congress on Nursing Informatics, held on 11th-14th June, 2006, at Seoul. 4. Kinsella, K. and Velkoff, V. (2001) US Census Bureau. An Aging World: 2001, US Government Printing Office, Washington, DC. 5. US Census Bureau (2008) 'International database. Table 094. Midyear population, by age and sex', available at http://www.census.gov/population/www/projections/natdet-D1A.html (accessed 20 July 2009). 6. Bazzoli, G. J., Brewster, L. R., May, J. H. and Kuo, S. (2006) 'The transition from excess capacity to strained capacity in US hospitals', Milbank Quarterly, Vol. 84, No. 2, pp. 273-304. 7. American Hospital Association (2005) AHA Hospital Statistics: 2005, Health Forum, Chicago, IL. The total for 2005 is based on the authors' tabulation of 2005 AHA annual survey data. 8. Institute of Medicine (2006) Hospital-Based Emergency Care: At the Breaking Point, National Academies Press, Washington, DC. 9. Bazzoli, G. J., Brewster, L. R., Liu, G. and Kuo, S. (2003) 'Does US hospital capacity need to be expanded?', Health Affairs, Vol. 22, No. 6, pp. 40-54. HENRY STEWART PUBLICATIONS 1753-3031. Journal of Management & Marketing in Healthcare. VOL. 2 NO. 3. PP 238-252. JULY 2009 251 MMH074.qxd 4/8/09 5:09 PM Page 252 Darkins and Sanders 10. Jacobzone, S. (2000) 'Coping with aging: International challenges', Health Affairs (May/June), Vol. 19, No. 3, pp. 213-225. 11. Anderson, G. F. and Hussey, P. S. (2000) 'Population aging: A comparison among industrialized countries', Health Affairs (May/June), Vol. 19, No. 3, pp. 191-203. 12. Available at: http://www.institute.nhs.uk/option,com_joomcart/Itemid,26/main_page,document_ product_info/products_id,338.html (accessed 1st April 2009). 13. Heinzelmann, P. J., Kvedar, J. C. and Kibbe, D. C. (2008) 'Beyond EHRs: How technology can help you treat chronic illness', Family Practice Management, Vol. 15, No. 3, pp. 29-32. 14. Center for Connected Health, Partners and Beyond Progress Report 2008, Boston, MA, Available at: http://www.connected-health.org/media/214782/2008%20progress%20report.pdf (accessed 1st April 2009). 15. Johnston, B., Wheeler, L., Deuser, J. and Sousa, K. H. (2000) 'Outcomes of the Kaiser Permanente tele-home health research project', Archives of Family Medicine, Vol. 9, No. 1, pp. 40-45. 16. Hebert, M. A., Korabek, B. and Scott, R. E. (2006) 'Moving research into practice: A decision framework for integrating home telehealth into chronic illness care', International Journal of Medical Information, Vol. 75, No. 12, pp. 786-794. 17. Darkins, A., Ryan, P., Kobb, A., Foster, L. et al. (2008) 'Care coordination/home telehealth: the systematic implementation of health informatics, Home Telehealth and Disease Management to support the care of veteran patients with chronic conditions', Telemedicine and e-Health, Vol. 14, No. 10, December. 18. Diamond, C. and Shirky, C. (2008) 'Health information technology: A few years of magical thinking?', Health Affairs, Vol. 27, No. 5, pp. w383-w390. 19. Deloitte Center for Health Solutions (2008) 'The medical home: Disruptive innovation for a new primary care model', available at: http://www.deloitte.com/dtt/cda/doc/content/us_chs_ MedicalHome_w.pdf (accessed 14th July 2008). 252 HENRY STEWART PUBLICATIONS 1753-3031. Journal of Management & Marketing in Healthcare. VOL. 2 NO. 3. PP 238-252. JULY 2009 Health Marketing Quarterly, 30:246-262, 2013 Copyright Taylor & Francis Group, LLC ISSN: 0735-9683 print/1545-0864 online DOI: 10.1080/07359683.2013.814503 Development of a Quality Ranking Model for Home Health Care Providers JUSTIN W. GRESSEL Department of Marketing, College of Business Administration, The University of Texas-Pan American, Edinburg, Texas This research aims to increase transparency and simplify consumer decision-making regarding the selection of a home health care provider. Currently, quality information on home health care providers is fragmented and difficult to interpret. In this study, a quality-ranking model is developed by selecting multidimensional quality indicators across multiple sources and respective weights using expert judgment. Given the weights and providers' performance on each quality indicator, a composite score is calculated that summarizes a home health care provider's overall quality level. This quality information empowers consumers to narrow their search and select the best-performing, most efficient providers. KEYWORDS health-care quality, ranking, composite score Recently, there has been a flurry of calls for transparency in health care related performance (e.g., Probst, 2010; S. M. Evans, Lowinger, Sprivulis, Copnell, & Cameron, 2009; Jessee, 2007; M. Evans, 2007) to increase the quality of decision-making and to improve overall health care quality. With greater transparency and tangible quality indicators, consumers can take on a greater role in ensuring that the marketplace operates efficiently by consciously seeking out those providers that offer the right balance of quality outcomes and cost while preventing the appearance and growth of lowquality providers (Armstead, Elstein, & Gorman, 1995). The pressure of informed consumers should force health care professionals to be more competitive by reducing inefficiency and improving poor health outcomes (Evans, 2007). Studies support this assertion by showing that there are Address correspondence to Justin W. Gressel, Department of Marketing, College of Business Administration, The University of Texas-Pan American, 1201 West University Drive, Edinburg, TX 78539. E-mail: gresseljw@utpa.edu 246 Development of a Ranking Model for Home Health Providers 247 improvements in health care quality in the hospital setting when public reporting and financial incentives are combined (Lindenauer, Remus, Roman, Rothberg, & Benjamin, 2007). Currently, quality is measured and reported for hospitals, health plans, nursing homes, home health agencies, and physicians in the United States with the purpose of creating quality benchmarks, influencing consumer decision-making, and determining provider reimbursement (Werner, Bradlow, & Asch, 2007). A recent study estimates that more than 50 organizations are dedicated to researching hospital and physician quality indicators and that roughly 60 million people search the Internet for health care information on a daily basis, with roughly 5 million people per month visiting just one of the many physician evaluation websites (May, 2008). Despite the high number of consumers accessing quality information, many consumers who seek out health care information do not use it. A survey conducted by the Kaiser Family Foundation (2006) found that only half of those that accessed quality data for health plans, hospitals, or doctors used the information to make a health care-related decision. The difficulty of accessing and understanding relevant health care quality information remains an obstacle for consumers. Currently, quality information on health care providers is spread across various data sources (M. Evans, 2007), some of which are inaccessible in that consumers are not aware of them, they have to be formally requested, or they require purchase of data. The fragmentation of health care information or \"scatter\" of relevant health care information causes searchers to retrieve incomplete information and end their searches prematurely (Bhavnani & Peck, 2010). Another obstacle to consumers in accessing health care information, even when the information is contained within one source, is the sheer amount of the available data, which makes it difficult for consumers to synthesize and summarize the information. In addition, the overly technical nature of quality indicator labels and descriptions makes it hard for the average consumer to understand (M. Evans, 2007). The present study focuses on home health care (HHC), a particular area of health care that constitutes a substantial portion of the health care market and for which there are few organizations that publicly report quality outcomes. HHC refers to medical services, such as occupational therapy and nursing services, which are offered in a home setting. HHC expenditures in 2009 reached $68.3 billion in 2009, rising by 10% over the previous year (Centers for Medicare and Medicaid Services [CMS], 2010). The largest trade association for HHC in the United States, the National Association for Home Care and Hospice (NAHC), estimates that over 12 million people currently receive HHC services from more than 33,000 HHC providers (NAHC, 2010). The HHC industry has experienced explosive growth over the past decade, and with that growth there are concerns that some providers are unqualified, poor performers, or even fraudulent. In a 2007 interview with the New York 248 J. W. Gressel Times, then state attorney general of New York said, \"It's a burgeoning industry that is ineffectively regulated, that is expensive for taxpayers, and can victimize consumers\" (Confessore & Kershaw, 2007, para. 8). Therefore, given the increasingly larger number of HHC providers, the lack of regulation within the HHC industry, and the relative lack of an informative source of health care quality, this study aims to improve accessibility, transparency, and understanding of HHC quality information for consumers. BACKGROUND Evaluating HHC Because 65% of HHC payments originate from Medicare and Medicaid (CMS, 2010), the Centers for Medicare and Medicaid Services (CMS) have a vested interest in assessing HHC provider quality. In addition, rising health care costs and utilization of medical services have influenced efforts to control spending growth and improve service quality (O'Leary & Walker, 1994). For this reason, CMS developed a comprehensive assessment of health care outcomes and treatment quality. This survey, known as OASIS (Outcome and Assessment Information Set), currently consists of 41 items and is administered only to Medicare-certified HHC providers. A select amount of the OASIS information can be accessed through Home Health Compare, a website-based comparison tool created by CMS that allows consumers to search for HHC providers within a specified geographic region. Upon inputting a location, consumers can choose up to three HHC providers for comparison and examine their respective performance across various quality indicators (see Appendix A). The data is adjusted for differences between providers, such as patient demographics, case mix weight,1 and primary diagnosis group. While helpful in making specific outcome-based comparisons between a few providers, Home Health Compare is unwieldy in understanding overall quality and in comparing multiple HHC providers. The quality information displayed on Home Health Compare consists of 23 quality indicators, an overwhelming number for consumers unfamiliar with home health who are trying to assess overall quality. Zaslavsky, Shaul, Zaborski, Cioffi, and Clearly (2002) stated that the proliferation of numerous quality indicators makes the information more difficult for consumers to interpret, prioritize, and understand. In addition, Home Health Compare only allows consumers to view side-by-side comparisons for up to three HHC providers at a time. The limited comparison set makes it difficult to determine the highest quality HHC providers when there are many providers, potentially hundreds, within an area. It would be helpful for a comparison tool that allows consumers to narrow down the consideration set to a more manageable size before making comparisons. Development of a Ranking Model for Home Health Providers 249 Another limitation of Home Health Compare is that although it allows for average performance-based comparisons, it does not include information on other relevant quality indicators, such as cost, efficiency, or compliance with laws and regulations. Other health care quality assessments, such as CMS' ranking of nursing homes and AETNA's ranking of doctors, incorporate other quality dimensions, such as cost and survey inspections, which leads to a more multidimensional assessment of quality. Sources for such data do exist for HHC providers, but they are inaccessible to the average consumer because they require the purchase of data (i.e., Medicare claims), formal requests to state agencies (i.e., survey inspection results), or directed Internet searches to known accrediting agencies. Besides CMS' Home Health Compare, there are a few organizations that provide benchmarking statistics to HHC providers, such as OCSHomeCare and Strategic Health care Programs. However, the data are targeted to HHC providers, as opposed to consumers. In addition, the data are not publicly available and the research methodology is not specified. OCSHomeCare creates various annual rankings, but the description of the data methodology is vague, only mentioning that the methodology incorporates quality of care, quality improvement, and financial performance. Regarding ranking models, they have become popular within the health care industry in synthesizing and processing relevant quality metrics for the public and ultimately outputting a number that represents overall quality. These models have been created by government agencies (e.g., Medicare's Hospital Compare and Nursing Home Compare), health services companies (e.g., Aetna's Aexcel and CIGNA Care Network), and thirdparty organizations (e.g., U.S. News Best Hospitals and Consumer Reports Health.org). However, no such ranking model exists within the HHC industry. Due to the lack of a user-friendly HHC ranking model that uses a comprehensive set of quality indicators, the present study lays out the methodology for developing a HHC ranking model that is multi-dimensional in nature and that simplifies the evaluation and decision-making process by combining the quality indicator dimensions into a single composite score. Specifically, this research aims to (a) increase accessibility and transparency of quality information by pooling relevant sources of information on local HHC providers into one domain and (b) simplify the processing of HHC quality information by displaying the overall ranking of HHC providers within a given geographic area. Quality Indicators Health care quality indicators typically relate to outcomes-based and process-based measures. From a consumer perspective, the use of outcomesbased measures is consistent with value-based purchasing, which involves making market decisions based on quality, costs, and the efficient use of resources (Lied, Malsbary, Eisenberg, & Ranck, 2002). The emphasis in outcomes-based measures is on the expected benefits to the consumer as 250 J. W. Gressel opposed to processes that presumably lead to outcomes. However, processbased measures are also relevant because superior processes should theoretically translate into superior outcomes, and the potential time lag between the enactment of processes and respective outcomes may cause the superior service to not be registered as a superior outcome for a significant period of time (Shaughnessy, Kramer, Hittle, & Steiner, 1995). In that case, a superior HHC provider at a particular point in time may not appear as good as it is outcome-wise because of the time required for the outcome to be manifested. Shaughnessy et al. (1995) acknowledged that both outcome-based and process-based measures have strengths and weaknesses and recommend that quality indicators that combine outcome and process measures are beneficial in the overall context of quality improvement. Another consideration for quality indicators is that they should be multidimensional in nature, giving a well-rounded view of a provider's quality from the patient's perspective, such as patient satisfaction with the provider, quality of life, and other structural measures (Sangl, Saliba, Gifford, & Hittle, 2005; Clemes, Ozanne, & Laurensen, 2001). Furthermore, the quality indicators should be independent of each other, not redundant or highly correlated with existing measures (Shwartz, Burgess, & Berlowitz, 2009). A study that examined service quality factors within the health care context found that relevant factors of health care provider quality are: reliability, tangibles, assurance, empathy, food, access, outcome, admission, discharge, and responsiveness (Clemes et al., 2001). Three of the factors (tangibles, food, and access) do not align well with assessing HHC quality because they relate to physical facilities and tangibles that are irrelevant in the context of HHC. The other factors, however, do apply to quality of service received by HHC providers. As evidence of the limitations of Home Health Compare, it addresses only outcome-based measures, leaving the other health care provider dimensions (i.e., reliability, assurance, empathy, admission, discharge, and responsive) unmeasured. Creation of a Composite Score An important consideration in displaying quality information is that it must be meaningful and understandable to consumers (Friedman, 1995), whether consumers are patients or health care related businesses. A popular methodology for simplifying consumers' understanding is the creation of composite scores that weight individual quality indicator ratings and sum the results to create a scalar that represents overall quality. Schwartz et al. (2009) stated: A composite score provides a useful summary of the extent to which top management has created a culture of quality and designed processes to ensure quality throughout the organization. Payers and consumers can use composite scores to compare performance across providers, and policy makers can use them to design incentives to encourage high quality, cost-efficient care. (p. 235) Development of a Ranking Model for Home Health Providers 251 Although composite scores can be controversial because of the selection and weighting of quality indicators (U.S. Government Accountability Office, 1994), their popularity with consumers across various industries reflects consumers' reliance on them for synthesizing and summarizing overall quality levels. The implicit assumption in creating a composite scale is that overall quality is a function of the quality indicators used in the model and that the weights represent the overall contribution of each indicator to overall quality. Thus, the composite score can be seen as a formative construct (Shwartz et al., 2009). Because there is no overall quality variable upon which the quality indicators can be regressed to determine their contribution, the weights cannot be determined through statistical estimation. Nonetheless, considerable care should be taken in the weighting, aggregation, and assignment of ranks because these decisions may significantly influence consumers' evaluations of provider quality and may have important policy implications (O'Brien et al., 2007). To overcome the inability to statistically estimate weights, researchers have proposed several options to assign weights to quality indicators, among them: giving equal weights, estimating the weights through an expert panel, or by assigning weights based on the relative variance of the respective quality indicators (Schwartz et al., 2009). To ensure fair comparison and benchmarking among providers, quality data must be standardized and risk-adjusted (Friedman, 1995). Standardization is important to adjust for different units of measurement and varying levels of dispersion across quality indicators so that, assuming equal weights, no quality indicator is more influential than another. Similarly, the quality indicator data must be similarly valenced so that positive (negative) ratings for all quality indicators are in the same direction. Risk adjustment allows for fair comparisons among providers by accounting for factors beyond the control of providers that may affect outcomes, such as the patient's age, primary diagnosis, and severity of illness. Within the health care field, risk adjustment has become a research field in and of itself due to the complexity and difficulty of accounting for relevant risk adjusters, collecting the data, and statistically making the adjustments (Jencks, 1995; Zimmerman et al., 1995; Shaughnessy et al., 1995). For the purpose of parsimony and focus, the present study will not delve into the specific methodology of how data should be risk-adjusted. METHODOLOGY Selecting Data Sources and Quality Indicators The first step in developing the ranking model involved identifying relevant data sources and quality indicators. For this, contact was made with HHC providers, consultants, government employees, accrediting agencies, and NAHC both by phone and by e-mail. They were asked what variables are relevant indicators of a HHC provider's quality, as well as what data sources 252 J. W. Gressel have relevant quality information. A conscientious effort was made to include sources and indicators that were multi-dimensional, providing glimpses of quality from different perspectives. In this initial stage, three primary sources of HHC quality information were identified: Home Health Compare, Medicare Claims data, and state survey inspections. These data sources shed light on different aspects of HHC quality, and therefore the integration of the different data sources should provide for a more well-rounded determination of overall quality than any single source. For example, Home Health Compare consists of patient outcome-based results and will soon also display patient satisfaction scores; Medicare claims data consists of cost and efficiency statistics; and state survey inspections provide data on accreditation status and on violations that show deficiencies in following established protocol. Home Health Compare organizes various indicators of patient health care outcomes into five categories or constructs: (a) managing daily activities, (b) managing pain and treating symptoms, (c) treating wounds and preventing pressure sores, (d) preventing harm, and (e) preventing unplanned hospital care. Each category consists of multiple items, from two to eight, and the data is in the form of proportions (e.g., percent of patients showing improvement). For the purpose of this study, the items within each category (seen in Table 1) are averaged to create an overall percentage for each TABLE 1 Five Categories of Home Health Compare and Respective Items Categories Managing daily activities Managing pain and treating symptoms Treating wounds P