Case Study High-Performing Health Care Organization • April 2010 Lessons from a Health Information Technology Demonstration in New York Nursing Homes S hana L ieberman K linger and S cott W hite The mission of The Commonwealth ABSTRACT: The New York State Nursing Home Health Information Technology (HIT) Fund is to promote a high performance Demonstration Project is a publicly subsidized initiative to implement comprehensive, health care system. The Fund carries point-of-care electronic medical records in 20 New York City nursing homes. Because of out this mandate by supporting independent research on health care an innovative union–employer partnership, direct-care staff of the homes were heavily issues and making grants to improve involved in the planning process. Union employees were assured upfront that no layoffs health care practice and policy. Support would result from HIT implementation, and training was a high priority in vendor selec- for this research was provided by tion. All participating homes successfully replaced paper records with electronic ones, and, The Commonwealth Fund. The views after the intensive pre-implementation planning period, it took less than six months on presented here are those of the authors average for facilities to make this transition. Despite this shared success, variation existed and not necessarily those of The Commonwealth Fund or its directors, between homes regarding: 1) organizational aims for adapting HIT; 2) the technology’s officers, or staff. perceived or real effects; and 3) implementation of quality improvement efforts as a result of newly available data.      For more information about this study, The National Context please contact: President Obama’s economic stimulus legislation, the American Recovery and Shana Lieberman Klinger, M.A. shanalk2003@yahoo.com Reinvestment Act of 2009 (ARRA), committed unprecedented new resources toward creating a health information technology (HIT) infrastructure for all U.S. health care sectors, establishing HIT as a federal priority.1 ARRA issued the national goal of “utilization of an electronic health record for each person in the To learn more about new publications United States by 2014.”2 With that goal in mind, this case study sheds light on when they become available, visit the HIT funding, deployment strategies, and outcomes in the long-term care nursing Fund's Web site and register to receive Fund email alerts. home sector in the New York City area, in the hope that the nearly 1.5 million Commonwealth Fund pub. 1380 Americans with complex health care needs who currently reside or are rehabili- Vol. 42 tated in nursing homes will benefit.3 2T he  C ommonwealth F und New York’s Nursing Home HIT Demonstration: Background “It is essential the homes utilize the available The New York State Nursing Home Health technology to enhance the quality of resident care. Information Technology Demonstration Project is a In turn, such technological advances will improve the publicly subsidized initiative implementing compre- working environment, advantaging the Employers’ hensive, point-of-care, clinician-centric HIT systems in ability to retain and recruit direct-care staff.” 20 New York City–area nursing homes. The demon- In the Matter of the Interest Arbitration between stration has simultaneously created an electronic medi- Southern New York Associates, L.L.C., et al., and cal record (EMR) for every resident in participating 1199SEIU United Health Care Workers East, March 2006. homes (totaling 4,467 beds) while automating the workflow and record-keeping for every direct-care “the acquisition of electronic monitoring and data col- staff person and clinician. An innovative union– lection equipment; professional training of staff mem- employer partnership included direct-care staff at all bers in the use of such electronic equipment; . . . revi- stages of program planning, and union staff were sion of computerized systems and network communi- assured that no layoffs would result from HIT cations; automated assessments, care plans,” and implementation. related tasks.5 Adoption of HIT in these homes has been The arbitration award defined a funding mecha- marked by variations in how the technology has been nism to establish a research and demonstration project used and how it has affected the organizations. from within the employer–union structure. However, Comprehensive research evaluations will define and the mutually beneficial goals of improved quality of measure these variations, and will address questions care and workforce retention laid the foundation for about the impact HIT adoption has had on quality of labor and employers to powerfully partner in seeking care for nursing home residents, workforce retention, public funding. Their representatives together labor and employment relations, organizational cul- approached the New York state legislature with a pro- ture, and the financial impact on homes. posal, and in late 2006 the legislature authorized a The demonstration originated in a unique part- grant of $9 million to support the adoption of HIT in a nership between unionized labor and nursing home group of nursing homes in the New York City area. employers centered on quality improvement. In 2002, The oversight committee was charged with directing the union representing workers in 95 percent of nurs- and monitoring the project. ing homes in New York City, 1199SEIU United Healthcare Workers East (1199SEIU), and 140 primar- HIT Vendor and Product Selection ily for-profit nursing homes in the greater New York The oversight committee supervised a rigorous process metropolitan area, agreed in their collective bargaining for selecting an HIT vendor, requiring that HIT tech- agreement to establish the three-member Quality Care nology be “clinician-centric” and available at “point of Oversight Committee.4 care.” The goal was to completely automate nursing The oversight committee was originally home clinical and workflow functions, including intended to study and review practices designed to patient care notes at the bedside, physician orders, improve the quality of resident care in nursing homes. medication administration records, care plans, nursing However, it remained inactive until March 2006, when instructions, and certified nursing assistant (CNA) the impartial arbitrator and neutral chair of the com- assignments. These requirements were innovative, mittee directed it, as part of an arbitration award, “to since much of then-existing information technology develop and commence research and demonstration deployed in nursing homes was directed at administra- programs” in a sample of nursing homes, including tive tasks, rather than automation of patient care and daily direct-care workflow. L essons from a H ealth I nformation Technology D emonstration in N ew York N ursing H omes 3 Another criterion in selecting a vendor was its can access different sets of information than CNAs; capability of providing each home with comprehensive residents’ physicians can see critical information at implementation support, including effective staff train- any time and place via the Internet.) ing. This was crucial, because the HIT demonstration The SigmaCare system offers extensive capabil- design required that all elements of the software pack- ities for analyzing information for quality improve- age be implemented fully throughout the entire home ment via “dashboards,” computer screens that summa- and used by all direct-care staff. An important factor in rize data in real time. Data can be categorized at the ultimately choosing the vendor was its commitment to level of resident, unit, or facility-wide. The system helping homes analyze and restructure their daily work- reduces or eliminates the need for separate paper cop- flow as part of HIT pre-implementation, and informing ies of orders, records, or care plans. staff of the positive benefits of the new technology. In late 2006, the oversight committee awarded Terms of the State Subsidy the HIT demonstration technology contract to New The oversight committee structured the contract York City–based eHealth Solutions, Inc. The selected between the HIT vendor and the homes so that HIT product system, SigmaCare, consisted of a centrally implementation costs, such as software, hardware, managed Web-based software application that enables installation, and staff training, would be merged with staff to use durable handheld devices (personal digital all subsequent yearly operating and maintenance costs assistants, or PDAs), and computer laptops and desk- over a five-year contract period. This created a fixed tops to record all details relating to the care provided per-bed, per-day rate for the entire course of the agree- to residents in real time, manage workflow, and main- ment. The fixed rate was two-tiered to recognize effi- tain a complete electronic medical record (EMR) for ciencies of size: homes with more than 220 beds each resident.6 Information is stored in an offsite received a slightly lower rate than smaller homes. With secure data center. Different categories of caregivers this strategy, homes could calculate a predictable amount have access to types of information in the resident into their operating costs during the contract period. record that they need specifically (e.g., onsite nurses SigmaCare Software What It Is SigmaCare is a secure wireless mobile electronic medical record system designed specifically for nursing home staff based on their workflow. What It Does • Automates physician orders, medication administration records, treatment administration records, care plans, progress notes, nursing instructions, and assignments for nursing assistants. • Allows clinicians to monitor real-time quality measures. Reports on clinical exceptions. • Provides clinical decision-support features, such as medication interactions. • Provides task reminders and customizable drop-down menus with care protocols to assist caregivers in daily workflow. • Gives different levels of facility staff access to their appropriate “dashboard”—summary information about resident(s), unit(s), or the whole facility. • Includes interoperability components for potential collection and transfer of data with any Regional Health Information Organization (RHIO), and exchange of information with other providers in the long-term care continuum, such as labs or hospitals. 4T he  C ommonwealth F und Subsidized Start-Up and Adoption Period • Coordinating with the 1199 Training Fund to The New York HIT demonstration contract was create a labor–management committee to designed so that the first 17 months of the 60-month guide the preparation and early HIT imple- agreement (28%) was paid for in full with the grant mentation in each home. subsidy from New York State via an intermediary • Committing to no layoffs of any union staff as administrative designee. Thus, homes were offered a a result of improved efficiencies resulting “free” start-up and adoption period, removing hurdles from HIT implementation. Reflecting the orig- related to obtaining initial capital. However, since they inal union–employer goal of increased quality were required to commit to staying with the contract of care and workforce retention, this was a de for the remaining 43 months, homes would eventually facto agreement aimed at reinvesting potential pay some part of that initial cost over time as part of savings in staff time back into quality of care the average of all costs. The fixed per-bed, per-day for residents, or to reduce the need for agency rate allowed a 200-bed nursing home, for example, to (non-union, part-time) staff. predict costs of about $17,700 per month (plus taxes) for the five years of the contract, with the first 17 • Implementing HIT software throughout the months paid by the subsidy. A portion of these expen- entire home and for use by all direct-care staff. ditures were eligible for reimbursement as capital costs Partial implementation (of selected clinical under Medicaid. features or in some units) was not permitted. • Cooperating with researchers connected with Administrative Coordination the project. The oversight committee acted as a program coordina- • Dedicating a management staff person as the tor to help facilitate HIT adoption by individual facility-level coordinator and contact for HIT homes. The committee first researched which HIT fea- implementation. tures and contract terms homes desired, set the vendor criteria, and conducted the vendor selection process. • Absorbing the cost of staff time required for Then, in coordination with the vendor, the committee training in the use of the technology participa- educated eligible homes about what the transition to tion in a labor–management committee, and HIT would entail. The committee also selected a des- managing the start-up and ongoing coordina- ignee to coordinate the financial and management tion of HIT. aspects of the HIT demonstration. That designee, the 1199SEIU Training & Employment Fund (1199 Application Process Training Fund), was experienced in working with These contractual and organizational terms, created unionized staff and employers in training and quality and overseen by a union–employer partnership, were improvement initiatives. The 1199 Training Fund coor- attractive to many unionized nursing home operators. dinated logistics between the homes, vendor, and Among the 140 nursing homes that were parties to the research teams. It also helped to prepare staff in each collective bargaining agreement and thus eligible to home for HIT implementation. apply, 83 submitted letters-of-interest for the 20 subsi- dized openings in the HIT demonstration. These inter- Requirements for Home Participation ested homes were sent a questionnaire designed by the To remain eligible for participation in the demonstra- vendor to gather additional information, and from the tion project, nursing homes were required to honor the 54 homes that returned the questionnaire, final partici- financial terms of the contract. They were also pants were chosen by the oversight committee in col- required to comply with workforce and organizational laboration with the vendor. commitments specified in the contract, including: L essons from a H ealth I nformation Technology D emonstration in N ew York N ursing H omes 5 Exhibit 1. Snapshot: 20 HIT Demonstration Homes Greater New York City metropolitan area: Brooklyn, Bronx, Manhattan, Queens, Staten Island (12); Long Island (6); north of New York City (2) Total number of beds in demonstration project homes = 4,467 Average* High* Low* National Average Number of beds per facility 223 320 120 107*** % of residents paid by Medicaid 78 98 54 64** % of residents paid by Medicare 13 26 2 14** Sources: * 1199SEIU Training Fund, March 2007. ** Henry J. Kaiser Family Foundation, http://www.statehealthfacts.org/comparebar.jsp?ind=410&cat=8, downloaded 4/23/09.*** Centers for Disease Control and Prevention, National Center for Health Statistics, http://www.cdc.gov/nchs/data/nnhsd/nursinghomefacilities2006.pdf, downloaded 4/23/09. Demonstration Homes National Average+ For-profit ownership 95% 67% Nonprofit ownership 5% 27% Government ownership 0% 6% + Source: American Health Care Association, http://www.ahcancal.org/research_data/oscar_data/Nursing%20Facility%20Operational%20Characteristics/Nursing_Facility_ OwnershipDec2008.pdf (downloaded 4/23/09). Successful Implementation and rapid adoption of HIT in nursing home settings In March 2007, the first HIT demonstration homes (Exhibit 2). The components of this rapid, full, and signed a contract with the vendor and began to roll out logistically successful implementation strategy are the HIT system through a multiphase process of plan- summarized in Exhibit 3. ning, training, and monitoring. Within little over a Both the demonstration project administrators year, by the end of March 2008, 17 of the 20 homes and the vendor viewed two factors as crucial to suc- had “gone live” with the technology in all units and cessful implementation: 1) strong support from home for all staff. An additional three homes were deployed leadership in planning and preparing for the technol- between November 2008 and April 2009. Key ele- ogy; and 2) the involvement of multiple levels of ments of the project’s implementation experience con- home staff in understanding, accepting, and offering tradict two commonly perceived barriers to successful input about the technology. Exhibit 2. HIT Adoption in Nursing Homes: Perception vs. Fact Perceived Barrier HIT Demonstration Facts It takes years to transition an entire facility from a paper to an The average time for all units in a nursing home to transition their electronic medical record. clinical and workflow functions to a health information system, from the start of pre-implementation planning to the moment all staff “went live,” was less than six months. This occurred after groundwork was laid by collaborative labor–management committees and involvement of multiple levels of staff in pre-planning. Direct-care staff will be unable to learn to effectively use the All direct-care staff in the nursing homes learned to access and technology at all or in a reasonable timeframe. record clinical and workflow information in real time on handheld PDAs or laptops/desktops, with an average formal training period of 12 hours for RNs/LPNs and two hours for CNAs. Training took place over a period of one to four weeks, with additional post-training support provided onsite from the vendor. Source: 1199SEIU Training & Employment Fund. Data are from the first 17 nursing homes implemented between March 2007 and March 2008. 6T he  C ommonwealth F und Exhibit 3. Support from Leadership 1 2 3 4 Pre-Implementation Labor/ Hardware & System and Project Planning Management and Network Configuration and Change Management Infrastructure Integrations Program 5 6 7 8 Onsite Training Pre Go-Live Week Go-Live Support Ongoing Program Support and Monitoring Role of Administrative Leadership Engaging All Levels of Staff Administrative leadership dedicated a good deal of The labor–management committees created by homes work to the pre-implementation phase. Within a short as a participation requirement for the HIT demonstra- time, home leadership worked with the vendor to doc- tion provided a structured approach to engaging all ument and analyze their facilities’ workflow, staffing levels of staff in the implementation process. These needs, user information, physician order and care plan committees were facilitated by the 1199 Training Fund back-order entries, as well as physical plant issues during the implementation phases before homes went related to hardware and network infrastructure. This live with technology. They aimed to engage nursing process presented an opportunity and potential catalyst home staff and administrators in collaborative deci- for reviewing organizational structures and priorities, sion-making related to HIT implementation, familiar- although not all homes perceived it or used it in this ize and educate all levels of staff about the benefits way. Leadership also faced choices about how to and characteristics of HIT, including allaying fears or encourage their staff’s willingness and enthusiasm to misconceptions, and promote a network of peer sup- learn the technology. port for learning the technology. “No health care worker left behind” became the unofficial motto of the committees. HIT Labor–Management Committees HIT labor–management committees were designed to consist of a minimum of eight members, four from different job occupations among union staff, and four from management. In practice, their sizes varied from home to home, with some as large as 25 members. The committees met six to eight times for an hour each time over the four- to-six-month period before the home went completely live. They were a conduit of information to facility staff during the process. Committees developed festive kickoff events to introduce and celebrate the coming arrival of HIT to that facility, the first of several informal opportunities prior to training for staff to see and experience the handheld PDAs and laptops they would be using. Another key activity of the committees was to identify and create peer mentors, that is, “go-to” people who were available to answer questions and offer support. L essons from a H ealth I nformation Technology D emonstration in N ew York N ursing H omes 7 Does Subsidization Lead to Voluntary Spread of HIT? An interesting development barely two years after the start of the subsidized New York HIT demonstration is that other nursing homes have apparently perceived benefits in the investments their competitors and peers are making in HIT. As of April 2009, 40 additional homes in New York voluntarily implemented technology by purchasing products from eHealth Solutions, Inc. Thirteen purchased complete HIT packages equivalent to those used in the demonstration. Another 30 homes were in the process of contracting for various applications.7 The implications of this phenomenon are worthy of consideration by policymakers: Does subsidization catalyze voluntary spread and replication of similar technology? Common Challenges as well. Additional intensive training was provided Staff from the 1199 Training Fund reported that the onsite from the vendor during the “go-live” week. most common challenges they encountered when facil- These multiple trainings accommodated individuals itating the labor–management committees were: who might need more time or one-to-one attention to learn the technology, and provided the opportunity for • fear of change and losing control from both mentors within each facility to provide on-the-job workers and management; assistance. Within this environment of preparation and support, all frontline staff learned to record information • fear of layoffs resulting from increased effi- in real time on handheld PDAs or laptops/desktops. ciency from the technology; After staff training, homes activated the tech- • fear of disciplinary actions, owing to a greater nology throughout the facilities one or two units at a ability of management to monitor workers; time, usually over a period of days or weeks, with a • fear that government would have access to strong presence from the vendor’s staff. Ongoing sup- confidential information; and port and monitoring from the vendor after the go-live period included round-the-clock customer support. The • fear by staff of its own inability to learn the 1199 Training Fund withdrew from its onsite role as technology because of such factors as learning labor–management facilitator after the go-live period, difficulties, language barriers, and unfamiliar- but remained a fiscal intermediary agent during the ity with computers. subsidy period, available to resolve issues between the The opportunity and willingness to address vendor and homes. these issues honestly in advance of the introduction of the new technology was a distinctive feature of Preparation for Annual Regulatory Surveys the project. The leaders of the HIT demonstration recognized the importance of informing the New York State Survey Staff Training Agency about the new HIT technology and its implica- Onsite staff training by the vendor commenced after tions for regulatory visits. In August 2007, at the time organizational planning, network infrastructure, and the first homes were going live with the HIT system, system configuration were completed. Over a period project coordinators briefed New York State of one to four weeks, RNs/LPNs received 12 hours of Department of Health (DOH) officials, who requested formal training, while CNAs, whose recording respon- that surveyors be trained in the use of the technology sibilities were less involved than clinical staff, for survey purposes. In December 2007, approximately received two hours. Other professionals, such as physi- 100 representatives from the federal Centers for cians, social workers, rehabilitation therapists, and Medicare and Medicaid Services and DOH surveyors dietitians received training specific to their specialties were trained by the vendor in accessing information 8T he  C ommonwealth F und from the HIT system relevant to regulatory visits. The time-sensitive tasks, such as medication administra- vendor provided six months of additional webinars to tion. Some homes have viewed such data as an indica- DOH to update new survey staff. Homes participating tor of staff ineffectiveness and sought to resolve the in the demonstration were encouraged to inform the perceived problem through that lens; others have survey team upon arrival about the new HIT features, undertaken root-cause analysis of how all factors in and to provide tutorials or guidance, if needed. In their system might be contributing to delays in medi- addition, the vendor was notified by nursing homes cation distribution, such as numbers of medications when a survey team had arrived onsite, and sent one of per resident, or per unit, in relation to staffing patterns. its staff to stand by to assist with instructing surveyors. Another difference has been in perceived time savings. David Lipsky and Ariel Avgar of the Cornell HIT Adoption: Variation Is the Theme Scheinman Institute for Conflict Resolution (ICR) con- The project’s implementation process successfully ducted interviews at demonstration homes one year launched homes into the adoption stage of HIT use, after the installation of the HIT system, which revealed marked by the replacement of paper records with elec- contradictory findings. Comparing electronic and tronic ones and use of the technology across all direct- paper records, the administrator of one home stated: care staff. In March 2009, for instance, among the 17 It takes more time to access the record. You nursing homes that had been deployed as of a year ear- have to go into the system. You have go through lier, the aggregate certified nursing assistant on-time all of the prompts to get to whatever it is that documentation rate was 99 percent, representing a you are looking for . . . a care plan or to order total of 118,600 CNA tasks per day. Medication medications . . . . Whatever it is you need to do, administration records and treatment administration it takes longer to do the job you have to do. I’m records for the 17 homes collectively totaled 69,100 not talking about hours, of course, but every per day, again with 99 percent documented on time.8 minute counts. So everyone has confirmed— However this overall statistic masks the significant all departments—that it takes more time to do variation across homes. their job.9 Same Technology, Different Effects A supervisor in another home offered a different Despite the fact that each home implemented the same report: software and hardware via the same vendor, there have been notable variations observed both by early I see the CNAs have more time now to sit research findings and by the 1199 Training Fund coor- down one-on-one with residents; they are spend- dinators about how the adoption of HIT has affected, ing more quality time with them. They are not and has been used by, homes. Examples of these dif- rushing to do their books between two and three ferences range from how homes responded to bugs in o’clock.10 the HIT system, to whether the technology was funda- Meanwhile, a frontline staff person in a third mentally perceived as a means of improving clinical home reflected: indicators, financial outcomes, employee efficiency, or the entire culture of a home. I think I’m spending more time with the For instance, data from the HIT system increase residents, being that I don’t have to go back to transparency in workflow and production documenta- the office to document that they’re not feeling tion. Because all entries can be viewed in real time and right . . . I can just click off and go into a prog- are required by the end of the shift, it becomes easier ress note . . . . It’s less time in the office, and you for administrative staff to identify delays in delivery of can profile medication, consults—everything is right there in front of you.11 L essons from a H ealth I nformation Technology D emonstration in N ew York N ursing H omes 9 Variation in Use of Available HIT Data • stimulate efficiency among staff by awarding An important area in which homes have varied signifi- recognition for highest levels of complete on- cantly is how they have used the range of data avail- time documentation by shift and unit. able through the HIT system. While homes met the Project coordinators did not originally foresee basic logistical benchmarks of creating an EMR for the need to specify benchmarks related to quality each resident, automating workflow tasks and physi- improvement analyses, or the customization of fea- cian order entry, and activating some clinical decision tures that might maximize resident quality of care, support menus, coordinators of the HIT demonstration increase organizational efficiencies, or improve staff now understand that this represents only one step working conditions. Thus, there was no expectation or toward achieving the potential benefits of HIT in nurs- ongoing training process during the subsidized period ing homes. HIT can also summarize complicated data for all administrators or staff to learn and use these in real time, customize clinical decision-support and capabilities. This raises an important question for poli- workflow prompts, and perform multivariable analyses cymakers: Should national HIT funding strategies of individual residents, units, facility-wide trends, and include targeted benchmarks and/or incentives to staff. The quality improvement possibilities inherent in encourage maximization of the quality improvement these capabilities are very rich. Not all homes, however, uses of technology, in addition to focusing on the have engaged in these types of analyses and customi- replacement of paper records with electronic ones? If zations, and those that did pursued different strategies. so, what should those benchmarks be? Are Data Analysis Benchmarks Needed? Diverse Organizational Priorities and Perceptions The following are examples of analyses some homes The variation in the perceived and real effects of HIT have reported they perform with HIT capabilities. adoption across homes may be a function of different Administrators said these functions are quicker and organizational goals and leadership styles that predis- easier to perform with HIT than with a paper record: pose homes to perceive any new development, HIT or otherwise, through particular lenses. This is a theory • check nosocomial infection rate per unit on a that David Lipsky and Ariel Avgar have formulated daily basis, resulting in timely interventions based on their early review of qualitative organiza- with increased precautions to contain and pre- tional and workforce data from 10 demonstration vent the spread of any identified infection; homes and five control homes collected before the • analyze daily direct-care staff record of HIT implementation and one year after. Seeking to changes in care needed by residents, resulting explain why the same hardware and software have in quickly and accurately updating residents’ apparently been regarded and used differently among care plans; the demonstration homes, they identify different mana- • revise billing categories more regularly to gerial strategies and goals for adoption of the HIT sys- accurately reflect levels of care and improve tem among the demonstration homes. Lipsky and reimbursement rates; Avgar then correlate these strategies and goals with the effects of HIT on resident care, financial, and work- • more closely track incontinence in residents force outcomes. for analysis of potential causes and solutions; Examples from their interview data illustrate the • maximize use of existing staffing to balance range of responses their theory encompasses: workloads and resident needs; and 1. In a post-implementation interview, one adminis- trator described a primary focus on resident care 10T he  C ommonwealth F und as the lens through which the HIT system was impact on these four areas: quality of care, workplace perceived: issues, business impact, and culture of the home. Sigma Care is like a hub, and everything Together, these evaluations will be the first to capture we do everyday, the dashboard drives our day. the effects of HIT adoption on so many dimensions in But what is the dashboard? The dashboard is so large a sample of nursing homes. A unique charac- filled with resident information, so the resi- teristic of the final analyses will be integrated and dents are driving our day. So it has kind of multidisciplinary comparisons of targeted data brought us all together. I think that there’s just designed to elicit findings from a wide range of per- a greater emphasis placed on resident care spectives (Exhibit 4). in general.12 Self-Reported Data from Homes 2. Another administrator focused on the opportunities While research data are not yet available for the proj- for improved financial returns through use of the ect as a whole, some homes have individually tracked HIT system: and reported cost savings and clinical efficiencies, including reduced formulary costs, fewer diagnostic Speaking fiscally, you are being paid for code errors in medications, and improved accuracy in the care that you provide—that is going to be recording acuity levels for reimbursement purposes. the future. The only way to capture it is elec- The Appendix summarizes self-reported information tronically. The more information you capture, from one demonstration home. the more money you are going to make. Shouldn’t you get paid for everything Conclusions you do?13 Replication Considerations 3. Yet another administrator spoke prior to imple- The New York State Nursing Home Health mentation about how the HIT system was intended Information Technology Demonstration Project offers to facilitate staff empowerment and education: a logistically successful model for implementation of We are trying to give people the opportu- point-of-care electronic health records for a substantial nity to manage themselves, which means giv- sample of nursing home residents, using partial public ing them the tools to work as best they can subsidization. Potential replicators of the project, and in their environment. The technology will those who mold future HIT funding strategies, would serve as an educational tool helping us reach do well to take note of the defining elements of the these goals.14 project before embarking on similar designs. The proj- ect’s successful union–employer partnership was moti- Multifaceted Research vated by the dual goals of improving both quality of Lipsky and Avgar’s research is only one component of resident care and staff retention and recruitment. These an unusually comprehensive range of evaluations two goals shaped the framework for the project. sponsored by New York’s HIT demonstration and The Specifically, the prohibition against laying off union Commonwealth Fund to fully investigate the effects of staff and the creation of labor–management commit- HIT adoption on participating nursing homes. The ele- tees appear to have contributed to overcoming barriers ments of the research reflect the multifaceted nature of to acceptance of the new technology by direct-care nursing homes as organizations. Nursing homes are staff. An implied result of this prohibition is that simultaneously clinical care systems, complex work- improved efficiencies would be reinvested into better force environments, business enterprises, and homes to resident care. If these priorities and guarantees had residents. Research will focus on the technology’s L essons from a H ealth I nformation Technology D emonstration in N ew York N ursing H omes 11 Exhibit 4. Research Questions Resident Quality of Care and Quality of Lifea Workplace Impact: Employment and Labor Relationsb What impact will the installation of HIT have on: What are the effects of the new technologies on: • specific measurable resident outcomes of falls, skin breakdown, • workforce retention and recruitment; hospitalizations, behavior problems, and change in functional • employee perceptions and attitudes, including job satisfaction, status; stress and commitment; • resident mood and quality of life measure; and • organizational effects, such as communication among and • facility-wide indicators including skin breakdown, incontinence, between staff, organization of work, and organizational culture; decline in cognition, UTIs, and fecal impaction? • resistance to change and conflict; and • labor relations, such as organizational and bargaining unit What are subjective resident reactions to the introduction of HIT, effects, and labor-employer perceptions and attitudes? including: • awareness of the new technologies; • attitudes toward the new technologies; and • satisfaction with care by and relationships with staff? The Business Case for Nursing Home HITc Resident-Centered Care (Culture Change)d How does HIT affect nursing home productivity? What impact does the installation of an electronic health record What factors cause the productivity effects of HIT to vary have on the level of resident-centered care (culture change) in across homes? participating facilities? What is the business case for investing in HIT by nursing home operators and for other long-term care stakeholders, such as payors and residents/families? a Karl Pillemer and Rhoda Meador, Cornell Institute for Translational Research on Aging. b David Lipsky and Ariel Avgar, Cornell Scheinman Institute on Conflict Resolution. c Lorin Hitt and Prasanna Tambe, Wharton School of University of Pennsylvania. d Cornell Institute for Translational Research on Aging and Cornell Scheinman Institute on Conflict Resolution. not been present, HIT implementation may have • Many nursing homes were willing to imple- unfolded differently. ment major transformations to adopt a point- There are important implications here: of-care, clinician-centric HIT system when 1. Assuming that resident quality of care and work- financial, logistical, and administrative support force retention are national health care priorities, is were offered by a union–employer partnership. there a need to ensure reinvestment of some por- • It took less than six months to transition an tion of the financial benefits of publicly subsidized entire facility from paper to an EMR in all HIT into these areas? units for all staff. 2. In homes where union–employer partnerships are • This transition took place with the support of not present, how can employers best engage staff clear union–employer agreements, including to overcome potential barriers to accepting HIT the designation of a respected coordinating implementation? body for oversight of the project. • Engaging multiple levels of staff in prepara- Lessons Learned tion for implementation via labor–management Key lessons learned from the implementation and committees was an important part of the early adoption experiences of the HIT demonstration groundwork laid prior to adoption. include: 12T he  C ommonwealth F und • Within this context, all direct-care staff were • Homes significantly differed in how they used able to learn to use the technology effectively available HIT data for quality improvement within a short period. purposes related to resident care, financial, and workforce outcomes. • The hurdle of initial capital outlay for HIT was overcome by a subsidy structure spread- ing all the costs of the HIT hardware, soft- Next Steps ware, implementation, and maintenance over a The research outcomes of the HIT demonstration proj- multiple-year contract to create a single per- ect will provide information about HIT implementa- bed, per-day rate and a subsidized start-up and tion in nursing homes in New York City (Exhibit 5). adoption period. Already, the logistical lessons learned from the proj- • The ability to support comprehensive organi- ect’s design, implementation, and early adoption peri- zational planning, workflow analysis, and staff ods provide information about health information tech- training prior to implementation was an impor- nology funding and deployment strategies. This infor- tant criterion for selecting the vendor, in addi- mation should be useful for policymakers as they grap- tion to the ability to provide secure, point-of- ple with how to implement a HIT infrastructure across care, clinician-centric software and hardware. the United States. • Home leadership appeared to have diverse For More Information organizational aims for adopting the Further information about the New York State Nursing technology. Home HIT Demonstration Project is available by • Home leadership seemed to have diverse per- e-mailing Scott White at the 1199SEIU Training & ceptions of the effects of HIT technology on Employment Fund at ScottW@1199.org. the organization. Exhibit 5. Research from the HIT Demonstration Project Will Help Clarify Key Issues Issue Potential Solution from HIT Exchange of complete admission and discharge data between Interoperable nursing home and hospital patient clinical records may nursing homes and hospitals is key to continuity of care, since improve quality of care upon admission and discharge, and contain residents frequently transfer between the two venues. costs due to unnecessary duplication of services. The long-term nature of care in nursing homes requires Clinician-centric HIT systems may allow multimember care teams, multidisciplinary, multifaceted care planning that includes ongoing including direct-care staff and the resident, to input and access consideration of quality of life and resident choice. Direct-care staff accurate information in a timely manner to optimize care planning are the closest link to the resident in terms of gathering information and care delivery. and providing care on a daily basis. High turnover rates in nursing home workforces currently plague HIT may have a positive effect on both workforce retention and long-term care facilities and impact their ability to deliver quality recruitment, thus contributing to an adequate and stable nursing care. There will also be a need for increased recruitment of long- home workforce. term care workers in the next decades to care for the influx of baby boomers. Current funding mechanisms for nursing homes are complicated HIT systems may improve the speed and accuracy of capturing of formulas that require accurate capture of each resident’s activities of reimbursement and regulatory data. Time saved through automated daily living and acuity levels. Federal and state regulatory reporting, processes may give nursing home staff more time to provide care documentation, and inspection requirements are extremely complex and interact with residents. and take a large amount of nursing home staff time and effort. L essons from a H ealth I nformation Technology D emonstration in N ew York N ursing H omes 13 Appendix. Nursing Home Hospitalizations and Hit: One Facility’s Experience The following is provided by Administrator Caroline Rich of the Four Seasons Nursing & Rehabilitation Center in Brooklyn, New York. Four Seasons is a 270-bed facility with six units: long-term care, short-term rehabilitation, IV therapy, ventilator, dialysis, and adult day care. It was among the first two nursing homes to install the SigmaCare technology as part of the HIT demonstration in August 2007. Can point-of-care, clinician-centric HIT help homes avoid unnecessary transfers of residents to the hospital or emer- gency room? Caroline Rich, administrator of Four Seasons Nursing & Rehabilitation Center, says, “The technology makes it possible for the resident’s physician and the nursing home care team onsite to simultaneously access and confer about the resident’s full clinical status in a matter of minutes. Then they can continue to monitor and commu- nicate in real time as a standard intervention for fever or cough, for instance, is implemented, just as would be done in the ER. A paper record does not afford that kind of simultaneous, instantaneous access to clinical information, nor can the offsite physician easily monitor the resident’s status via Internet.”15 Given the importance of close monitoring by a physician when a frail elder becomes ill, it has not been unusual for physicians to err on the side of safety by ordering a transfer to the ER. A scenario described by Ms. Rich illustrates the clinical alternative with HIT. “Last week Mrs. S. spiked a fever of 100.2 and was not eating much. The nursing supervisor immediately contacted the resident’s physician offsite, who, via Internet, viewed Mrs. S.’s full clinical record over the last week, including the real-time data being entered at the bedside by the nursing team and direct-care staff, and a record of all her med- ications and when she had taken them. In response to her symptoms, a plan was made between the nursing home care team and physician to give Mrs. S. intravenous fluids for 24 hours to avoid dehydration, give fever-reducing medication, monitor her vital signs, and inform the physician of developments. The physician could view from off- site, at any time, the resident’s ongoing treatment and care records. If hospitalization was indicated, it could have immediately been carried out, but Mrs. S.’s fever became normal over the next 24 hours and she began to eat and drink. The treatment plan was appropriate and no hospitalization resulted.”16 14T he  C ommonwealth F und N otes 7 E-mail communication from Steve Pacicco at eHealth Solutions, Inc., April 28, 2009. 1 The American Recovery and Reinvestment Act of 2009, TITLE XIII, Health Information Technology. 8 E-mail communication with eHealth Solutions, Inc., April 20, 2009. 2 Ibid. § 3001(c)(3)(A)(vii). 9 Unpublished data provided in e-mail communica- 3 Center for Disease Control, National Center for tion from Ariel Avgar, June 17, 2009. Health Statistics, Nursing Home Current Residents June 2008, http://www.cdc.gov/nchs/data/nnhsd/ 10 Ibid. Estimates/nnhs/Estimates_Demographics_Tables. 11 Ibid. pdf#Table04, downloaded April 13, 2009. 12 Ibid. 4 Private e-mail communication April 27, 2009, from Jay Sackman, former executive vice president of 13 Ibid. 1199SEIU Nursing Home Division. Parties to the collective bargaining agreement were five nursing 14 D. B. Lipsky, A. C. Avgar, and J. R. Lamare, “Orga- home associations and other represented homes. nizational Strategies for the Adoption of Electronic The original and current membership of the Qual- Medical Records: Toward an Understanding of ity Care Oversight Committee include a nursing Outcome Variation in Nursing Homes.” Presented at home employer representative, William Pascocello, annual meeting of the Labor and Employment Rela- a union representative, Jay Sackman, and a neutral tions Association, Jan. 4, 2009. chairman, the New York nursing home industry’s 15 Interview and e-mail communication with Caroline impartial arbitrator, Martin Scheinman. Rich, April 2009. 5 In the Matter of the Interest Arbitration between 16 Ibid. Southern New York Associates, L.L.C., et al., and 1199SEIU United Health Care Workers East (Mar- tin F. Scheinman, Impartial Chair), March 2006, pp. 14–15. 6 http://www.ehealthsolutions.com/index.asp. L essons from a H ealth I nformation Technology D emonstration in N ew York N ursing H omes 15 A bout the A uthors Shana Lieberman Klinger, M.A., provides project management, research, written products, and program evaluation to nonprofit, government, and academic organizations. She coauthored the Individualized Care Pilot Toolbox (2008) at the Rhode Island Department of Health, summarizing an innovative project to promote resident-centered care in nursing homes via the mandated federal survey process. Current projects include the Optimization Study of the DEKA Arm at the Providence VA Medical Center. Scott White has been an organizer in the labor movement for nearly 20 years, representing an array of health care workers in New York, New Jersey, and Connecticut. In late 2006, at the request of the 1199 Service Employees International Union (SEIU) leadership, he led the New York State Nursing Home Health Information Technology (HIT) Demonstration Project. Mr. White was subsequently nominated by the SEIU to be the health care workers’ labor representative on the initial national HIT Policy Committee convened by the federal Office of the National Coordinator. In February 2009, the Government Accountability Office appointed him to the HIT Policy Committee, where he represents health care workers and their issues in the creation and advancement of a national health information network. Editorial support was provided by Christopher Hollander. This study was based on publicly available information and self-reported data provided by the case study institution(s). The Commonwealth Fund is not an accreditor of health care organizations or systems, and the inclusion of an institution in the Fund’s case studies series is not an endorsement by the Fund for receipt of health care from the institution. The aim of Commonwealth Fund–sponsored case studies of this type is to identify institutions that have achieved results indicating high performance in a particular area of interest, have undertaken innovations designed to reach higher performance, or exemplify attributes that can foster high performance. The studies are intended to enable other institutions to draw lessons from the studied institutions’ experience that will be helpful in their own efforts to become high performers. It is important to note, however, that even the best-performing organizations may fall short in some areas; doing well in one dimension of quality does not necessarily mean that the same level of quality will be achieved in other dimensions. Similarly, performance may vary from one year to the next. Thus, it is critical to adopt systematic approaches for improving quality and preventing harm to patients and staff.