The Business Continuity Planning Iniatives at Johns Hopkins Health System
IN 2001 THE JOHNS HOPKINS HEALTH SYSTEM(JHHS; Baltimore, MD), began an earnest examination of its ability to recover patient care and business processes when operations have been disrupted unexpectedly. By the end of 2002, pilot continuity plans were developed for the payroll, patient financial services, and hospital IT departments. The Business Continuity Planning (BCP) Initiative, a project to facilitate contingency planning across entire health system beginning with patient intake and admissions functions, was launched in January 2003. Today, the BCP Initiative is 60 percent complete across the health system, with completion targeted for mid- to late 2006. Every clinical, ancillary, and administrative department in the entire Johns Hopkins Health System will develop a comprehensive, integrated, and easy-to-follow business continuity plan. The mission and objective of the BCP Initiative at Johns Hopkins is to allow department directors, administrators, and managers-as well as directors of core processes and ancillary functions-to continue to lead their operations under emergency or disaster conditions. Furthermore, executing these plans should achieve recovery objectives through the implementation of viable, integrated, and rehearsed contingency strategies, allowing uninterrupted patient care, medical education, research, and customer service.
In early February 2004, the Johns Hopkins Hospital experienced a small taste of the need for continuity planning when its Cardiology Care Unit (CCU) became infected with the highly contagious, but relatively harmless, Norwalk Virus of cruise-ship fame. First, the clinical staff and nurses became sick, followed shortly by almost all of the doctors. The CCU patients were smoothly relocated to other units without any exposure to the illness, but the functional unit director in conjunction with the hospital epidemiologist made the decision to close the CCU completely because there were insufficient healthy staff to safely care for patients on the unit, and because this was believed to be necessary for appropriate decontamination. The hospital epidemiologist recommended to the Department of Medicine administration that they discard CCU supplies and send non-disposable equipment to Central Sterile for decontamination to ensure patient and staff safety. Such major organizational upheaval-and the loss of half a million dollars in revenue and supplies-made the local evening news. Would the situation have been more easily and economically handled had the CCU administration had a business continuity plan? Decisions and actions could have been planned, validated, and rehearsed long before any infection struck the CCU. A media message could have been preemptively developed. In fact, many advance actions could have been evaluated to mitigate the impact of the crisis on the administrative, managerial, logistical, and business functions of the CCU. Today, healthcare organizations are operating with slimmer revenue margins and can ill-afford a disaster that adversely impacts their bottom line. The issue of risk in the business of delivering healthcare has risen to the attention of executive- level leadership, which has, in turn, undertaken efforts to mitigate and supplant potential disaster with comprehensive continuity strategies designed to protect their patients, organizations and their reputations.
The Johns Hopkins Health System Board of Trustees' Audit Committee is the body that identified business continuity planning as a health system priority and which now holds ultimate authority over all program matters. Through the Initiative's Steering Committee, the JHHS Board of Trustees conducts the ongoing management and governance of the Business Continuity Planning Initiative. This structure is designed to ensure Johns Hopkins' business continuity requirements are assessed, resourced, and funded; and that recovery and continuity strategies are comprehensive, coordinated across the Johns Hopkins Institutions, and effectively tested and maintained. The Steering Committee is charged with operational oversight of the JHHS BCP Initiative. It is chaired by the executive vice president of JHHS and COO of the Johns Hopkins Hospital. Committee members are updated regularly (approximately once every two months) by the BCP project lead. Every Johns Hopkins entity is represented on thecommittee and the leadership of most core processes withinthe system is represented as well. To date, the Steering Committee has had a very positive impact on the project by keeping it a system-wide priority and resourcing it appropri-ately. The JHHS Department of Operations Integration (OI) was assigned the task of leading and facilitating the entire BCP Initiative. Specifics of the department's role are discussed later in this article.
Investigating Industry Trends and Best Practices
During the planning phase of the BCP Initiative, Johns Hopkins found that few organizations in healthcare were pursuing detailed business continuity plans, and of these fewer still were academic medical institutions like JHHS is. Expanding the scope beyond healthcare, Johns Hopkins searched across American industry for common threads and proven solutions. The following best practices were observed and adopted:
- Effective governance of business continuity planning initiatives takes place at executive levels within successful organizations. When business continuity planning receives focus at the upper echelon of leadership, plans are generally well developed and integrated with business operations.
- Almost every successful business continuity program is scenario-based. A scenario-based approach to contingency planning keeps the organizational focus on disaster results, rather than on the endless permutations of what initiated the crisis. In other words, an unavailable building is an unavailable workplace, regardless of whether the cause is a bomb threat,terrorist attack, gas leak, burst water pipe, or fire. The business continuity plan must address the unavailable building, not the incident that caused the building to become unavailable.
- Scenarios are based on the vulnerabilities most relevant to the organization. For example IT/IS reliant companies plan for information system, network, and computer interruptions in order to address critical vulnerabilities. Organizations along the Southeastern seaboard develop plans around interruptions resulting from seasonal weather extremes. Similarly, hospitals and healthcare facilities must address vulnerabilities that impact the delivery of patient care, while academic health systems also focus on continuance of research and teaching efforts.
- Successful BCP programs include a testing phase during which written plans are set in motion and "operationalized." Frequently, "disinterested parties" render objective, unbiased assessments of the plans, their execution, and staff knowledge.
- All successful business continuity programs include a staff education component where employees are taught their roles and learn what is required of them during contingency plan execution. New staff members are introduced to the business continuity plans as part of their orientation. This prevents plans from becoming shelf-bound binders viewed but once a year.
To develop business continuity plans that achieve Johns Hopkins' mission, the first step was for the organization to conduct a formal risk analysis. According to the Business Continuity Institute, "Prior to creation of the [business continuity] plan itself, it is essential to consider the potential impacts of disaster and to understand the underlying risks: these are the foundations upon which a sound business continuity plan or disaster recovery plan should be built." As a prelude to the Business Continuity Planning Initiative, JHHS used risk analysis tools to identify critical vulnerabilities and exposures based on the internal and external environment in which they operate as an urban-based institution located in the heart of Baltimore. Initially, a vulnerability analysis was conducted, followed by a more comprehensive business impact analysis (BIA). The BIA evaluated areas of impact in two primary categories: patient care delivery and revenue capture. Within these categories, criteria such as recovery time objectives, recovery complexities, management aspects, and facility and equipment requirements were analyzed. Information was evaluated for every department in each of the acute care hospitals and across affiliate organizations such as physicians' practices, the home health company, and other corporate affiliates. Departments most vital, first to patient care and medical research, and secondly to revenue capture, were identified.
KEY DATA APPLICATION UNAVAILABLE. This scenario assumes a department's key data application/s (or computer program/s) become unavailable for some reason. A "key" application is one whose loss causes the department's business process to stop or slow to the point of preventing effective patient or customer care. Scenario 1 describes the actions that the department takes to overcome the interruption, and resume service using an alternate data application or a manual back up. For example, the Pathology Data System (PDS)-in the Department of Pathology-or the Physician Order Entry (POE) System-in the Department of Medicine-is unavailable for some reason. How does the affected department continue to process laboratory samples or record and process physician orders despite these outages?
SCENARIO 2: NETWORK CONNECTIVITY LOST.
This scenario assumes the main computer network used by a department becomes unavailable. External communication via this network is not possible. Any data applications housed on an external server become unavailable. Any data applications housed on department computers or local servers at the department remain available. Scenario 2 describes the actions that the department takes to overcome the interruption, and resume services. Often, a network outage equates to a "Key Data Application Unavailable" (i.e., Scenario 1) because key data applications are usually housed on external (to the department) servers. In this case departments defer to plans developed in Scenario 1. Loss of network connectivity also creates a loss of access to shared computer drives-a vulnerability addressed by this scenario.
SCENARIO 3: BUILDING SYSTEM UNAVAILABLE.
This scenario encompasses several interruptions. First a specific floor the department is located on becomes unavailable. This impacts departments differently as some are housed entirely on one floor while others operate over multiple floors in multiple buildings. Secondly, the entire building is unavailable causing the department to relocate to a predesignated alternate location. This is by far the most complex aspect of BCP development. Lastly, several utility interruptions are addressed. These are heating, air conditioning, electrical power, water, medical and laboratory gases and vacuum, and telephony.
SCENARIO 4: KEY VENDOR(S) UNAVAILABLE.
This scenario assumes the department's key vendor, business partner/associate, or supplier (generically referred to as "vendor") is unavailable. A "key" vendor is one whose loss causes the department's business process to stop or slow to the point of preventing effective patient or customer care. Scenario 4 describes the actions that would be taken to overcome the interruption in service, and either resume service by that same vendor or gain the services of an alternate vendor. The scenario feature tools that assist departments in determining which vendors are key.
SCENARIO 5: PERSONNEL/STAFF UNAVAILABLE.
This scenario assumes that a very large portion (as determined by each department) of the department's staff or workforce is unable or unwilling to come to work for a period of 72 hrs or more-a similar scenario to the 2002 SARS outbreak in Toronto, Canada. This scenario details distribution of work for available staff, minimum staffing requirements for the department, and the department's ability to shift available personnel to assist others. Since it is not possible to know in advance specifically who will be absent, this scenario seeks to capture the decision making criteria for how personnel and workload are distributed rather than the particulars of "which position does what duty". This allows anyone placed in a leadership role to sustain department or unit operations. of the direct patient care units already had evacuation plans in place (mandated by the Joint Commission on Accreditation of Healthcare Organizations), along with trained clinical staff, that allowed them to respond effectively to life-threatening disasters. Direct patient care departments each already had internal Disaster Coordinators who could implement these plans and coordinate the department's response to a disaster. Led by the Disaster Control Administrator, the health system also has a robust disaster response and crisis management infrastructure that would provide resources and coordinate department-level disaster response and business continuity plans in the event of a catastrophe. With this high level of preparedness already in place, direct patient care departments (e.g. Medicine, Surgery, Neurosciences) would not need to develop BCPs immediately. With these factors in mind, the BIA yielded a list of all departments, functions, and core processes in the health system-ranked from the most to the least critical to patient care and revenue generation. Because patient care units already had disaster plans, they were ranked in the upper middle of the list. At the top of the list were infrastructure support departments that impacted everything and everyone in the health system. Examples are the Department of Facilities and [Power] Plant Operations; Corporate Purchasing (responsible for everything from paper to OR supplies) and the Department of Materials Management and Distribution.
Ancillary patient care departments such as Radiology and Pathology were also at the top of the list. These were immediately followed by all the direct patient care functions such as Emergency Medicine, Surgical Sciences, and Ambulatory Services. The middle of the list held critical services like Medical Transport, Infection Control, Pharmacy, Medical Records and Nutrition. Also in the middle of the list were financial functions like Financial Planning and Budget, Financial Analysis and Cost Accounting. Last on the list, ancillary functions such as Pastoral Care, Nursing Recruiting, and various administrative functions and administrative departments were ranked as being least critical to patient care and revenue generation.
Which Track to Take?
Possible business continuity solutions vary, both in their effectiveness and in their cost. The first option JHHS considered was one used by several other industries and the federal government: Enterprise-wide business continuity plans. Gauging these models in the financial sector as well as with several large health insurers; implementing one health system-wide business continuity plan covering all aspects of the organizational structure and operations was considered. The Johns Hopkins Health System is composed of two acute care hospitals in the City of Baltimore, and one community hospital in Howard County. The combination of these three acute care hospitals with other integrated healthcare delivery components, multiple ambulatory care facilities, a homecare company, a managed care contracting entity, and a network of primary and specialty care practices throughout Maryland make up the Johns Hopkins Health System. Along with the Johns Hopkins School of Medicine, they are collectively known as Johns Hopkins Medicine. After scrutinizing the feasibility of applying the insurers' model of an enterprise-wide business continuity plan, it was decided that Johns Hopkins Medicine was too disparate in design and organization and too decentralized in operation for one overarching plan to be effective. Indeed, one plan would have answered the question "Do you have a BCP?" It would not, however, be a plan that could bring to bear employable, flexible, and effective crisis mitigation and continuity strategies. Next two other options for implementing the Business Continuity Planning Initiative were examined. The first was just "beefing-up" or augmenting existing disaster preparedness and response infrastructure and capabilities. This would include preparation and preemptive strategies to allow more rapid business resumption, but nothing in the way of actual business continuity plans. This course of action was also discarded because it essentially represented throwing money and people at the problem to create temporary solutions under the guise of business continuity planning. The other option considered involved each hospital and entity within JHHS developing its own business continuity plan. There are several for-profit hospitals and non-profit hospitals and academic medical centers in Baltimore, DC, and New York that have a singular business continuity plan for each institution. Such plans are highly integrative of patient care, administrative, and logistical functions, and must be exceptionally comprehensive to be effective. They also tend to be bulky and difficult to manage. It was decided that, although all-encompassing singular institution plans may work effectively at smaller hospitals, for a decentralized group of large institutions (Johns Hopkins Hospital alone has over 900 patient beds) something different was needed.
The Ultimate Decision
The approach ultimately chosen as the best fit for the organization required each department within each hospital and institution of the health system to author its own business continuity plan. Development at the department level allows those most familiar with unit operations to tailor solutions to their own needs and organizational dynamics. In the face of a crisis, these plans would likely be executed under the leadership of the plan authors themselves. A key benefit of planning at the departmental level is that departments encompass inpatient, outpatient, and research functions. This composition would translate directly to BCP structure. For example, at Johns Hopkins Hospital, the Department of Pediatrics includes inpatient functions such as the Pediatric Intensive Care Unit (PICU) and Pediatric Cardiology; outpatient functions such as Outpatient Renal Dialysis, and Pediatric Speech; and research functions that span both inpatient and outpatient functions. A BCP developed by this department will incorporate, and hopefully integrate, the crisis response of all these functions. Furthermore, plans designed for execution at the lowest organizational levels tend to be more agile and flexible. Other advantages include a personal stake for plan developers in the success of their continuity strategies; a detailed understanding of the plan's intricacies; departmental-or user-control over plan development, management, and implementation; and a clearly defined scope for each plan. Department administrators are responsible for conceiving, coordinating, writing, and testing their own business continuity plans. A disadvantage anticipated with this approach was the fact that a department-by-department methodology would create a lengthy timeline for project implementation. This is because only a fixed number of departments can be facilitated through the development process at any given time due to resource constraints. Another disadvantage was the burden put on departmental administrators. Administrators already had numerous initiatives before them with which to contend. Lastly, a department-by-department strategy would tie up resources from the Department of Operations Integration, which would facilitate the process, for a lengthy time period. None of these disadvantages was deemed a significant obstacle to the success of the BCP Initiative at JHHS. The Board of Trustees felt that the department-by-department approach best fit the Johns Hopkins culture of decentralization, and arrangement of highly independent functional units and departments. They also thought that this solution would best address and balance the considerations of operating in both urban and suburban environments. Each organization could thus tailor solutions unique to their own needs, organizational dynamics, and operating environment.
Developing the Plan
Once a decision was made to pursue department-level BCPs, developing a plan format was next. Five scenarios were chosen as the core of each business continuity plan. These scenarios kept business continuity development focused on situations with which the organization must contend, rather than their cause.The five scenarios are the core of the Johns Hopkins BCP. Each BCP also includes sections on scope and purpose, chain of command, assessing the magnitude of emergency situations, departmental chain of command, organizational charts, plan communications (internal and external); and a section detailing department hotlines, their activation and use. Each BCP also incorporated a system through which departments contacted their personnel. This typically was an emergency call tree-an ordered list of contact numbers delineating call down responsibility. The plan covered the activation procedure and specified individual staff responsible for periodic testing and updating of the call trees. Lastly, each BCP covered regular plan maintenance and updating; staff awareness including familiarization of new employees; and plan testing and drilling timeframes, methods, and responsibilities. All of this is captured in standardized templates which serve to provide a consistent platform from which all Johns Hopkins business continuity plans are generated. Thus any department looking for information in another department's BCP knows exactly where to look as they are all structured alike. Standardized BCP architecture also amplifies other development efficiencies and facilitates interdepartmental collaboration. A senior project administrator from the JHHS Department of Operations Integration facilitates the BCP development process. This person has an extensive background in contingency planning, is formally trained in BCP development and evaluation, and is assisted by another senior project administrator and a project administrator from the department. All are skilled at project management and group facilitation. The project administrators coordinate, lead, and facilitate group meetings, share information and ideas across groups, provide process feedback to participants, review BCPs, and make recommendations to plan authors. Prior to initiation, department leadership participate in an education session where the concepts and framework of BCP development at JHHS are explained. A planning meeting then precedes each of the five scenarios addressed in the plan. Here, the project administrator explains the specifics of the scenario and how others have addressed it in the past. The group then enters into a discussion of what approaches and strategies are best suited for particular functional areas or core processes. Department administrators return to their own departments and form internal sub-groups to develop and write their business continuity plans. At the next meeting scenarios are submitted for review and assessed by the Operations Integration project administrators. Feedback is provided and plan authors given an opportunity to address recommendations. This process continued until all components of the BCP are complete. The BCP then undergoes a final review and is submitted for approval and "sign off" to each department's senior leadership and the appropriate entity's representative on the BCP Initiative's Steering Committee-typically a vice president. In the post-September 11 world, no one questions either the ever-present and credible threat to civilian institutions or the need for contingency plans. Nevertheless, in order for business continuity concepts and "contingency thinking" to be effective, healthcare institution must change the way they think about disaster. Considering the number of times a year the electricity, water, and computers experience outages, "contingency thinking" is already second nature to most. Still, in any large enterprise resistance to this new way of thinking may be encountered. Two factors that have helped overcome the barriers to contingency planning are leadership buy-in, and responsive funding/resourcing. When senior leadership demonstrates commitment, not only to project implementation but also to the ongoing success of the effort, leaders throughout the organization tend to follow suit. At Johns Hopkins the genuine and sincere executive leadership commitment to this project has translated powerfully to the departmental and core process leadership. In turn, these leaders have echoed and expanded upon this message by clearly portraying continuity planning as another facet of effective patient care and safety. Actions speak louder than words, and few actions resound more thunderously than funding an organizational effort. The Initiative's Steering Committee members consider very seriously the funding and resourcing required for some plans to be viable and has pledged to fund most reasonable resource requirements for BCP development or implementation. Some of this is being accomplished over multiple fiscal years, some immediately.
When undertaking a detailed, department-level planning effort across a highly complex and decentralized academic medical institution, failing to coordinate actions with others can result in failure. "Planning in a vacuum" means that a department creates a series of business continuity tasks, but fails to consider the impact on other departments. This pitfall must be cautiously avoided. Here is one example. A stent is a wire mesh tube used to prop open an artery that's recently been cleared using angioplasty. Scenario 4 of the Corporate Purchasing BCP states that in the event of a vendor loss, stents will be purchased from an alternate vendor. This alternate vendor can begin supplying Johns Hopkins within 72 hour of notification and contract signing. Contract signing including legal review is fast-tracked in this case to a mere 72 hours. This means that there is a six-day gap in stent purchasing should the primary vendor become unavailable. Corporate Purchasing leadership knew stents were shipped daily to the Department of Surgery. The Senior Director of Purchasing coordinated with the Department of Surgery to ensure that a six-day gap would be feasible. The Department of Surgery, in turn, evaluated the on hand supply of stents at the Cardiovascular Interventional Lab. They identified that, at that time, there was a seven-day supply on hand and that a six-day gap was "cutting it too close." This led to two actions: First, Corporate Purchasing reduced the delay from six days to five, and secondly, the lab increased its stent stocks to eight days. This made the Corporate Purchasing plan viable for the Department of Surgery. It also led the Department of Surgery to realize it doesn't need to spend money overnighting stents every day. Now they are shipped more economically by ground whenever possible-an added bonus to conducting some vital cross-functional coordination. Interdepartmental coordination is emphasized at the onset of the project with all working groups. One of the key roles played by the Department of Operations Integration project facilitators is to act as cross-functional, cross-entity coordinators. A crucial aspect of the project facilitator's job is to ensure that ideas and solutions are shared among departments; especially those that have an impact on one another's operations. When the Department of Urology develops innovative solutions for operating room (OR) management in the face of an OR Management Information System (ORMIS) outage, these solutions must be shared and coordinated with other departments such as Pediatrics, Otolaryngology, and Surgery who are planning for the same contingency. This allows Johns Hopkins to capture internal best practices and streamline the workload for BCP developers. As a solution to overcome some of the obstacles associated with maintaining a BCP, a Web-based BCP repository is stored on the secure Johns Hopkins intranet. Password protected, this site allows departments to look at one another's BCPs and update their own. Additionally, the Web site has features that automatically remind plan authors or their designees that regular plan updating and testing is required.
One of the organization's largest administrative undertakings, business continuity planning at the Johns Hopkins Health System is a Board of Trustees mandated initiative with significant focus. The process began with an analysis of vulnerabilities to identify how and where the organization is most likely to encounter an interruption in the patient care delivery system or revenue capture mechanisms. This resulted in prioritization of mission-critical operations.The enterprise-wide initiative was then launched to formulate and implement viable recovery plans that ensure continuity of patient care services, physician education, and research efforts-our core missions. These strategies and plans are being written and developed by those who will need them in a crisis-the functional units, departments, and core processes that make up the health system. The business continuity planning effort throughout the Johns Hopkins Health System requires new thinking on the part of many in the organization. This new way of thinking quickly has become an organizational imperative. Through uncompromising preparedness, of which business continuity planning is but one aspect, Johns Hopkins will continue to care for tens of thousands of people.Gai Cole is Senior Project Administrator, Operations Integration for Johns Hopkins Health System (Baltimore, MD). He can be reached at (443) 287-5254. ©2005 Communication Technologies Inc. Reprinted from Continuity Insights Magazine.