EM Forum Presentation — November 17, 2010
GIS in Hospital and Healthcare Emergency Management
The Evolving Role
Ric Skinner, GISP
The Stoneybrook Group, LLC
Amy Sebring
EIIP Moderator
The following has been prepared from a transcription of the recording. The complete slide set (Adobe PDF) may be downloaded from http://www.emforum.org/vforum/GISDay10/HealthGIS.pdf for ease of printing.
[Welcome / Introduction]
Amy Sebring: Good morning/afternoon everyone and Happy GIS Day! Welcome once again to EMforum.org. I am Amy Sebring and will serve as your Moderator today. We are very glad you could join us. For our newcomers, we will be providing some instructions as we go along so you can relax and participate with us.
In observance of GIS Day, our topic is "GIS in Hospital and Healthcare Emergency Management" which is the title of a recent book that was edited by today’s guest. The book is available in both hard copy and electronic form, and the publisher, CRC Press, has decided to make a limited time offer on the hard copy in connection with our program today. We will have specifics at the end of the presentation.
[Slide 1]
Now it is my pleasure to introduce today’s guest: Ric Skinner is the owner and principal consultant with The Stoneybrook Group, and as I mentioned, editor of the book. He is a certified GIS Professional, and has been working in the field of "Health Geographics" since the mid-nineties. Most recently, Ric served as Program Manager at Baystate Medical Center -- the only hospital-based full time GIS department in the U.S., where he also served on the hospital’s Emergency Management Committee.
Please see today’s Background Page for further biographical information and links to related materials. Welcome Ric, and thank you very much for being with us today to celebrate GIS Day. I now turn the floor over to you to start us off please.
[Presentation]
Ric Skinner: Thanks, Amy, and I want to thank EMForum for the opportunity to participate in today’s recognition of GIS Day. I’ve been asked to talk about my book, and really the topic of the book, the material is "GIS in Hospital & Healthcare Emergency Management" which was published earlier this year.
[Slide 2]
So in the next 25 min or so I’d like to tell you a bit about the book and really more about how GIS is being used in hospital & healthcare emergency management.
This book is something I had wanted to do for a number of years but just didn’t have the time. Then in 2008, when I ended 36 years of working for other companies, I decided to start an independent consultancy. But as I’m sure you can understand the severe downturn in the economy which was unfolding in 2008 and into 2009 was certainly not the best time to start a new business – especially consulting. However, since clients were not lining up at my door, it was actually a good time to launch my book project to fill a lot of non-client time.
I researched potential publishers who had published books on GIS and on Emergency Management and selected one that I thought had a good process and marketing program. I also decided to be an editor rather than a book author because I wanted to let the experts in their respective areas do the writing. So I began to put the word out through my professional connections and numerous listservs and announcements in online newsletters such as Directions Magazine.
I asked for chapter ideas that would be relevant to the title of the book. At one point I had over 35 potential authors come forward with ideas and chapter abstracts. However, something else happened in 2009 which took precedence over a book – H1N1 flu. Even after some authors had submitted draft chapters, a number of them found it necessary to drop out of the project to focus on H1N1. In the end I had 16 chapters, including the ones I decided to write.
One thing that was important to me was to provide a high degree of professionalism in the chapters so I arranged for each chapter to be reviewed by at least two and in some cases three of the author’s peers from outside their organization. In reality, the book is a peer reviewed publication.
Another thing I wanted to do was to provide the reader with value beyond what the chapter authors provided, so I researched materials and developed a CDROM of hundreds of relevant documents and references which is included in the book. To make accessing the information easy, after taking an online XHTML course I developed a user friendly interface that is essentially a webpage.
[Slide 3]
It was an honor to have Dr. Tom Sizemore agree to write the Foreword for my book. This validated the book and it’s topic and illustrated how GIS in hospital & healthcare emergency management has support from the highest levels. This is an excerpt from what he wrote in the Foreword.
In the short time I have today I’m not going to spend any time on what is GIS or even how it fits in the broad topic of emergency management since you all probably have a pretty good understanding and experience in either or both of those areas, and there’s plenty to be found by Googling the internet. My contact information is provided on the last slide so you can always contact me as well.
[Slide 4]
Let’s talk about the topic and contents of the book. I divided the book into three sections, Concepts, Applications, Case Stories.
[Slide 5]
In the concepts section, you’ll find chapters on hazard vulnerability assessments and particularly the geographic aspects of that and how geography should be factored into the HVAs. There is an interesting chapter by Jeffrey Miller on using GIS in workplace and worker safety crisis management.
Another chapter is on infectious disease surveillance and how GIS can be used to help understand that. There is one on logistical support for healthcare emergencies.
[Slide 6]
In the applications section, there is a chapter on using optimization modeling in siting of trauma centers. There is another one on identifying alternate care sites that can be used during disasters for additional sheltering or evacuation points for hospitals.
The state of South Carolina has two chapters on an enterprise GIS which is used heavily by their hospitals and healthcare facilities. There’s another chapter done by Loma Linda University Medical Center in an enterprise system they have there for improving pre-hospital and disaster response.
[Slide 7]
Finally, in the case stories, there is a chapter by a hospital in Illinois that is using GIS to understand and help plan for upcoming pandemics situations so that they have resources that are necessary on hand. Another chapter has to do with the San Diego wildfires that occurred in 2007 and how GIS was used to help evacuate hospitals and nursing homes.
There is another chapter on using GIS in natural disasters and how it is used for evacuation planning—moving patients and supplies from one point to another and deciding the best routes to use. There is another chapter on mass casualty incident planning and then finally, one that is on an enterprise system that is integrating EMS and hospital emergency management systems.
[Slide 8]
As I’m sure most of you are aware, assessing vulnerabilities of a community, business or organization is not a simple process. The same is true for hospitals and healthcare facilities. It involves careful thought about how each of these hazards might impact people (patients, staff, and visitors), physical facilities and business operations, as well as how well the hospital is prepared for, has taken steps to minimize and mitigate impacts, and measures it can take to return to normal business operations as quickly as possible.
Usually, the hazards that are assessed are divided into three or four categories as shown here. Sometimes HazMat events are included in technological or human-caused event categories.
[Slide 9]
This an example of one way that a hazard vulnerability analysis is reduced to numbers to guide development of the emergency response plan so that is adequately addresses the high priority events for a hospital. A lot of the hazards have a geographic component to them.
What happens is, what you do here on this particular spreadsheet, for example, the probability for a particular event would be assigned a objective range of low, medium, to high—in other words, what is the probability of this particular event happening? Then, it is rated or evaluated against its impact, if it were to occur – the impact on the human, the property and business aspects of the hospital, and also how prepared they are internally and externally to handle that type of event.
At the end, there is a ranking that is created. There is a similar page for technological and human-caused, and so forth that has gone through. In the final worksheet of this particular Excel application is a summary of all the hazards and it gives you a ranking of the most predominant type of hazard that particular hospital will have to deal with.
[Slide 10]
Here’s an example of using GIS to determine the impact of flooding on a hospital. This map shows Midland, Michigan, and you have the MidMichigan Regional Medical Center located here and some nursing homes and schools. The Tittabawassee River, which flows through Midland, is located here. This is a normal river stage map.
[Slide 11]
When factoring in elevation and water level, we see a different picture in terms of the impact on particular features in the area. In this case, we are looking at the typical 24-foot level flood stage. That is the prediction for the river. That is when it reaches flood crest. You can see the expansion of the flooded areas.
In some cases you’ve got particularly schools that are impacted. The hospital is still in the dry area, but it is pretty much surrounded by water with one escape route to the northeast. In a case like this, the hospital might say they’ve made the decision to shelter in place. Twenty-four feet is really a disaster for the community but they are still high and dry.
The problem is, it could get worse than that. In fact, it did.
[Slide 12]
In 1986 the river crested at almost 34 feet – more than four feet above the previous record set in 1916 and 10 feet above flood stage. This is the predicted flooded area. Now, notice that more schools, nursing homes and even the hospital are in the potentially flooded areas. Knowing this, or having this predictive modeling capability, the hospital might at this point decide to evacuate.
[Slide 13]
Another example of location-based hazard vulnerability assessment is being able to map the locations of hospitals and where the fault zones, the active faults are located. This is an area around Everett, Washington. The darker orange or pink areas are areas where there is a potential for high level of ground disturbance—earth shaking, if you will. Hospitals would probably rank earthquake as a fairly high probability in their hazard vulnerability assessment. They would want to make sure it is adequately addressed in their emergency response plan.
[Slide 14]
Another way to use GIS in an HVA is for simulation and modeling. Here is simulated a chemical tank rupture on a rail line east of a hospital. The two colored ellipsoids represent the plume spread at a half hour and one hour based on certain weather conditions—wind speed, wind direction, other meteorological conditions, and the type of chemical.
If the hospital had this information, they might decide to evacuate the hospital. We’re obviously not going to go driving through the plume. With GIS, certain road barriers can be established, which then the GIS takes into account and creates a route that is the quickest way to get out of the plume and get over to another facility.
One of the powerful things about GIS is you don’t have to produce these maps and they sit for an hour or two. If the weather conditions were to change so that the wind was no longer coming out of the west, but maybe it was coming from the southwest and pushing the plume in another direction, having that advanced predictive information, the hospital might decide to shelter in place and see what would happen.
[Slide 15]
In this example GIS is used to do evacuation planning for a hospital that might be affected by a fire, hostage situation, probable flooding, so that the evacuation routes from the hospital could be developed—maybe a staging area created here for triaging patients or determining courses of action relative to particular medical statuses of patients.
Those that needed to be evacuated could then be moved over to the helipad and evacuated that way. Here we can place road barriers that could prevent traffic from coming down the road so that you have unimpeded access across the road to the evacuation point.
[Slide 16]
A hospital should know where its entire staff lives. Here a chemical tank explosion on a rail site. It could be accidental or a terrorist type of thing, and we have a predicted plume that is passing over the city. In addition to potentially affecting residents who may be in their homes, the little blue-green dots are hospital employees – trauma surgeons, ER nurses, environmental services staff, security, whatever—that the hospital may want to call in to help handle the victims from this kind of disaster.
The hospital needs to know that the yellow dots indicate hospital staff who may be in the plume and therefore could likely be victims themselves. That’s an important thing the hospital will want to consider in determining its staffing needs during the disaster.
[Slide 17]
Obviously, I’ve been talking about location and the effects on hazard vulnerability assessment. Do you think this hospital in Florida would have location as an important factor in some of its decision making?
[Slide 18]
A baseline decision-making model becomes helpful when establishing criteria for logistical support prior to an event occurring. This is part of the pre-planning and could be included in the mitigation in the planning. During the initial response effort for an event there will be little time for deciding among events and will be a period of extremely rapid supply consumption that will likely taper as the length of an event extends.
Any function that supports the delivery of essential services must be considered an element of the logistics chain. Much like preparation, mitigation involves the identification of common characteristics of hazards most likely to affect operations in supply chain management.
[Slide 19]
This graphic shows how the GIS integration of multiple information sources can provide insight during a disaster for a healthcare logistics manager. In this case we have a disaster occurring here with a downwind southeasterly direction of a plume or impact zone. A medical center is located near the impact area, and it depends on supplies from various locations.
One supply node is affected directly by the disaster. Another supply node may be affected because of the downwind effect. This supply node, which may be a critical supply source for the hospital can’t get to the hospital directly. They may have to route their whatever it is they are bringing to the hospital—it could be blood supply or surgical equipment—they may have to re-route that around the disaster area to get that to the hospital.
[Slide 20]
The difference between life and death for severely injured people often depends upon the amount of time it takes to get them to a trauma center. The siting of trauma centers, however, is more complex than just maps of land area with mileage bands around each hospital. The speed and location of helicopters and ambulances, the number and location of trauma centers in a region, and the spatial relationships between these facilities need to be considered.
In this example GIS was used in a project for the American Trauma Society to create a map using mathematical optimization modeling that uses population and access to existing trauma centers based on geographic relationships to ambulances and helicopters. This can be used to simulate new trauma centers, or the potential locations for new trauma centers to see how that will change the picture.
In this particular map, the purple areas represent 60 min. travel times for ground or air ambulance, to a trauma center. I don’t think I’d want to be in southwestern Texas or northeastern Nevada—there doesn’t seem to be a lot of quick response capabilities there.
[Slide 21]
Some GIS-based healthcare preparedness decision support systems can be very complex as illustrated by the Advanced Emergency Geographic Information System (AEGIS) created by Loma Linda University Medical Center and some of its community partners.
The AEGIS interface can be used universally at the response or command level, and across political and jurisdictional boundaries. It allows users to consider a diversity of pertinent information, interpret and analyze trends and threats, share information, and communicate with other responders.
It is especially suited for operations where a great amount of information from multiple sources must be distributed and analyzed quickly for appropriate decision-making. It is designed to be agency independent and can be used for daily operations as well as for incident response. That’s one thing I like about this system—it’s not the kind of system that gets turned on when there is a disaster. It’s functioning all the time.
You don’t have to change hats or change your focus or pull a different book of instructions off the shelf because you’ve started a new system. People are used to using this on a daily basis. They just ramp it up for more robust use during a disaster. The system is designed to provide secure access to information sources, and to expand and contract to include only the information that is necessary for the level of response needed and operational success.
[Slide 22]
Unfortunately, it too often takes a disaster to provide the justification for emergency management initiatives that seek to lessen the effects of future disasters. AEGIS is one of the outcomes from The Esperanza Fire in San Diego County in 2006 which resulted in a number of deaths, including 5 fire fighters who were trapped in Engine 57 near "Octagon House" shown here. The Fire Incident Report cited a primary causal factor for the deaths as a lack of situational awareness.
[Slide 23]
Situational awareness is knowing what is going on around you and how it might affect you. GIS is an essential asset for visualizing situational awareness. For example, being able to see a fire perimeter in time and space and appreciate its relationship to a community or its relationship to public safety personnel cannot be overestimated. The red dots on this map are locations of fire equipment and personnel at a particular point in time.
[Slide 24]
AEGIS incorporates both static and dynamic information into a single user-friendly interactive map. Static information includes a variety of typical maps with the location and attributes of key facilities and resources. Dynamic information includes hospital diversion status, real time traffic information, weather conditions, and updated major incident information. AEGIS also visualizes ground and airborne emergency assets in real time.
[Slide 25]
At the South Carolina Department of Health and Environmental Control GIS plays a critical role in all emergency response, preparedness, planning, and routine public health activities. Every hospital and nursing home in the coastal counties has an automatically created account when they are licensed by the state. When they log into this system they instantly have access to a map with the location of all state evacuation zones, hurricane surge zones, facility location information, evacuation routes, and other pertinent information in relation to their facility.
[Slide 26]
Here’s an example of one of the applications called the Critical Data Sheets application which all hospitals and nursing homes in the coastal counties are required to use. They must enter their shelter in-place and evacuation plans into it if they want to request the ability to shelter in-place for a Category III or less hurricane.
The application gives the hospitals the ability to access these plans from anywhere at any time and provide updates to the state in real-time while allowing the state to more fully comply with federal requirements. States also have the ability to do full reporting and spatial queries from any location on the plans currently in place for each facility.
[Slide 27]
In 2007, a massive wildfire – termed a "firestorm" -- destroyed nearly 400,000 acres of San Diego County, and more than 500,000 individuals were evacuated from fire-threatened areas. This is a view from space of the smoke plumes in California driven westward by the Santa Anna winds.
Evacuation of medically fragile individuals from residential health care facilities and two acute care hospitals using GIS to pre-plan the evacuation of threatened facilities and to ensure that patients were evacuated to comparable facilities or appropriate temporary evacuation points.
Critical to the multi-level emergency response plan was a GIS database containing information on medical resources such as facilities, ambulance agencies and shelters.
The locations of medical facilities, the number of patients housed, medical and transportation needs, contact information, and other information were all available. Furthermore, the GIS database contained data about the location and needs of other vulnerable populations including the young, the old, and those that were medically or physically disabled.
[Slide 28]
GIS-trained epidemiologists drew perimeters around rapidly changing fire zones, and predicted fire paths and so forth so they could do some advance planning. County EMS staff then called each facility to assess individual evacuation plans, determine transportation and receiving facility needs, and coordinate the deployment of ambulances and other transportation. The EMS mapping capabilities also provided real-time updates on facility status, shelter status and road closures.
As the slide indicates, in one day there were 2,100 medical evacuations completed successfully from two acute care hospitals, one psychiatric hospital, and 12 skilled nursing facilities. What contributed to the success was that each patient was moved with three days of medication, with their medical records, and with staff from the sending facility. There was a continuity of care provided that way.
[Slide 29]
All this information is gathered at the local, municipal, regional and state levels. A lot of it finds it way to fusion centers or fusion cells. Many states have these fusion centers. This happens to be what I call the "decision support wall" down at the Health and Human Services Secretary’s Operation Center in Washington, D.C.
There are a lot of maps that can be displayed based on information that is being gathered around the country and the world to help provide situational awareness of what is going on and a common operational picture for everyone who is making decisions related to any of the events that are going on.
[Slide 30]
So with that, let me leave you with my mantra and encourage you to think of what you do and the decisions you need to make in a spatial context. The opportunities to engage GIS may not always be obvious but I believe they are certainly endless.
[Slide 31]
If there are any questions, this is my contact information. The book publisher is offering a 20% discount if anyone wants to purchase a copy of my book between now and the end of the year, December 31. If you place the order, you use this promotional code—803DA. That will give you the 20% discount. With that, I will turn it back over to Amy to field any questions.
Amy Sebring: Thank you very much Ric. We appreciate that overview. Now, to proceed to our Q&A.
[Audience Questions & Answers]
Question:
Amy Sebring: Can you tell us a little bit more about the CD-ROM that accompanies the book?
Ric Skinner: The CD-ROM contains a number of documents, exercise plans that are relevant to hospitals. There are quite a few documents and links to other websites that have information about GIS in the context of health or hospital emergency management. There are probably several hundred links or documents that can be reached from that CD-ROM.
It is reference materials, and also color images of all the figures in the book. The book is published with black and white figures, but with a map, you’ve got to have color. That’s one of the reasons I wanted to put them on the CD-ROM. All the figures also have a color version on the CD-ROM.
Question:
Amy Sebring: You’ve been around long enough to see advances in the technologies, especially the Web services. Any comments on how that is evolving and where it might be headed for the future?
Ric Skinner: From a hospital’s perspective, GIS is developing to the point where it can be available to almost anyone who wants to use it. You don’t have to be a multi-year trained GIS professional to work some of the GIS software that is out there. There is everything from some of the ESRI products, the software that is not GIS that requires some training and there is cost involved.
There is the WebEOC approach to emergency management. Those are on the upper and middle end. Then there are other systems like Google Maps that can be done. There are applications that can be developed that way. The Depiction software is another one that I’m finding is very user-friendly way to produce maps and support emergency management and collaboration.
Over the years, the range has broadened, so you almost have a GIS depending on what your needs and budget are.
Question:
Dwight: Where did you get the fault line data for the Everett, WA, area?
Ric Skinner: There are some websites. Most of the data I get is open source. I can’t give you the URL right now, but if Dwight will email me I will email him back the link. It is the Washington State Environmental Division, or something like that. There is a whole list of spatial data for Washington—the faults and shake maps, liquefaction zones—all that stuff is available.
It has become a much more open source, which is good and bad. Just because it’s free doesn’t mean it’s good. You have to know what you are accessing and what the original source is. It makes it easier for generating maps that meet your needs without having a pay a lot of money for data development
Question:
Isabel McCurdy: Ric, how up to date is the information on these GIS maps, e.g., road name changes or closures?
Ric Skinner: If you’re accessing open source data, like Open Street Map, that is updated basically by the user group. I mentioned Depiction software—it uses an open source map, Open Street Map, as its primary map base. That literally can be changed almost on a daily basis, or updated, I should say.
Other sources, where you are getting data on a DVD or CD-ROM or downloading it from a state website, like MassGIS—Massachusetts has a very rich set of data, but it may only be updated once a year or every other year, depending on the type of data. A lot of data doesn’t change very much, but you need to know what you’re asking for.
It’s like the aerial images you see in Google Earth. Those images are not real time. They may be a year or two, or even three years old. You need to keep that in mind if you’re looking for current information.
Amy Sebring: You also have the capability to update. You’re basically talking about your base maps, your data layers—but then, on top of that, you can input in real time your evolving situation, right?
Ric Skinner: If you’re mapping things like movable ambulances or helicopters, or something like that, your map can capture the GPS coordinates and literally provide updates in real time. Or, if it’s a hospital reporting bed counts or ER status—that information can be captured in almost real time and displayed on the map.
Question:
Amy Sebring: I would like you to elaborate just a little bit more about using GIS in epidemiology and disease surveillance.
Ric Skinner: I had a whole set of slides that I started but I took out of here because of the time factor, but really the GIS and epidemiology started back in the 19th century. The first example being used is the cholera epidemic in London.
While there wasn’t a computer back then, Dr. John Snow mapped on a paper map the locations of cholera victims, the locations of public drinking water wells and determined there was one particular pump, the Broad Street pump, which is where most people were getting their water. Those were the people coming down with cholera. When he removed the handle of the pump, the cholera epidemic subsided.
That was the first often-reported version of epidemiology and geographic information. Today, we still use some of the same thinking that Dr. Snow used in the 1800’s—that is looking at cause and effect based on location. That is really the foundation of a lot of epidemiology.
Question:
Avagene Moore: Ric, are you finding more emergency managers and healthcare facilities using GIS? If not, why not in your opinion?
Ric Skinner: There are not as many as I would like to see. I’ve been involved with GIS and health for 15 or 20 years and it has been a very slow process. As you mentioned, at one point I was at Baystate Medical Center and for 10 years we were the only hospital that had a full time GIS department. We used it for many different things—epidemiology, the routing of vehicles, health care resources, security issues.
I think it is being used more and more, but it is a slow process in terms of the healthcare preparedness sector. I think a lot of it has to do with the presumption that you need to have a skilled, dedicated GIS professional to run the system for you. Because of what I said earlier, a lot of the software is coming to the point where that is not necessary. You could have somebody—it takes a little bit of training—but it can produce intelligent, actionable maps to support emergency management decisions.
Question:
Knox Andress: Can you talk about potential use of GIS in Mass Fatality Planning?
Ric Skinner: I think the way that would be used would be if you have mass fatality, you’ve got more victims than what you can send to any particular hospital. If you have a GIS that is tracking regional hospitals, whether they are a sub-state region or a multi-state region, but in terms of their capabilities—whether their ER is open or closed, whether or not they have a pediatric burn center, whether they have a decontamination unit—that sort of thing.
That information can be fed to the primary scene where the casualties are occurring and used in the decision of how to get victims from point A to a medical care facility. Tracking the patients to those facilities, and oftentimes they get moved from one facility to another, so GIS helps track that information.
Down the road, when family members want to know, "Where did Uncle John go?", you can pull it up on the system and say that he initially went to hospital A, but now he is at hospital B.
[Closing]
Amy Sebring: Time to wrap for today. Thank you very much Ric. We appreciate your taking the time to be with us today and share this information. We do wish you much good luck with your book.
Contact information: Ric Skinner, GISP
The Stoneybrook Group LLCThe LLC
Sturbridge, MASturbridge, MA
[email protected]
http://www.healthGISguy.comwww.com
http://www.linkedin.com/in/ricskinnerwww.ricskinner
Again, the recording should be available later this afternoon. If you are not on our mailing list and would like to get notices of future sessions and availability of transcripts, just go to our home page to Subscribe.
Now PLEASE take a moment to do the rating/review! I am going to load the rating/review form into Live Meeting so you can complete it on the spot. Note: We are asking you to rate the relevance of the information, and this will assist us in our future programming.
Our next program will be on December 8th. Please make plans to join us then, and in the meantime, Avagene and I wish you and yours a safe and happy Thanksgiving holiday!
Thanks to everyone for participating today and have a great afternoon.