Sarah Kramer is the provincial vice-president and chief information officer of Cancer Care Ontario in Toronto. CCO is focused on improving the quality, accountability and innovation of cancer services, and is rolling out a wait time information system to support the reduction of wait times in the province. Kramer has a background in public policy, with experience in the civil service, health-care sector and private sector. Before joining COO, she was the CIO for the Department of Health in Nova Scotia.
Timothy Beasley is the program director for health surveillance with Canada Health Infoway in Toronto. Infoway’s vision is to build electronic health records and promote their use in Canada. Beasley runs the Public Health Surveillance Program, which has been added to Infoway’s mandate as the result of a funding agreement signed with the federal government in 2004. Prior to that, Beasley worked as an IT consultant to public sector clients in both Canada and the U.S., and helped build surveillance systems for the U.S. Centers for Disease Control and Prevention.
Catherine Aczel Boivie is senior vice-president of information technology with the Pacific Blue Cross in Burnaby, B.C. Pacific Blue Cross is a not-for-profit organization that extends health and dental benefits to groups and individuals. About half its claims are automated and never touch human hands. Prior to this, Boivie was CIO of the British Columbia Automobile Association and CIO of the City of Etobicoke, Ont.
Computing Canada: Could you discuss your technology platforms in terms of hardware, software and databases?
Beasley: We’re investing in a health surveillance solution that we hope will assist all the provinces and territories in Canada, and that solution is going to have a Web-based user interface and N-tier architecture. It should be database-independent, (so) in other words it ought to be able to run on Oracle and DB2 and SQL Server. We’re in the process of selecting a winner in a request-for-proposals process, so I can’t be too specific about what the eventual solution will be because we haven’t selected it. We’ve received bids from a number of integrators who have proposed different off-the-shelf products for functions such as case management, immunization registry, alerting and outbreak management, and they’ll be integrating these into a fairly integrated suite of modules that would then be implemented in the jurisdictions.
Boivie: We use Oracle as our main database.
Kramer: Ours is a mixed bag because we’re a very small organization. We are investing a fair bit in data warehouse technology, which is our main platform. We have a MicroStrategy and Informatica ETL database environment that is just being built and we (launched) our first data mart set just a little while ago, which is our main function at Cancer Care Ontario. We’re in the midst of building out tools that are hospital information system agnostic, so we can grab and provide information back and forth in the health-care system, and we have a number of ways of doing that. We are looking at consolidating some of our own technology so that we’re not launching separate applications every time we move forward. We have a server that does a lot of the main functions with the specific functionality built out, but we don’t actually run a technology platform for health care – we advance the use of it.
CC: Is integration the No. 1 challenge?
Boivie: Integration is a challenge in our area as well, but we also have a challenge integrating the legacy systems to something more flexible (that will) accommodate our business requirements and goals. We have been looking at how to integrate our existing systems so that we don’t spend as much time on the interfaces, but are able to channel our resources to get different functionality and further our initiatives.
Beasley: If you can’t really integrate a couple of systems, at least do a cross-walking so you can feed data from one into the other. In public health there are a number of layers, (such as) the public health nurses who do case management on the front lines. If you look at individual people with hepatitis or AIDS or whatever communicable disease, (nurses) collect the details, and the trick is to get that data aggregated up to the level where the epidemiologists can deal with it on a population basis so they can spot trends and patterns. This is a field where the interfaces and aggregation . . . and all the other steps that have to happen in that process is problematic and expensive, so it is a challenge.
Kramer: I think from a technology perspective, integration is the biggest challenge, but from a sector perspective – probably more so in Ontario than elsewhere in Canada – the strategic goals and business focus (are) actually the bigger challenges. Technology will come along and has actually come along from an integration perspective. The business focus on what we need to do, the five most important things to do in this sector from an information perspective, is a huge challenge. We either have huge gaps or we have lots of people running down a lot of parallel roads and not really advancing the system as a whole. Ontario is renowned for that because of the way we’ve got health care organized, (which is) not very regionally focused.
CC: You read about electronic health records and reducing wait times, but what needs to happen before you can achieve that?
Kramer: It’s not a simple task, but it’s not the most difficult challenge either. There are more difficult technology questions than this one. (The) electronic health record is a big question. What is an electronic health record? Who contributes to it? Who owns the information? What are the privacy issues? Where are the users in this? Where are the physicians and other clinicians in defining what it is they need on the front end instead of us focusing all the time on the technology back end? Canada Health Infoway has done a huge job in moving that along and all of the provinces are working hard at that, but I think that’s where some of the more negative (headlines have come from), I think that’s the impetus. There’s been a lot of money spent on the technology, but the uptake isn’t there yet . . . it isn’t all that clear who is supposed to take it up and how it’s supposed to be used to improve patient care.
CC: Is it the same thing with case management, where there has to be a lot of work done before something like that can be implemented?
Beasley: With case management you’re dealing with people who are at the retail end of the business, (who) are dealing with the individual tracking of sick people. You really have to design systems that are user-friendly and will get their buy-in, otherwise the systems won’t get populated with the data elements that they need to be populated with, so the whole thing will collapse because no data will get fed in at the front end. To design or to select off-the-shelf case management products, you really need to directly involve the users who are going to be using these things all day. This isn’t rocket science, it’s not sophisticated bleeding-edge technology – we’re just talking about storage and retrieval. We’re not talking about artificial intelligence or anything all that exotic, but we need to have a reasonable set of screens that people can enter data into, we need to have flexibility built into the workflow so you don’t have to always navigate A, B, C. It can be A, D, G or different routes through the user interface. That’s what a lot of users complain about – some case management systems are too rigid and (users) need to be able to customize their own paths.
Boivie: There are (business intelligence) tools to assist in making better decisions, and we have been using that in order to access data from a number of systems and bring it together in a cube where (we) can use the information available more readily in making decisions.
CC: Is it as simple as having one large database with lots of people accessing it?
Beasley: Public health has a (relatively) free rein compared to other health domains. If you have a communicable disease, your consent is implicitly given for public health authorities to access data about your case. You can’t say no because you might pose a threat to the population. But still, you don’t want people accessing any data they don’t need to, so there needs to be security and encryption and all those kinds of technological fixes built in to ensure there’s no unauthorized access or use of data. The rubber hits the road when the solution is implemented in an organization and what business rules and procedures they apply to how they’re going to use the system. You can’t build in any set of safeguards that will guarantee there’s never going to be any unauthorized disclosure.
Kramer: From a business perspective, it’s not one business, it’s a lot of different businesses. It’s a lot of different providers and users. People don’t have accountabilities across the line, and that’s changing but it will never be complete. This is not England where every doctor ultimately works for the Minister of Health. This is a system with a bunch of players and to envision a single database even beyond the privacy issues, which are huge, from a user perspective it isn’t a realistic goal. I think it sets up a mountain we can never climb, as opposed to saying, ‘You have your information sets for your own purposes, let’s work at the integration and interface level, the data centre level and overarching where-do-we-want-to-go from a systems perspective’ and focusing on those.
CC: It sounds like technology is not the problem and other issues are coming to the fore.
Kramer: There are(elements) in technology that need to be improved, but we’re not bleeding edge.
Beasley: I think Sarah’s let the cat out of the bag – the dirty secret that we technologists are reluctant to admit – that our jobs are the easy ones. It’s (relatively) straightforward to build or buy the technology. The real challenge is political (and) organizational, to put the agreements in place to use standards, to use common procedures, to call things by the same name. They’re huge obstacles.
Kramer: I got that when I was in Nova Scotia. Most of the work we spent in implementing a province-wide lab system was in getting the pathologists and technologists to agree to a dictionary of lab terms. So imagine what it’s like in Ontario.
Beasley: Another problem with health is not a lack of standards but too many standards. Having a plethora of standards is the same as having none. You’ve got competing ICD10 and Snowmed and all these different standard terminology sets – sometimes when the systems are incompatible, it’s not because of the software or hardware, it’s because of the terminology they use and embody.
CC: Presumably you’re either part of a senior executive management team or you’re dealing with senior executives. How do you get them on side?
Beasley: The challenge in my job is 10 per cent technological, 10 per cent financial and 80 per cent political. Surveillance is a good idea and we’ve got a lot of money to do this and the technology’s there, but it’s putting the deals in place that will result in projects being carried out and successfully completed and systems being implemented. You don’t frequently run into anyone saying no, but you run into the need for extensive negotiations over the plan and how you’re going to execute it and what the path is you’re going to take.
Kramer: I sit on the senior executive committee and I did in my previous job. One of the biggest challenges in health care – and it’s a good challenge – is every dollar you put into technology, you’re not putting into a nurse, you’re not putting into a new drug, you’re not putting into treating in a new way. We shouldn’t kid ourselves, these are really challenging questions, and when I was in a hospital I started to ask those questions every time an investment came up, and I expect my colleagues to ask these questions of me as well.