Q: What issues are providers coming to you for help on?
Most of our work is with not-for-profits or with state and local governments. The core economic issue that many of these healthcare organizations face is that their costs are rising more rapidly than their reimbursements and it's a very capital-intensive business. They’re in a constant squeeze play between the need for more technology and new facilities and the pressure on the income side. We do a lot of economic improvement, strategic planning, prioritization, and organizational alignment work. Our sweet spot is the integration of process improvement with technology.
Healthcare is characterized by tremendous fragmentation. You see that fragmentation in the fact that you don't receive one bill for hospital services. You receive bills from physician providers, the hospital, the ancillary service provider, and so on. The revenue cycle is very inefficient in healthcare. We try to create more efficiency and cooperation, and so we've developed and implemented a solution which electronically connects providers of services with payers of services. A lot of the paper can be completely eliminated. It reduces the cost of a transaction many fold, is low cost to implement, uses open source software we’ve developed — but it requires cooperation amongst the participants that are traditionally adversarial.
Our benchmark for cooperation is a project that we’ve been involved with in Massachusetts for over 10 years to connect all of the players electronically. It's an early form of what is now called HIE or Health Information Exchange. The idea is to create a health information highway to connect all the different parts of the healthcare industry together electronically to support patient and financial information exchange. The government has had a major role in this through its Office of the National Coordinator for Health Information Technology (ONCHIT). It could eliminate a lot of administrative and bureaucratic costs. That requires participants in healthcare to agree to collaborate. The structure for that collaboration has been through the formation of regional health information organizations, which are new non-profit collaboratives formed from provider organizations and payer organizations who agree to share and communicate electronically together.
The first level of information sharing is financial information-sharing, particularly around payments, authorizations, eligibility. The second level is around clinical and patient information. If you are taken into an emergency room, you want them to be able to access your information right when you walk in the door. The portability of information, the ability to communicate that information is very important. And so we’ve engineered those systems for patient information flexibility and interoperability.
We have several pilots. But the most important initial emphasis in Massachusetts and in the other states where we’ve implemented these systems is to focus on medication administration, particularly around the emergency room, and providing a complete medication history on the patient to the physician. Just to give you a sort of brief snapshot of what the technology is, a pilot set of institutions in Massachusetts — major hospitals, physicians, and pharmacies — agree to a set of standards for information and security so that when a patient presents at a facility or at an emergency room, the system sends a request out to all participating institutions, locates information about the patient wherever it may be, and brings it back instantaneously. It's like an internet search methodology. You don't actually centralize the information. The information remains where it was created, so we call this a "federated" model.
It's more efficient from an organizational standpoint. These federated systems are much more appealing to the participants than centralized systems. People in healthcare are very concerned about big centralized databases of information. Added to that, it's actually pretty efficient from a technology standpoint to go get it. The search algorithms are good, and it's cheaper because you don’t have to maintain a central infrastructure. New York is promoting this kind of interoperability as well.
Interoperability is very appealing from a patient-service standpoint. The root problem is who pays for it. Until now all these different initiatives have been either state or federally funded as demonstration projects primarily to establish proof of concept. So there's not inherently an ROI [return on investment] sustainability model yet in place and this is causing a lot of concern as to whether the interoperability "movement" will ultimately succeed. Since portability of health information is a cornerstone of healthcare improvement, this is a big concern. So it's still somewhat up in the air as to how sustainable these initiatives will be. They are going on across the country. The only ones that have proven sustainable from a financial standpoint are the ones that reduce administrative and transactional costs.
Q: What are some examples?
We think there's a good ROI in terms of clinical patient interoperability such as e-prescribing where an automated network can manage the prescription drug process. When you see your physician, they can order the prescription right there. It can be delivered to the patients' home or their office. Refills and reminders all can be done on an automated basis. The key is to have good participation from the outset and a tightly focused scope that allows the participants to derive value.
Another promising development is clinical information systems. Many errors in healthcare are caused by incorrectly administered medications: the wrong drug, wrong dosage, or a drug that's prescribed that interacts negatively with another drug that the patient is taking. There’s a lot going on to create systems that order drugs correctly, dispense them correctly, and manage any potential interactions with other drugs before they occur.
Reducing these drug errors and improving quality is one of the main reasons many hospitals make the investment to implement clinical information systems. Maybe 25–30% of the larger institutions have some kind of system up and running. There’s a desire to spread that to many more facilities, community hospitals in particular.
Health information technology is considered by many to be a panacea. I'm skeptical about that. I think it’s great, but I don’t think that health information technology necessarily has all the answers, certainly not to the degree to which you see it portrayed in national debates. The traditional disciplines of process improvement, quality management using available data, silo-busting, and good resource allocation and management still can pay tremendous dividends. Better work flow automation and analytics can turbo charge these but not replace them.
Q: Where else is technology playing a role?
It’s really hard to know what's going on in healthcare. Who is a good provider? Was that quality care or not? Was a mistake made? Did the patient improve as a result of the care? Answering all these questions systematically requires data, and there are few tools today to do this but a tremendous desire to "pay for quality" — that is, tie payment to outcome.
The main mechanism for achieving this is to aggregate data and mine it and then tie the results to payment systems. Data mining is a huge trend right now within healthcare. It's a bit of an arms race for who will get the best information fastest. The government is trying to get information, so they're mining their databases.
The government recently decided not to pay providers for procedures or hospital days incurred as a result of mistakes that could have been prevented. This goes into effect and is the vanguard of many more efforts to pay for quality. Collaboratives, often pushed by employers, have come together to pool their claims data and to evaluate the data so as to identify "effective" providers of care. We have implemented the data warehousing technology for one big one, Care Focused Purchasing, that represents over 50 employers including my own, Computer Sciences, covering in total 30 million people. Payers are mining their data to support pay-for-performance efforts. The Blue Cross Association calls its effort "Blue Health Intelligence" — we support that one too. There is also more and more demand for a Consumer Reports-like scorecard. HealthGrades, a recently started company leads the way in this area – it's mining government data sets to provide scorecards for consumers. The CEO Kerry Hicks was the compelling keynote speaker at the 2006 SOM Healthcare Conference — and since then HealthGrades has grown by leaps and bounds.
Lastly, providers are mining their data so they can understand what's going on and reduce hospital acquired infections, reduce their readmission rates, or understand where patients are going once they leave their institution and go into a system of care. All so that they can survive in a coming period of intense scrutiny on results.
It's a fascinating period in which the major focus is on how to get data and how to utilize data effectively to understand performance relative to expected performance and then use that information. That all starts with a good understanding of what the baseline is and then using that information to improve the process.
The single biggest potential for healthcare improvement is around adopting an operations research framework for managing resources within healthcare. It's so fragmented and there's no systematic process improvement across the healthcare system, and as a result there's a lot of adversarialism among agents in the healthcare system as well as a "Swiss cheese" delivery model full of holes and handoffs that result in errors.
One way to find operational efficiency is to look at where all the parts of healthcare are under one umbrella. So a good example would be Kaiser Permanente. Kaiser has been and continues to be a leader because it includes within itself a health plan as well as a provider organization as well as an efficient physician services organization. They're wrestling with this stuff too, but they are able to look at the entire lifecycle of a person and manage resources effectively to optimize across the healthcare system. In Massachusetts, the Blue Cross plan is acting in many ways in that capacity by forging much closer working relationships in a so-called pay-for-performance model with its major participating providers. And that kind of very close coordination and cooperation between the payer and the provider, I think, is critical for managing that system improvement because you can align incentives.
An important area that's developing with the introduction of new technology is patient self-management. You want to use the information from better diagnostic testing for the patient to make changes in their lifestyle to be healthier. That requires the ability for the patient to have access to all their medical information — in a personal health record. At a certain point, if they get sick or develop a chronic disease and that needs to be managed, you want them as involved in that management as possible. You want to do that, as much as possible, in the outpatient setting. You want to reserve your inpatient and acute-care services for the least amount of time needed because they're the most expensive.
Kaiser is doing more and more in the patient management area, and has teamed up with Microsoft to offer personal health records to its employees and eventually its beneficiaries. The vision is to eventually link in remote monitoring – for instance, for diabetes and chronic conditions – as well as e-prescribing; and of course, ubiquitous access to health information. Moving to a healthcare model in which the consumer is at the center will be a long term trend in healthcare and essential to deal with the many health issues our society suffers from which include medical issues growing out of lifestyle choices and environmental problems.
Q: What other aspects of healthcare would benefit from improved management?
There's tremendous inefficiency in the supply chain. About a third of a health service organization's budget is made up of supplies and services. It's a very big part of the cost structure, and it's very inefficient in terms of which products are selected and used. Clinical, administrative, and financial demands coexist in an organizational agenda for improvement. In a multi-stakeholder setting a third party can come in and support improvement both from an organizational development standpoint and from a content standpoint. We bring best practices as well as technology solutions including trying to standardize the supply chain all the way out to the supplier.
Studies have shown tremendous operational benefits from optimizing the supply chain across an industry. In other industries a major feature of the business-to-business e-commerce boom was to create full, automated integration of the entire procurement-order-through-delivery process. Wal-Mart introduced and perfected the paradigm that when "you take the paper towel off the shelf they are cutting the next tree down in Washington." That is true supply chain management! Healthcare has been slow to adopt advanced supply chain techniques including e-commerce, collaborative commerce, and demand forecasting.
Q: How much inefficiency has been taken out of the supply chain?
Well, I would say just a fraction of what's there. The inefficiency is worth billions of dollars. The best study was done under the auspices of the EHCR [Efficient Healthcare Consumer Response] back in 1996. CSC was the consultant who supported this effort. It identified $11 billion in supply chain efficiencies that could be gained. I believe that generally the findings still hold; the number needs to be grossed up to account for the passage of time.
There's a lot of inefficiency in terms of delivering the right product to the right place at the right time because the supply-chain industry in healthcare is so broad and so massive. A lot of times the product is either over- or under-engineered. Product A is used when Product B might just as well be used. We work with institutions to evaluate products, relative to each other, for efficacy and make choices on the drug side as well as on the device side. Once you use standard products, you can order them much more effectively as a buyer. Particularly if you then join forces with other institutions and buy as a block. Another part of the inefficiency is that it takes a long time to get from the supplier to the point of use. The industry is moving to just-in-time, logistics-management inventory. Very few of those techniques are broadly deployed within healthcare with the same level of efficiency as in manufacturing or high-technology manufacturing and distribution. Typically when we work with a hospital system we are able to reduce supply costs and services by 5-8% per year. This is money that can be reinvested in patient services or in information technology, or in new imaging equipment.
Interview conducted and edited by Ted O’Callahan