Who Will Cut The Gordian Knot of Healthcare Billing?

In a story about open source medical records systems, I couldn’t help but be struck by the irony in the following statement:

Referred to by health care quality guru Philip Longman as an “unrecognized national resource,” VistA’s open source code is constantly being improved and updated by its users. However, John Halamka MD, Chief Information Officer at the Beth Israel Deaconess Medical Center in Boston, is quick to note that VistA is not designed for complex billing scenarios that challenge large hospital systems because the VA is a single payer system unlike the health system for the general public.

It’s true that VistA is designed for hospital management and patient care rather than billing, but isn’t it a sign of something wrong when the billing tail wags the dog of care?

For so many problems in our society, solutions are dismissed as impossible because they would require changes that people don’t want to make. That’s why change so often comes from outside. Perhaps the simplicity of VistA is a feature, not a bug. In its early days, the internet was cited as inadequate — too lightweight for serious networking — by proponents of complex, over-built systems. Where’s Alexander when we need him? Gordian Knots are everywhere.

  • Electronic medical records have benefits of instant access, but right now, one of the problems is getting your doctor’s office records, test results, etc. integrated with a larger hospital system,(assuming your doctor has switched to an EHR) and making records between hospitals and their individual systems easily compatible. Say you get sick and you’re on vacation- can you really get easy access to your medical records? While we’ll get to the point where you can carry your records around with you including digital imaging, (and this is available now to some extent) it’s also currently a mess of individual proprietary formats, and not enough agreement as to basic standards like we have for URLs online.

    Billing needs to be integrated into the EHR because billing is currently based (in the US, at least) on diagnostic codes used by medicaid and medicare. These diagnostic codes are also part of the EHR- it’s an integral part of how things like tests are ordered and tracked.
    It’s likely that what will need to happen is a system to be built for medicine and health records that operates like the web itself- universal “grammar” like the composition of URL’s, file transfer protocols, etc. Maybe even a parallel medical internet would work. This may be the solution that would allow adequate security and privacy of records, yet make them all universally exchangeable- it’s a dream, at least.

    Physicians are great at practicing medicine, and some of them are even experimenting with new media as a way to reach out to their patients- but what they need are ways to make their practice faster and easier, not any more complicated. But I think many of them are hesitant to adopt changes because on an individual basis, any change is expensive to their individual practice, time consuming, and until the standards are agreed on,there’s little incentive to do so. Why should anyone lay out the money, effort and time required to implement a system that may be equivalent to betamax in another year?

    The biggest thing we need are docs who are also “geeks” who can help the programmers really understand the needs of the doctors, nurses and patients on the front line, and make the code meet the need. Too often the geeks and the docs need someone who know both sides and can act as translator- because too frequently, what a client wants and what’s possible may not be the same thing.

    A classic example was at our local hospital, where the administrators decided to move a ward because they had other plans, and did not realize the ward was located where it was because of its proximity to operating rooms, required as a standard of care. As a result,the hospital had to go through the expense of designing and building a special high speed elevator, including time trials of same, to make sure the relocation of the ward would still allow patients to get to the operating room as quickly as necessary. So what seems like a simple administrative decision about where patient’s rooms would be located had medical and legal implications, all caused by all sides not understanding the others requirements before they had acted.

    While this is a non-digital example, you can easily expand it into the EHR arena. What if the data and pictures turned out by the CT Scan or ultrasound can’t be added to the EHR, or not read on the office computer? What is the resolution is bad and leads to a misdiagnosis? What if the lab results get misreported or aren’t added correctly by hand into the EHR? There are even problems with EHR’s not being able to adequately flag concerns to make sure the doctor addresses them at a visit.

    While we all want this problem fixed tomorrow, and those of us with a geek bent can see that we have the technology capable of doing so, the roll out in a piecemeal, proprietary format way will cause more problems than it will solve, with actual medical impacts for patients and doctors alike.

  • The real Gordian knot is the strangling web of government regulations that limit all health care in this country. Cut through that by getting the government out of healthcare, and you’ll quickly see entrepreneurs step up to solve these problems.

    Unfortunately, I’m afraid we’re going to move in the other direction instead–to a socialized medical system that will gain the same reputation for efficiency, innovation and customer service as Amtrak and the post office.

  • Jason,

    Could you be specific about “the strangling knot of government regulations”? I’m sure they exist, but I don’t know what they are. What I do experience every day is the strangling knot of insurance billing. It’s nearly impossible to know what you’re really expected to pay, and it’s clear that different people are charged radically different amounts for the same services.

    While it is certainly true that regulations can make things more difficult, it’s not just government regulations. Private companies are just as good (or better, er-worse) at inventing bureaucracy, and without any public good in mind.

    In any event, if you have real knowledge of the field, rather than just libertarian ideals, please supply details. Thanks.

  • In my opinion, one thing that government can do to accelerate electronic record keeping in the healthcare arena is mandating that solution providers must make means available to both import and export data to/from their systems. Mandating a specific format is less of an issue and should not be the main goal.

    If vendors were obligated to open their systems and encouraged to do so via some [fiscal/tax] incentive then the problem would solve itself. IT professionals either on their own or contracted would create solutions that connected the hodgepodge of systems out there.

    These systems are infinitely difficult to implement on any scale (by scale I mean number of people using the system – not technical challenges). At the moment I am tangentially involved in a major enterprise installation at a very well known institution. From my vantage point one of the major stumbling blocks is the difficulty in which the many, many systems already in place will interact with the new system (and with each other). The inability of these systems to interact with each other directly affects the cost of doing business in the modern day hospital.

    Many of the people in positions to make purchasing decisions are not fully cognizant of the concept of “vendor lock in” and “data portability”. These poor decisions are paid back many fold not only in cold hard cash but also in opportunity cost over the lifetime of the system purchased, which may very well be decades. Government can make this all go away by simply requesting that all players in this space provide mechanisms for programmatic, unfettered data access. Providing fiscal incentive will all but ensure that those providers in compliance will be preferred.

    Mandating a specific system, or data exchange format is not the solution or, frankly, the “American way”. Evening the playing field by ensuring openness through financial incentive is much more palatable to a broader spectrum of participants. The market place will ultimately decide and coalesce around its own standards and formats.

  • Tim, you asked about Jason’s comment: “the strangling knot of government regulations”?

    As the Whitney has already indicated, one of those is the us of Diagnostic Codes for treatment. These are not just labels for treatments, but part of a labyrinthine set of rules about which codes can be used, in which combination and which order. There are people who are employed just to manage which code will be assigned to which treatment to maximize allowable billing. This is not just the insurance industry problem, the correct codes must be used for the government paid Medicare program which typically under reimburses patient costs and which the government has absolute buyer power.

    The best way to understand this is to think about how your own company would deal with a printer for your books. Imagine that the authors submit books directly to the printer when they feel they are ready to do a print run and that you pay for the printing, and printer trying to maximize returns by deciding which print format, paper and inks he would use to print the books. The authors would want the most expensive quality, and the printer would want to give them that while working under some sort of rules that your company imposed to control costs. Inevitably you would end up with a complex set of rules for the book quality and how errors would be handled, which the printer would game to maximize his returns. You don’t do it this way. You maintain control of book supply and you both manage the complexity level by limiting the universe of choices and standardization.

    There is indeed a Gordian knot to be cut, and this is the billing problem to which you allude to. This will not be solved by any of the proposed healthcare solutions currently being promoted. It will require that healthcare is not delivered in an “adversarial” way, i.e. that private providers like doctors and hospitals have to deal with a funding system, whether private or public that seeks to limit reimbursement costs after the fact of service delivery.

    One way around this problem is to go with a government run system like the UK’s NHS, where the government both supplies the funding and controls the system. This would not be a popular way to go in the US, and it is not particularly popular in the UK either unless you have no recourse to private insurance.

  • To cut the knot I suspect you’ll need to take a closer look at the tax code, rather than any particular set of regulations.

    The allowance made to companies – but denied to individuals – for the deduction of health care expenses has had the effect of concentrating buying power in the hands of group plans, and the organizations that participate in them.

    Inevitably, costs are transferred from those with strong plans to those with weaker plans, or no plans at all. This has led to the creation of mechanisms that price services based on evaluating consumers, not services. “For you, my friend” is now the order of the day.

    The results are ugly – and go well beyond the nightmare of hospital billing.

    In the employment market, this has created a situation where some companies are able to use their negotiated access to the health care system as a form of scarce currency. Because the “bill by person, not by service” system has spiraled so far out of control, companies can limit their labor costs by including health care access in their compensation plans.

    Between the discounted rates, the additional tax breaks, and the inflated costs that fall to those outside these arrangements, employees find that these plans are extraordinarily valuable, in that they’d have to pay two to three times the amount the company pays if they were to meet their health care needs on their own.

    On the surface, it seems like a good deal, but a number of very perverse incentives result. For one thing, this only works well for companies that hire on a salary basis, or for hourly workers who are unionized. Independent contractors and non-union hourly workers (who have the least ability to pay for anything at all) are thrust into the highest cost bracket. Needless to say, this results in an explosion of expensive and chronic conditions that could be prevented with reasonably priced access to basic, preventative care.

    Even within the salaried world, the outcome of this arrangement is less than ideal, as decisions that would normally be made purely on a career-assessment basis (changing jobs, fields, returning to school, joining a start-up, taking time off to care for family, etc.) are suddenly filtered through the potentially life-or-death consideration about maintaining access to affordable care. Just imagine, if you will, having your career options limited by the question “yes, but will I still be able to get into the grocery store, or buy gas for less than $10 per gallon?”

    Combined with the horrific legal provisions that allow group plans to capitalize on breaks in coverage by denying care for any pre-existing conditions, you find that large employers (along well as unions) are given an extraordinarily unfair advantage in managing and negotiating with employees. Wage stagnation is just one of the results. Severely curtailing the ability of employees to leave abusive, overbearing, or simply mismanaged operations is another.

    The result is that bad companies are able to retain talent that would otherwise go elsewhere, while the startups or transitional ventures that could attract them are faced with a daunting – and totally artificial – barrier.

    This is a loss for the economy at the best of times. When faced with the need for serious restructuring before a solid recovery can take place, it’s especially damaging to the national interest.

    Another perverse outcome is the loss of effective market pressure within the medical profession itself. With buying power totally centralized in organizations that care overwhelmingly about cost of service and not at all about quality, it’s no surprise that the average physician visit is over in a matter of minutes, with a focus on what the doctor can prescribe, as opposed to the health benefits he can produce. This only compounds the overall lack of health that has come to define Americans.

    Moreover, the total marginalization of customers who can buy (i.e. exert market pressure) based on quality of care, means that the only competition group plans face is from other groups with the same perverse incentives. And since the only way people can leave groups is by quitting their jobs, they have little incentive to change for this reason – especially when those same people will have to rejoin another group dealing with the same economics before they can regain access to the health care system. Normal market pressures are muted entirely.

    Individual plans aren’t much help either, since the transaction costs for insurers are exponentially higher than those for employees whose companies bear a lot of the costs via payroll deductions and consolidated payments. Given a choice, most insurers wouldn’t offer them at all. Politically, this would be unacceptable, so they’re available, but priced at very unattractive levels, with the bar for rejections based on health conditions set very low. The message is clear – join a union, get a salaried job, or get married to someone who has either of these. Otherwise, get lost.

    In short, not only is the tax code at the heart of the atrocious billing practices, it’s also a prime mover in a wide range of social ills that stem from a fundamentally inequitable situation.

    Politically, it’s unrealistic to expect that the corporate care tax deduction will be eliminated (it’s in the same third-rail category of middle-class entitlements as mortgage interest deductions). However, if it were extended to everybody, we could sever the link between employment and membership in group health plans. And since you’d no longer have to quit your job to change plans, these non-employer based options would have to compete on quality as well as cost, and not just provide rudimentary access.

    Presented with this option, the health care industry will raise the straw man argument about people who game the system by not participating at all until they hit their thirties. They’ll point out that, unlike home or auto insurance, there’s no choke-point like a mortgage or registration where they can demand early buy in, which is essential to any viable insurance operation.

    But this concern could be resolved by creating a positive incentive instead. Specifically, the creation of individual tax free health-savings accounts that people would be foolish to pass up. Combined with discounts for services paid from these accounts, there’s no reason to think that insurance companies couldn’t generate a positive incentive to participate in the system. Combined with direct government subsidies for individuals with so little income that ‘tax free’ is no incentive at all, we should be able to come reasonably close to universal coverage.

    Moreover, we’d do so in a manner that preserves all the benefits of a free and open market. Because let’s face it, if you’re comfortable with single payer health care, you probably think that government is fair, efficient, and reasonably uncorrupt.

  • Tim,

    Furthering Alex, Whitney, and Jason:

    I think the state of the debate around health care is yet another example of the perils of “cheer-leading” but in this instance, perhaps until recently, not so much your cheer-leading.

    Several political constituencies have come together around the question of “health care reform”:

    Social justice folks concerned with the non-universality of access to health care want something – anything – to fix that problem.

    Overburdened professionals and the fiduciaries of taxpayers concerned with the “abuse” of emergency rooms want increased access to health care, especially preventative care.

    Big business wants to escape from the cost of health care benefits for employees and especially for retirees.

    Public health officials want greater access to and use of care and greater “monitoring” of the public health.

    Some with social democratic beliefs want a representational democratic government to have the market power to dominate the prices of health care products and services.

    Venture capital investors eye the seeming economic inefficiencies of the system we have, believe that new products and services can out-compete what is out there. They look to the other interest groups’ as a source of marketing points.

    UN types such as those that originated the catalytic W.H.O. report about which I have no polite characterization to offer.

    Punditry (hi Tim) who sense that action is afoot and are looking for something to say.

    Now, with those groups interested in the “health care problem” you can kind of see the differentials of relative political influence, yes?

    In the first few years of the 21st century a rumor (and it is nothing more or less) arose mainly among the punditry and the investors that there was some “sound basis” for believing that if only everything used the VA’s IT system or something like it, all of those constituencies could be easily satisfied. It would lower costs and thus promote universal access and take heat off of emergency rooms (our de facto clinic system). It could be wired up to promote centralized monitoring. Per some (bad) science it would “obviously” result in better care at the level of the population. etc. A huge cheer-leading-led hype bubble arose around this flimsy notion.

    Recently that pseudo-consensus has started to fall apart but not before becoming integrated with, for example, the platforms (now “agenda”) of the current administration.

    It has started to fall apart because (a) of privacy concerns and (b) of concerns of how such a massive scale IT transition can be funded. The notion that such a transition be funded by “stimulus” dollars is falling apart.

    Sensing this, the sentiment among the biggest influencers has been to look for some alternative, simpler justification for federal subsidy of new ventures in health care IT reform and, Tim, you ain’t the first one to suggest “billing” (and to do so with no understanding of what goes on in real life, for most people).

    So you get your Tim’s and your Esther’s and such waving around incomprehensible bills, and anecdotes of over-billing, and tragedies of misunderstood billing and such and saying “see? problem! give us leverage over the entire system and we can fix that.”

    Well, in my ideal: no. The new move of the influencers is just as much rubbish as the first push to turn the entire system into the VA. It’s ungrounded in solid science. It’s opportunistic self-promotion on a grand scale and without concern for tolls it takes human lives or even the public health. It’s especially offensive these days because it is so tightly tied to implicit pressure to incur and spend tax-payer debt on these private profit ventures. It’s reflexive “cheer-leading” about “something perceived as likely to happen and benefit the cheer-leader’s social class” rather than any concern about, oh, say, um… the quality and accessibility of health care.

    There is no Gordian Knot to be cut here! (in my ever so humble opinion). The current system has a lot of virtues. The “adversarial” and semi-random nature of billing? Yeah, that saves lives: providers cheat the system to save lives and to collect small subsidies from the well off. That will be hard to improve upon.

    Weaken (a bit, not entirely) drug patents. Let “moore’s law” type things continue to lower the cost of drug discovery and early trials. Perhaps give a tax credit in some progressive way but a restricted tax credit (refunds due to the health tax credit may only go into a health savings account). Get the vcs learning how to do lots of small investments in clinics – with a willingness to double down and fight regulatory issues (public and private) that get in the way. That’s efficient and good as it can get (and, from what I’ve seen, would be quite good and improvement.)


  • @Tom “The current system has a lot of virtues. The “adversarial” and semi-random nature of billing? Yeah, that saves lives: providers cheat the system to save lives and to collect small subsidies from the well off. That will be hard to improve upon.”

    I think this needs more explanation. Does it really save lives, or are the same lives saved at a higher cost? Is this really a better system than others, given what we know of other health care systems, their costs and outcomes?

    However, I do have an immense distrust of “global solutions”, they often turn out to be prime examples of Murphy’s Law.

  • @Tom “The current system has a lot of virtues. The “adversarial” and semi-random nature of billing? Yeah, that saves lives: providers cheat the system to save lives and to collect small subsidies from the well off. That will be hard to improve upon.”

    I think this needs more explanation.

    Even the W.H.O. had a lot of difficulty explaining away what customer satisfaction metrics they bothered to measure among those in the U.S. with access. Also, hey, like every close friend or family person I know I’ve seen the “fudging” of the billing system in both directions – the staff make an honest best effort and I can’t imagine a regulator or legislator exhibiting better or more universally acceptable ethics. Freedom (even if subterranean underneath the *last* bad experiment in billing) kinda sorta works in a way that is hard to compete with. Just expand access, along the lines I suggested at the end of my earlier post – everyone wins in scenarios along those lines.


  • Tim,
    Nice to see that you are paying attention to VistA. I am widely regarded as the top expert on Open Source medical billing http://linuxmednews.com/1113401258/index_html

    I am afraid that if the previous comments are incorrect, it is only because they are not cynical enough about this problem.

    Rather than go into the details of the problem, which I do a little in my interview above, I will just comment on the basic motivators. If I am an insurance company, I make money on the ‘float’. which is the time that I have your money, before you need care, and I have to pay for that care. Generally, an insurance company is just an opportunity to exercise finance. So if they get to keep your money for a month or two later, that adds up to millions of dollars in interest.

    Essentially, doctors are doing electronic data interchange with a partner who has a substantial financial incentive not to accept the data transfer. So if I go to the doctor, and I tell them my name is ‘fred’ but my insurance company has be down as ‘frederick’, well that can delay payment by a month!! But an insurance company can also play this game with the rules of interchange itself. For instance, it could require that the last name be all upper-case, or all lower-case, or only the first letter of the last name should be upper-case and everything else should be lower case. So if you submit McDonald, it would reject. There is no incentive to sort out these kinds of technical issues on the payer side, since anything that delays the payment of a bill makes them money.

    You mentioned the need for evidence. And the best that I can give you is the existence of AthenaHealth, which is a company that exists only to track the variations in the payment processing structures of insurance companies. They make money because this problem is so complex doctors are willing to pay a percentage to deal with it.

    I have been working on Open Source software to solve this problem. I would love to stay in touch b/c eventually I will write a book about this!


  • @Fred – your insights suggest that what’s needed is an incentive alignment map.

    If you take three basic constituencies – doctors, patients, and insurance providers – and map the current system in terms of smooth flow and its impediments – it should be possible to identify clearly (a) self-serving impediments, (b) the parties that are ‘attacked’ by these arrangements, and (c) the extent of the damage that is typically done.

    It’s worth noting that each of these three groups (doctors, patients, insurers) is capable of relating to the others in a manner that can range from harm to help, so this exercise shouldn’t be seen as an attack on, or a criticism of, any one group in particular.

    Rather, it is an assessment of who can do what to whom, followed by a categorization of how functional (or dysfunctional) existing incentives really are. It should conclude with an assertion of what would need to change in order to optimize incentives across the board, so that each of three two-part relationships (i.e. patient to doctor + patient to insurer, doctor to patient + doctor to insurer, etc.) is optimized.

    An optimized map should then point the way to desirable outcomes for a range of groups – lower costs for government, development opportunities for entrepreneurs and industry, increased well-being for families, employers, and communities, development of a healthy job market for individuals with a broad range of skills and abilities, and a general expectation that costs will go down as results improve – just like any other field positioned to benefit from advances in science, data analysis, and engineering.

    In a word, health care – as a field – will become a net generator of value, and not a net destroyer as it is in its current, unsustainable form.

  • Dorothy Hoskins

    What does it mean that IBM was granted a patent on June 2,”Method for electronically maintaining medical information between patients and physicians” (patent 7,542,911) “Disclosed is a method, system, program, and data structure for maintaining electronic patient medical information. An electronic patient data structure is generated to include patient biographical information and one of medical history information, medication schedule information, and appointment schedule information. The patient data structure is electronically transmitted between a physician computer and a portable patient device. The patient data structure is capable of being modified. ” and “8. The method of claim 1, wherein the patient data structure further includes patient insurance billing information that can be used to generate insurance claims for patient services. “

  • nice,thanks for sharing.