Disqus - Latest Comments for Fazal_Raheman_MD

Re: 3. Prize Design Approach

Fazal_Raheman_MD — Fri, 24 Jul 2009 00:55:19 -0000

X Prize is a challenge to innovators to come up with revolutionary approaches to solve the serious healthcare problems. Construing that X Prize Design itself should answer all the difficult questions is misplaced. It is a means to an end, and not an end itself. It challenges the competing teams to design best solutions to address these difficult issues.

The plan design approach may not be perfect, but that’s the purpose of this blog. The purpose is also not only to identify the flaws but suggest solutions to fix them.

With that understanding here is my take on davecluley’s criticism:

1) Cost is definitely a winning factor in estimating the Health Value of a competing healthcare solution. It is for the X Prize competitors to factor cost into their solutions. Consumer engagement is just one of the arms of the diverse elements that X Prize competitors may incorporate in their strategy, and not a component that carries the entire burden.

2) At competition level the entity that optimizes health value the most gets a $10 million reward funded by the competition sponsors. At deployment level the entities that optimize health value become commercially viable enterprises,

3) It sure may be an illusion for some, but then, so was the commercial spaceflight until Ansari X Prize made SpaceShipOne a reality in 2004 with a $10 million X Prize. W.H.O.’s “Health for All” may also be an illusion, but that did not stop us from the pursuit since its first declaration in 1978. Incentivized competitions are for those who have the will to transform illusion into reality.

4) A viable solution to improve overall health need not presuppose that healthcare needs will be clustered in last two weeks. Neither treatment cost has to be reduced by limiting needed care or rationing care. An approach that cannot see beyond those assumptions does not enter the competition. Period.

Re: 4. Initial Prize Design

Fazal_Raheman_MD — Thu, 23 Jul 2009 23:56:31 -0000

Firstly, ask any insurance expert, and he will tell you that the misconception that increases in health insurance premium are solely because "costs are unknown until the fact" is fatally flawed.

Secondly, even actuarial science uses actual cost of a loss claim when it estimates the probability of that claim.

Therefore the increased premium is either because the actual cost of the claim has gone up or the probability of the claim has increased, and definitely not because the cost is unknown.

Scepticism can be reasoned, but negativism cannot. Negativism does not win battles anyway. With that I rest my case.

Re: 4. Initial Prize Design

Fazal_Raheman_MD — Sat, 18 Jul 2009 05:30:57 -0000

Depends on what you mean by a “larger group.” For a small insurer the larger group may be say 20K, but for most insurers that are likely to partner in the X Prize initiative the larger group is likely to be in hundreds of thousands, a size that reaches the actuarial costing plateau that does not get affected by pulling out of a cohort of 10K. Even if there is any excess cost burden it is an attribute of an inefficient system rather than the liability of an efficient solution under trial.

Re: 4. Initial Prize Design

Fazal_Raheman_MD — Sat, 18 Jul 2009 05:29:39 -0000

Why do you assume the price of the service will not be known at the time the health insurance underwrites the policy? When a product is on the market the price is definitely known, but when the product is not yet on the market, the price can only be estimated on the basis of manufacturing cost, volumes to be produced and marketing overheads. In my previous post I already provided estimates of what each of the healthcare component will cost in our X Prize solution. The cost can only be estimates today as the service is yet not on the market. Sure they will be firm costs when the 3-year trial validates them. X Prize is about conquering unchartered territory, how do you expect to know all about the territory unless you research into it?

Re: 4. Initial Prize Design

Fazal_Raheman_MD — Fri, 17 Jul 2009 02:19:04 -0000

Sure transparency is crucial to bring down not only the cost but up the quality. The buyer may not know the actual price before the healthcare is provided, but the insurer sure has an estimate of the overall cost and that's enough for pricing the health insurance premium. And, all that the buyer got to know is the monthly premium.

Re: 4. Initial Prize Design

Fazal_Raheman_MD — Fri, 17 Jul 2009 02:13:47 -0000

Here’s how we have addressed the four challenges that davecluley cites in our X Prize solution:
1) An insurer/payer always knows an estimated breakdown of the cost distribution of care categories per premium dollar. For example based on published data following is the breakdown, and in parenthesis is the corresponding reduction that our X Prize solution Rx0 targets:
Inpatient care = 25 cents (50-75% reduction) Rx0= 6.25-12.50 cents
Outpatient & Misc Clinical Support = 25 cents (20%) Rx0=20 cents
Prescriptions = 16 cents (60% reduction) Rx0=6.4
Administrative costs = 34 cents (60% reduction) Rx0=13.6
Total Rx0 costs 46.25 – 52.50 cents
So we can arrive at the X Prize target of 50% benefit and also a fair estimate of insurance premium after accounting for the loss ratio.
2) Since the study recruits subjects after a comprehensive screening, the health status of the cohort will be a lot better known than a usual situation that a health insurance company encounters.
3) I don’t think in our model the study cohort has any impact on the larger group, so no question of premium penalty.
4) The reduction in premium is not because of subsidizing, but on account innovations deployed in underwriting, implementation and administration of the policy itself. Again no question of premium penalty to the other group.

Re: 7. Philosophy of Prize Design Guidelines

Fazal_Raheman_MD — Sat, 30 May 2009 03:03:26 -0000

I really like your flexibility. It’s a great commitment to innovation. “Vendor to team in multiple communities” and “Demonstration community,” it’s like no bars to creativity. Cheers.

Re: 5. Impact Potential

Fazal_Raheman_MD — Sat, 30 May 2009 02:52:47 -0000

Thanks Dr Goel. I guess we can count on Healthcare X Prize sponsors not letting the vested interests who continue to profit from maintaining a status quo raise hurdles in the process of radicalizing healthcare that is on the roll now.

Re: 2. The Grand Challenge

Fazal_Raheman_MD — Sat, 30 May 2009 02:43:30 -0000

Firstly, Medicare patients are senior citizens obviously the sickest of the entire population. Rehospitalization is a function of their sickness rather than “less administration”. Apples cannot be compared with oranges.

Secondly, administration cost is not the “management cost” to reduce morbidity. It is purely non-care expense incurred on account paperwork for processing patients’ claims for the service.

Thirdly, it is already unequivocally established that minimizing that wasteful administrative expense improves the impact of healthcare dollar. As I have already stated elsewhere in this blog (http://healthblog.xprize.or... the US healthcare except a few areas does not look that bad if we don’t look at the cost of delivering it and the proportion of uninsured. It is the exorbitant cost that makes it look awful in comparison to other developed countries. Minimize the cost and healthcare as it is looks much better. If competitors don’t emphasize the cost element the expected 50% benefits are impossible to achieve.

Re: 4. Initial Prize Design

Fazal_Raheman_MD — Fri, 29 May 2009 16:09:16 -0000

I agree with Dr Goel, the dollars need to be real. But the monthly premium paid to the health insurance company is as real as reality gets. More so, because cost is the major culprit for the troubles the US healthcare system is in. Now, how do we deal with the test population’s health insurance premiums? Do we tell them that their healthcare is covered free of cost because they are subjects of a clinical trial? In the traditional clinical trial for a defined disease variable, providing the test intervention free may be fine, but not in this case, because cost itself is one of the variables under study. So if the test population is paying health insurance premium just like the control population how do we arrive at what the premium will be without actuarial estimates? Please understand that although actuaries provide estimates to begin with, they are real dollars at the end of the day. This is how it is. In the beginning of the fiscal year when the beneficiary signs up, the policy premiums are only actuarial estimates, but at the end of the year they are real dollars in terms of “loss ratio”, “administrative costs” and “profit” or “loss”.

In insurance a loss ratio is the ratio of total losses paid out in claims plus adjustment expenses divided by the total earned premiums. For example the average loss ratios amongst the US health insurance companies range from 60% to 110%. See James C. Robinson, "Use And Abuse Of The Medical Loss Ratio To Measure Health Plan Performance", Health Affairs, vol 16, No. 4, pp 176 - 187, 1997. This means if an insurance company pays $60 in claims for every $100 in collected premiums, its loss ratio is 60%. It also means that for the healthcare services that the population received the cost was $60 and the remaining $40 minus the administrative costs is the profit of the insurance company.

In any meaningful clinical trial the trialist should have a fair projection of the efficacy of the intervention under trial, only then he can design the elements of a robust study. For every “real outcome” to be proven in clinical trial, it has to be first the “estimated outcome”. If the estimates are positive that is a “GO” decision, if the projection aren’t positive it’s a “NO GO” decision. That’s how we develop new treatments in the healthcare industry. No clinical trial is planned without projecting its efficacy. I don’t think it should be any different with those who wish to compete for the X Prize. Every X Prize competitor has to come into the competition with a projection of what fiscal and morbidity implications of his solution will be in terms of percent improvement over the control. If those projections are close to desired threshold (50% in this case) the trial is worth it otherwise there is no point in conducting the study. At least we followed the same approach. For instance, it took us more than a month since the first announcement by X Prize Foundation on April 14th to figure out if the solutions that we developed over the years can theoretically achieve the targeted benefit of 50%. And when our estimates did give us numbers supporting a “GO” decision I joined this blog.

One more thing that I noticed missing from this blog’s discussion is that at the end of the day this competition is a major one-of-a-kind clinical trial the like of which the clinical researchers have never seen. Trust me designing the X Prize clinical trial to test the winning solution is more challenging than the solution itself, and I already made some suggestions in another space on this website in that regard. Thank you very much for your patience.

Re: 5. Impact Potential

Fazal_Raheman_MD — Fri, 29 May 2009 00:08:18 -0000

"In healthcare delivery, science of the moment misses the overall picture." You said it all DBruceD1. A breakthrough first-in-class anticancer drug gets approved by the FDA for deploying fancy science and in the process giving the terminal cancer patient a “statistically significant” couple of months of extra life of pain. That’s science of the moment costing the patient an extra $10,000-$20,000, and in the process making the big picture look bleaker. The super specialists go one popping in probes, stents and grafts into sick or not-so-sick hearts without doing a thing to change their long term morbidity or mortality risks. (Weintraub WS, Spertus JA, Kolm P, et al. Effect of PCI on quality of life in patients with stable coronary disease. N Engl J Med 2008; 359:677-687.).

Today’s hierarchy indeed can never reform the system. A radical change needs an initiative like the X Prize incentivized competition. Let the competition be conducted in an environment that is not influenced by the interests that benefit from the existing paradigms. Amen

Re: 2. The Grand Challenge

Fazal_Raheman_MD — Thu, 28 May 2009 13:28:06 -0000

Actually administrative costs may drain as much as one third of the total healthcare expenditure. A research conducted by Kahn et al concluded that, of the total insurance premiums used to cover hospital and physician care, 21 percent is spent on insurance administration. Another 13 percent is used to cover other administrative tasks. Only 66 percent is used for actual patient care. (Kahn James et al. The Cost Of Health Insurance Administration In California: Estimates For Insurers, Physicians, And Hospitals. Health Affairs, 24: no.6; 1629-1639, 2005.). It is indeed the single biggest category of wasteful expense.

Re: 9. Contact / Comment

Fazal_Raheman_MD — Wed, 27 May 2009 05:16:13 -0000

Scalability in the context of the current competition is broadly of two types, a) scalability of the technology infrastructure, and b) scalability of the population dynamics and outcome. Scaling up of the technology infrastructure is ensured using the established Lean Design and Six Sigma protocols during the technology development process. However, scaling up of the population base and still keeping the outcome close to those obtained during the trials depends on the outcome measures chosen for the trial. A very carefully designed clinical trial protocol is therefore crucial in translating the results of a trial population to the entire nation. The X Prize team needs to focus on that aspect.

Re: 8. Teams

Fazal_Raheman_MD — Wed, 27 May 2009 04:40:39 -0000

Have you developed the HXP LOI Program? Any timeline on when the HXP LOI Program will be flagged off? How much is the refundable LOI fee?

Re: 3. Prize Design Approach

Fazal_Raheman_MD — Wed, 27 May 2009 02:33:50 -0000

David,
Contrary to what you think health outcomes are always a consequence of the healthcare delivered, how can they be tangential? The “Health Value” comes from the health status achieved at a cost. Cost is always integral part of value of a service, and can never be separated in a competitive world. What good is an excellent service if it comes at a prohibitory cost? Please read my recent Posts in the “Measurement” section of this blog to see my point as to why in the first place we see US healthcare flawed. The bottom-line is it does not provide the same value per dollar spent as the other healthcare systems provide. See http://assets.opencrs.com/r.... Overall, if adjusted only for those Americans who have access to healthcare, in most areas, the US healthcare stats will rank above average or even lead amongst all the advanced countries. But if compared in terms of the price paid and morbidities contributed by uninsured, it fairs miserably. That’s the kind of problem the X Prize competitors will be facing.

I think your observation is influenced by the fact that the X Prize is still isn’t actually framed as you assume, but it is in the process of being framed. It is a preliminary announcement to challenge the innovators to pursue the so called “unassailable pursuit”. How the journey to the ultimate destination will be, is still clouded, and that is perfectly understandable given the complexity of the challenge. It is up to us to help define it, and that’s the reason this forum is open to public for discussion. Those who really want to take up the challenge in a positive spirit of winning the impossible race will eventually prove many a predictions wrong. So cheer up the X Prize Team.

Re: 6. Measurement

Fazal_Raheman_MD — Wed, 27 May 2009 02:31:10 -0000

In my previous post on this blog I shared my initial thoughts on endpoints that measure the health value outcomes of the intervention and hoped to continue the discussion. On second thought I believe that the most difficult part is structuring a well designed trial that can be deployed to show the beneficial effects within the constraint of sample size and time. I don’t see the expected shape to the Prize design coming as we come to an end of the 45 day period of public comments. It requires more concentrated efforts than a few healthcare professionals commenting and airing their opinions on this blog on micro issues without the bird’s eye view of what’s going wrong as compared to other healthcare systems of the world.

Firstly, how do we know that the US Healthcare system needs reform? Just imagine if the healthcare system of other advanced countries also spent 17% of their GDP or averaged $12,700 per family per annum in insurance premiums, and had 15% population with access difficulties (uninsured), would the US healthcare still look that bad? Not at all. Or just reduce the US Healthcare cost to half and adjust the population denominator to only those who are insured and have access to healthcare, would the US healthcare appear on par if not better than the best in the world? Sure it will.

Secondly, will you find any Alzheimer’s, neural tube defect, teen pregnancies, dialysis or even mortality in a typical 10,000 US community? May be a few at best, but that does not give us any analytical power.

Thirdly, what are the drivers of upping the cost and confounding the quality that are found in statistically significant numbers and can be easily measured in a 10,000 population? The answer again comes from comparing US vital statistics with the health stats from the advanced countries. For instance if you look at the hospitalization trends US already has one of the least number of hospital days per 100,000 population, but intensity of such hospitalization is the highest. Can we get any lower if we merely measure the change in hospitalization?

Finally, all these issues need to be taken into account upfront in designing the X Prize trials. Trust me the perfection of the study design is crucial and more intricate to get any meaningful outcome at the end of 3 years. Just selecting a population and measuring morbidities most obvious to the naked eye is not enough. I believe there are three ways this can be done:
a) X Prize team hires an experienced in house clinical trial specialist / epidemiologist to come up with the draft clinical trial protocol, which may not come cheap;
b) X Prize team retains a third party specialist CRO for designing the clinical trial protocol, which may be still more expensive;
c) X Prize announces and publicizes a global competition for a robust trial design that would demonstrate about 50% improvement in health value parameters in 3 year follow up of a typical 10,000 US community.

The last option can come out the cheapest and most productive. The winning study design can be awarded a nominal amount of $100K plus a guaranteed entry amongst the five finalists. The cash involved may not motivate the best, but guaranteed entry amongst the 5 finalist will definitely motivate the best in business. X Prize can very easily find a sponsor for the cash prize and the campaign will also create tremendous media presence for the initiative without much in-house advertizing expense. The Healthcare X Prize initiative will accomplish a lot in 90 days this way than any other.

Re: 2. The Grand Challenge

Fazal_Raheman_MD — Wed, 27 May 2009 01:27:41 -0000

Thanks for bring up the subject. Establishing Best Practices are essential not only for the practice of innovation, but for creating the innovation. No clinical trial data will bear any strength if not generated in fully compliance with GCP, GLP and GMP requirements. I did not see any mention of the issue either from the X Prize team members or anyone from the audience. Perhaps it’s still too early as the clinical trial protocol for the 5 finalist teams is not even on the drawing board yet. Nevertheless, its important though to bring the subject to the attention of the X Prize team.

Re: 4. Initial Prize Design

Fazal_Raheman_MD — Tue, 26 May 2009 16:11:16 -0000

Here is my take on tbe direct cost question. Each team's cost cutting and outcome improving approach will be clearly defined upfront before initiating the trial. Since each team will partner a health insurance provider whose actuaries can actually estimate the risk and the insurance premium upfront based on the team's outcome projections. Insurance companies routinely use the acturial algorithms to estimate premiums. For example the current national average for annual health insurance premium for an employee's family policy is $12,700, and coronary revascularization is 59 per 10,000 population. If the team projects reducing coronary revascularization to say 40 per 10,000 and arrives at an acturial estimate of $8,000 and copayment of $1,000, that's more than 50% benefit.

Re: 7. Philosophy of Prize Design Guidelines

Fazal_Raheman_MD — Tue, 26 May 2009 15:51:39 -0000

Of course winning company will retain the IP

Re: Health Value Messaging - The Single Sentence Statement

Fazal_Raheman_MD — Sat, 23 May 2009 14:14:10 -0000

Thanks for your feedback on my comments. The parameters you suggest are indeed measurable and perhaps quite media friendly too. But, I am not sure if any measurement can become tangible unless the dollar denominator isn’t associated with it, especially in the present scenario when cost is the major problem with the US healthcare. Just eliminate the cost element and you will find that the US healthcare compares fairly well with any advanced country. Cost is an essential element of any equation that measures the health outcome. I believe it should be the essential common denominator in any draft statement just as “less than $10,000” defines the challenge that X Prize Genome project faces.

I think everyone agrees across the globe that the standard of healthcare in the US is undoubtedly the best. The problems however are with the process of getting it delivered to the beneficiaries at a cost that is disproportionally higher than the outcomes. While the higher costs are attributed to the inefficiencies and wasteful practices of the current system, the over all lowering of the standard of outcome is significantly affected by the sizeable population of uninsured. The hidden cause of less than ideal outcome (such as higher infant mortality rate, life expectancy at birth etc) is obviously attributed to more than 15% population that is uninsured, and shall remain so unless a radical change alleviating the woes of the uninsured is brought in. The big question is can any care model that claims to bring a radical change ignore such a significant chunk of population that in the first place is the cause of the compromised quality of healthcare in comparison to other developed countries. Well, that’s the issue for another post, perhaps in the Measurement blog.

Clinical endpoints have to comply with two minimum essentials, firstly they should be found in significant numbers in the trial population, and secondly they should be most vulnerable to change in response to intervention. According to a recent Congressional Research Service Report the US fairs quite efficient amongst the OECD countries in terms of hospital utilization and average length of hospital stay. See http://assets.opencrs.com/r... (page 11 and 14). For a clinical endpoint which is already at the lower end of the expected range, achieving a further decrease with any type of intervention is statistically very difficult, certainly a tall order if the expected benefit is targeted at 50%. Finally, if the Healthcare X Prize statement articulates the indicators of measurement of health value, articulating the pitch before the clinical endpoints are finalized will be jumping the gun. What do you say?

Re: Health Value Messaging - The Single Sentence Statement

Fazal_Raheman_MD — Sat, 23 May 2009 00:58:37 -0000

Scott,
Healthcare X Prize is indeed a different animal. As much as one may think pitching “health value” in a single concise statement may be no less rocket science then rocketing three guys 100 kms into space twice in two weeks, it’s turning out to be mightier than the havens already conquered. Sure, it is easier to articulate measurable target parameters in a scientific exploration even if it’s near impossible to achieve the goal deploying the current state-of-the-art means. No wonder there are no comments/feedbacks on this Messaging blog. Sounds like an invincible territory that everyone is avoiding. Perhaps, that’s why an X Prize is justified.

The ardent researcher in me, trekking into this apparently unassailable footslog, tells me that research is going to the unknown based on what is known. What is still unknown in this quest for an elevator pitch for Healthcare X Prize are "whats" and "hows" of the parameters to measure the winning improvements in healthcare. So first things first, let’s get a working hypothesis for the unknown that we need to define, and then prove, amend or improvise that hypothesis based on resolving the health value measurement dilemma. While my insights on the “health value” measurements are posted on the “Measurement” blog (http://healthblog.xprize.or..., here is the initial draft suggestion for the working hypothesis statement.

“Wellpoint Healthcare X Prize: Comprehensive scalable healthcare interventions that establish net benefit of at least 50% in improved outcomes and lower costs in a population of 10,000 over 3 years.”

Let's keep the ball rolling.

Fazal Raheman, MD

Re: 6. Measurement

Fazal_Raheman_MD — Wed, 20 May 2009 23:19:56 -0000

Gentlemen,
A review of this blog tells me that Dr Randolph Seed first commented that the 10,000 sample size is “too small”, to which Dr Goel invited “suggestions for things that we should change”. Fred Davis posted comments that highlighted more problems downstream and suggested he had “several small and inexpensive changes that will affect significantly every one of those 4 main measures with zero ill effects.” But I didn’t see any comments on problems relevant to designing a robust study design that could allow deployment of epidemiologic tools to elicit desired X Prize deliverable i.e. 50% improvement in health value. Over two decades of intense involvement in clinical research and product development in healthcare industry I have learnt to accommodate robust research methodology to the demands of time, logistics, cost and administrative constraints. And, by those constraints I believe a target population of 10,000 and a 3 year follow up are reasonable limitations that the research methodology has to be worked around. The big question then is what that “target population” should be, and what the measurement metrics to assess the intervention are. But before that we got to understand the problem with a little better clarity than defined by the preceding posts.

In large national or regional surveys (target population usually in millions or at least 100s of thousands) up to two dozen indicators of community health status may be usually used. Most traditional indicators of community health, such as infant mortality, life expectancy at birth, all cause mortality, incidence of new cardiovascular events, incidence new-onset diabetes, so on and so forth, will not show any significant change in the envisaged 30,000 person years (10,000 X 3 years) of follow up of a typical US community. Other indicators of community health status such as percent uninsured, percent inadequately insured, percent unemployed, care accessibility, etc. cannot be evaluated in a study design that presumes assured access to healthcare, as is the case in X Prize competition. In fact, all of the traditional indicators of community health, except surrogate markers such as BMI, hypertension and hyperlipidemia, will hardly reach any level of statistical significance given the sample size and duration of the proposed study and of course the desired 50% improvement in the target population’s health status.

As an example let’s look at a major and most common health problem – Diabetes. Amongst the general population on an average 1785 person-years of follow up is required to find 1 case of new-onset diabetes. (See Martínez-González M, de la Fuente-Arrillaga, Nunez-Cordoba JM, et al. Adherence to Mediterranean diet and risk of developing diabetes: prospective cohort study. BMJ. 2008;DOI:10.1136/bmj.39561.501007). It means with 10,000 population size and 3 years (30,000 person-years) of follow up, one would at best expect just 1-2 cases of new-onset diabetes, which would hardly ever reach statistical significance compared to control population.
Therefore, if the study sample size and duration is limited one has to look for inclusion criteria, endpoints and indices that provide statistically significant outcome, especially if the target is to see 50% improvement. We also have to look for outcome measures that quantitatively measure the cost efficiency of each competitor’s interventional approach. Towards that end, I find the proposed measures of “Health Value” or “CHI” under consideration by the X Prize Team vague and do not clearly define clinical endpoints:
a) that are reasonably objective and reproducibly measurable,
b) that are found in statistically significant numbers given the study size of 10,000 followed up for 3 year period, and,
c) those endpoints that focus on health conditions that consume the most healthcare cost. For example just three chronic disease categories cost US economy about $900 Billion, which is almost 75% of the total health insurance revenues. With the 30,000 person years of intervention, it is impossible to get any meaningful data on each and every aspect of healthcare, many of which may not have much impact on health value or cost and aren’t worth defining the measurement metrics based on them.

Let’s look at two of the CHI measurement metrics under consideration by the X Prize team:

Optimization of individual vitality: Functional fitness, self-reported health status

Self-reported health status is a subjective soft endpoint and will only carry any deductive value if the trials of competing X-Prize interventions are double blind RCTs, which I very much doubt are possible in the given circumstances. Since the trials cannot be blinded only hard endpoints or surrogate endpoints that are more or less predictive of the hard endpoints, and are highly prevalent in target population will carry any analytical value. How these hard endpoints or their surrogates can be identified and maximized is the key to designing the trials.

Elimination of acute exacerbation: Hospitalization/ rehospitalization; ER visits; Communicable

Rosenberg et al (N Engl J Med. 1995 Nov 16;333(20):1326-30) studied a New York cohort of 7445 fee-for-service insurance plan enrollees who called for inpatient services. Only 10.55% of all hospital admissions were on account of medical emergencies. See Rosenberg SN et al. Effect of Utilization Review in a Fee-for-Service Health Insurance Plan. N Engl J Med. 1995 Nov 16;333(20):1326-30). (http://content.nejm.org/cgi..., accessed May 20, 2009). However, when accounting for the total hospital days approved for inpatient insurance coverage, only 286 days out of the total 20,792 hospital days (about 1.4%) were claimed for emergency admissions. In other words 98.4% of the approved hospital days were for elective procedures and emergencies were only 1.6%. In a general population cohort of 10,000, we may not find the incidence of acute hospitalization beyond single digit. That low occurrence of a clinical event endpoint can never reach statistical significance when compared with any comparable control. Therefore if the clinical endpoints pertain to cost savings or quality improvement in care or elimination of “acute exacerbation” or emergency hospitalization, the trial will hardly provide any meaningful data.

In my next post I will share my thoughts on the remaining two measures and appropriate strategy in resolving these issues in a way that best deploys the clinical epidemiology tools to effectively achieve the X Prize goal of picking a winner who can prove his/her approach produces 50% benefit within the constraints of 30,000 person years of intervention. Meanwhile, comments and input from the members of the X Prize Team and other experts will help placing the issues in proper perspective for the readers of this blog and for helping evolve a robust study design for a successful X Prize contest.
Fazal Raheman, MD