Disease Management Has Moderated U.S. Health Care Cost Inflation

Thanks to years of unapologetically quoting, citing, linking and blogging in over 1300 posts that disease management saves money, the Disease Management Care Blog has earned approbation, fear, respect and disdain nationwide.

But its audacity is mere child's play compared to this Wall Street Journal Opinion by J.D. Kleinke.

His claim? That "disease management" was one factor in the slowing of national health care costs.

Mr. Kleinke, as you can see here, is no lightweight. He not only has experience as a real-world health care entrepreneur and executive, he's the author of the groundbreaking book, The Bleeding Edge.  When he published it ten years ago, Mr. Kleinke predicted the rise of "Emerging Healthcare Organizations" (EHOs) that would harness the forces of risk-assumption, consumerism, consolidation, integration, and industrialization and transform the health care system. If you believe ACOs will succeed, you may want to thank JD for thinking of them first - even if he got two of the initials wrong.

But is he wrong about disease management?

Mr. Kleinke uses a copyrighted bar graph in the article but the image below (lifted by the DMCB from the White House's web site) shows the same data. After an uptick in 2000, the nation's annual percentage change in total health care spending has progressively declined and is the lowest it's been since the 1980s.

Mr. Kleinke asserts that when HMOs were defanged after their '90's decade of bad behavior, insurers not only rolled out deductibles, co-payments, health savings accounts, tiered drug plans and urgent care center coverage, but they were subjected to publicly available measures of how well their networks performed and started "the still emerging science of disease management."

He also makes the point that it's impossible to know the relative contribution of each of the initiatives described above. Assuming that they collectively had some impact, Mr. Kleinke calls for a doubling down and change the regulations and tax code that still is preventing insurers from building on their success and finding ways to innovate.

The DMCB agrees that he makes a good point. But it also brings up several caveats:

Depending on how you compare Medicare fee-for-service (FFS) (where there is no disease management) and commercial insurance costs (where there is), the rate of increase Medicare may be even lower.  It's comparing apples and oranges, but it's still an important point that deserves further analysis.  There's more discussion on why the comparison is not so simple here.

The remarkable drop in health care costs over the last two years may be result of a bad economy and not the advances described by Mr. Kleinke.  We'll find out more when the economy picks up steam.

Last but not least, a smaller rate increase on top of an ever expanding fraction of the economy is still an big increase.  We're now committing a whopping 17% of gross domestic product to health care.

Of Asthma, Inhalers and Patient Education (Humor)

Yes, it's a loathsome and unrealistic TV show, but the Disease Management Care Blog cannot resist this example of the merits of patient education and how not to engage patients in self-care, hm?

How "Texting" Would Work in a Real World Disease Management Program: How Many Nurses Do You Need To Hire? Can Physicians Deal With the Data Glut?

"My glucose is 149"

As population health providers such as care management vendors, home health agencies, medical homes, accountable care organizations and pharmacy benefit managers strive to increase both the quantity and quality of interactions with their patients, one thing is certain: traditional snail mail and phone calls are no longer equal to the task.

E-mail is an option (even as the lawyers worry about HIPAA and the CIO insists that it happens in a "walled garden") and then there's the emerging potential of social media (precontemplation in 140 characters or less?).

But how about cell phone texting?

The Disease Management Care Blog thought it would have been more common in population health settings, but it's not too sure. A quick literature search revealed some research on adolescent asthma care, diabetes, and lupus.  There's even this glowing review of the topic that describes multiple prospective studies in multiple countries that show that texting can have a positive impact on behavior change.

But before you run out and establish a texting option for your patients, you may want to read this hot-off-the-presses American Journal of Managed Care article first. Henry Fischer and colleagues from Denver Health recruited 47 persons with diabetes who were receiving their primary care in a Denver federally qualified community health center. The study subjects went through a short (3 month) pre-post study that piloted the impact of three-times-week text inquiries about blood glucoses plus reminders about upcoming appointments. If patients mistyped their answers, they were called by a nurse.

There are no data on the impact on blood glucose control, but that's OK. The DMCB learned some other useful stuff:

1. The Digital Divide: Cell phones are often the sole means of communication for patients who cannot otherwise afford a landline, internet access or a computer. If you're serving a socioeconomically disadvantaged population, texting may not be just the best option (because you need to get past caller ID), it's the only option

2. Ageism? Don't believe that this can only be used in adolescents. Nine of the persons in this pilot were between the ages of 60 and 69.

3. The Silent Treatment: The DMCB occasionally texts its spawn and, like many of the parents reading this post, gets ignored. But now you know that you are not alone because about a third (32% of the 1585 messages in this trial to be exact) will be ignored by patients.

4. Fat fingers? Nope. 99% of the texts were formatted correctly.

5. Oh, those pesky full time equivalents: Of course the only reason any health system CFO will OK the budget for a texting service is so you don't have to hire anybody and hopefully even "downsize." The bad news is that someone needs to monitor things: this study describes a "0.2" FTE nurse. The DMCB guesstimates thats one nurse for every 250 persons in a high volume (three times a week) texting-only care management program. Cutting the volume of messages to once a week may mean one nurse for every 750 participants. You get the picture.

6. The physicians?  You guessed it, the busy physicians were unable to get to the text message result a whopping 88% of the time.  A classic example of electronic data glut.  Better to let non-physicians manage this, says the DMCB, even if the CFO doesn't like it.

The DMCB Bottom Line: As far as the research goes, we already knew that texting can lead to behavior change.  This AJMC article was more of a real world effectiveness study that yielded some important insights on the target population, response rates, staff support ratios and the risks of burdening physicians with information overload.

Image from Wikipedia

White House: "Covering contraception is cost neutral since it saves money" - The Birth Control Controversy and Using Today's Costs for Tomorrow's Savings

"Cost neutral and saves money?"

"What the hell does that mean?" asks the Disease Management Care Blog.

How can a White House Fact Sheet simultaneously assert that an insurance mandate simultaneously costs nothing and also reduces health care costs? Is this cognitive dissonance? Clever obfuscation? Insurance amateurism?

The physician DMCB fully understands the importance of contraception in women's health.

The logical DMCB knows someone has to eventually pay for birth control pills and doesn't understand the material difference between a mandate for buying contraceptive coverage versus buying mandated insurer coverage of contraceptives.

The accommodating DMCB wonders if religious leaders could compromise by agreeing to cover birth control pills with a "100% co-pay."

But the blogging DMCB couldn't resist looking at the of real world difficulty of reconciling today's insurer hard costs and tomorrow's savings.  Over the years, it has learned it's never that easy. 

Read on and see why that rule holds here.

So, just what is the evidence on birth control pills (BCPs)?

While BCPs' $15 to $50 a month retail cost contrasts with the average $13,000 to $17,000 total charge for childbirth, it's not that simple. Add up the additional cost of thousands of BCP prescriptions, their associated provider visits plus unnecessary testing, a 15% failure rate and a low but appreciable rate of side effects, its very possible that the aggregate costs of BCPs could exceed any savings from a fewer number of unwanted pregnancies.

The WH Fact Sheet addresses this by saying that prior studies show.....

"...there was no increase in premiums when contraception was added to the Federal Employees Health Benefit System and required of non-religious employers in Hawaii. One study found that covering contraception saved employees $97 per year, per employee."

As far as "Hawaii" goes, the DMCB discovered that White House is quoting "empirical evidence" from a 2001 study from the state's Insurance Commissioner. It describes the experience of four commercial health plans (anonymously named A, B, C and D) that went through a state-imposed contraceptive mandate.  "A" already covered it, "B" saw a 14 cents per member cost of "contraceptive services and supplies" and "no material changes in eligible charges for maternity related services," "C" had a $1.50 increase in "contraceptive costs" and an approximate $7 decrease in maternity costs per 1000 members (which the Commissioner summarizes as of "little effect"), and "D" which had an 11.5% increase in pregnant members as the contraceptive mandate was implemented.

And where does the Fact Sheet's "$97" savings figure come from? The DMCB suspects the number comes from Table 4 of a 2000 CDC and Washington Business Group on Health Report here which quotes scenario economic modeling of $562 vs. $659 per employee per year with and without a contraceptive benefit, respectively.

While it was thinking about the topic of economic modeling, the DMCB found this interesting 1996 study of the cost-effectiveness of family planning services for the British National Health Service. Based on a cost of £39 and "0.82" avoided pregnancies per year, the authors estimate that the NHS saved £754 "per avoided pregnancy." The DMCB suspects the White House would probably prefer to not quote this one.

The DMCB's conclusion: there is no real world evidence that oral contraceptives save any money by reducing claims expense. Any evidence that there are savings are based on methodologically suspect economic modeling and actuarial studies. While the good news is that BCPs probably won't increase insurance costs, the bad news is that it does little to tame health care inflation. That's probably why this small recent survey of U.S. health insurers indicates that they don't believe it will save money either.

Two other DMCB  points:

Given the experience in Europe, even if the mandate for contraceptive coverage wins this round, it will remain vulnerable to political and economic meddling... probably forever.  Women's health deserves better.

BCPs are easy to prescribe.  Given the likelihood of not enough PCPs to meet demand,it can be argued a physician office visit is unnecessary - which would further lower costs. It has also been argued that BCPs should be made available without a prescription, which begs the question of why this should be an "insurance benefit" in the first place.

Image from Wikipedia

The Latest Health Wonk Review Is Up

The latest Health Wonk Review is up over at the "More Than Birth Control Pills" edition hosted by Jason of the Healthcare Economist Blog.  There's a lot of good stuff here, including observation hours, value-based purchasing, the coming Supreme Court battle, the Republican candidates, stealth marketing and wayward HIPAA triggering laptops.

Enjoy!

Web Crawling an EHR for Text: Natural Language Processing

"Crawling" an EHR for text?

Readers may recall a prior Disease Management Care Blog post that describes the potential of "natural language processing" (NLP) in population health and disease management.  Since much of the electronic health record (EHR) still relies on free text, it's difficult to extract and pool useful data from the system. 

For example, if the DMCB types that the patient "has a 10 year history of diabetes" in a clinical encounter note without otherwise officially documenting that condition in a "diagnosis field" or "billing code," the presence of diabetes won't be electronically recognizable.  The good news is that NLP is getting to the point where, much like a classic "web crawler," it can be used to "scan" free text and look for phrasing that identifies otherwise silent and undocumented diagnoses or treatments.

The good news is that there's this article courtesy of Fierce Healthcare that, in addition to saying nice things about the DMCB, gives a lot more background on the art and science of NLP or "computer assisted coding."  It turns out that its been under development for more than ten years, can generate physician reminders and by inferring a diagnosis, automate the tedious exercise of coding a patient visit.

The DMCB likes the notion and would make two other points:

1.  It doubts current EHR vendors will develop an NLP capability because they can't commercialize it and it's not in the meaningful use criteria.  As a result, this functionality will need to be "bolted on" by a third party.  The companies that can develop a single solution that not only fits multiple EHRs but can adapt to a health information exchange will win.

2.  Speaking of meaningful use criteria, it's time to include NLM.  Its ability to find patients at risk, facilitate population-based program planning, assist in predictive modeling, infer and automate diagnosis coding, generate reminders and act as an early warning system for outbreaks of disease make its  potential too great to ignore.

Image from Wikipedia

ePatients: The Disruptive Innovation for Shared Decision Making?

As readers may recall, the Disease Management Care Blog is a big fan of shared decision making (SDM). Using unbiased, state-of-the-art and interactive media that presents a set of treatment options, patients with the help of their doctors are surprisingly able to input their values, make trade-offs and come up with the best choice. Research suggests that patients tend, in aggregate, to be both reasonable and conservative.

The DMCB is also a fan of the ePatient movement.  Based social media, this is the networked and two-way sharing of medical information in a virtual community of like-minded patients that also facilitates informed patient decision-making. Compared to SDM, ePatients have been less well studied, but there is some good research that suggests that these on-line communities are remarkably disciplined and accurate. In fact, advocates argue that ePatient communities, compared to physicians, are better able to alert its members about the latest medical updates.

Which makes the DMCB wonder if the ePatient movement represents a classic disruptive innovation that is threatening the SDM business model.  Commercial SDM typically is made up of video content (much like a DVD) that is developed by credentialed experts and has to be updated periodically. Contrast that approach with the ePatient community (on-line, on demand whenever you want it) that harnesses the wisdom of crowds and is so organic, it really never stops being updated.

Think music CDs vs. iTunes. Or DVD movies vs. on-line streaming.

But the most important distinction? The former is being sold by physicians or content developers, while the latter is being given away.

The DMCB speculates:

The shortage of primary care services, increased cost sharing, spreading consumerism, the explosion of medical information, distrust of authority, ease of use for social media, less concern about privacy, the rise of self-educated experts and a curious immunity from actually having to make money all make the DMCB think that the ePatient movement is here to stay.  It's value will trump the inevitable anecdotes of waylaid web users, e-mistakes and death by internet.

In order to stay relevant, commercial shared decision support tools will begin to direct users to medical e-communities. Physicians will "outsource" their own SDM to online ePatients. This will give IBM's Watson a run for its money. Architects of clinical guidelines, point of care decision support, EHR portals and health information exchanges will successfully ignore the ePatient movement - for now.

On-line bulletin boards will be the first stop for any new symptom and to second guess physician advice. The doctor-patient relationship will turn into a menage a trois.

"eApps" will appear for the persons with diabetes, mothers of leukemics and the children of persons with Alzheimer's.

Health insurer preferred provider networks will have one more reason to die off.

e-Communities will add value in helping its ePatients figure out how to get insurers to cover services that would otherwise be denied.

Some mainstream health organizations will host ePatient communities.  Authentic and trustworthy ones will learn from them. None will successfully commercialize them.

Researchers will tap into e-communities to recruit patients for research. The pooling of users' data for observational research studies will grow.

Unfortunately, this will exacerbate the medical digital divide for persons without reliable access to the internet.

Preauthorization vs. Retroactive Audits for Patient Safety

"If you're telling the truth, hit 1; if not,
hit 2.  If you're not sure, please wait on
the line....."

If the Disease Management Care Blog accepted the common wisdom about commercial health insurance, it'd believe that the businessis is an evil empire controlled by pirates who 1) use weenie print to fool unwitting patients and 2) rely on opaque decision-making to deny coverage for life-saving treatments.

Once again, the DMCB eschews common wisdom:

The reason why insurers have used small fonts in their coverage documents has been because their state regulators likewise use small font in their rule making and require full word-for-word disclosure to the beneficiaries. Think of lawyers run amok and you'll understand the main reason for the arcane language.

The good news is that new regulations promise easy-to-understand "labeling." The DMCB thinks that's a step in the right direction, but that doesn't mean that a lack of preciseness won't lead to misinterpretation and wrong assumptions. Time will tell.

As for the opaque decision-making, check out (for example) Aetna's and Humana's open-access coverage documents. Yet, despite a high level of on-line transparency, critics may point out that commercial insurers will still require a hassle-laden preauthorization process.

That may be true, but the DMCB says who can blame them? There is literature here here and here that point out that physicians will readily resort to deception get a proposed treatment covered.

To really make the point, check out this just published Health Affairs article. Using a validated survey on a representative sample of physicians (obtained from the AMA's masterfile, which contains all practicing docs, not just AMA members), about 10% of the respondents indicated that, in the past year, they had lied to a patient. A whopping 50% admitted telling a patient that their prognosis was better than warranted.

Egads. This is not only driving medical costs, it's increasing variation and exposing patients to unnecessary and potentially dangerous treatments.

Keep in mind that the fee-for-service Medicare program does not engage in "preauthorization" that can lead to a refusal to pay for a treatment.  Rather, Medicare generally pays all claims and then relies on retroactive audits to "claw back"any improper payments.

While that has shielded the Medicare program from being lumped in with all the allegedly evil commercial insurers, the DMCB asks two rhetorical questions about the few bad apples who are hurting the profession:

Which would you rather have for you, your parent or your child: an insurer that scrutinizes a proposed treatment ahead of time and doesn't necessarily take the doc's word for it, or an insurer that lets things happen and uses hindsight after the damage is done?

And last but not least, how will the new payment mechanisms, like bundling and shared savings, reduce the likelihood of bad behavior?

Hashtag #diseasemgtvalentines

The Disease Management Care Blog just discovered the Twitter hashtag #healthpolicyvalentines and has been watching the love flow.  In 140 characters or less, Tweeps combined policy and passion into tweets like "Let's go on a mandate" and "Uwe got what I need."

With only a day to go before Valentine's Day, the amorous DMCB joins in under the just-invented #diseasemgtvalentines hashtag:

I have the disease, you are the management

I want to populate your health.

You put the oh! in my ACO.

Let me in your risk corridor

PMPM's rising!

Let's make some outcomes

Honey, can I telephony?

Let's go off protocol

I want you at my point of care!

More to follow.....

The Medical Benefits of Exercise (and are the taxpayers getting their money's worth?)

Disease Management Care Blog readers may recall this post on the benefits of exercise in combating depression among persons with chronic disease. Thanks to the GlassHospital, the DMCB became aware of this entertaining, informative, accurate and savvy video on the benefits of putting limits on sitting, eating and sleeping to 23 1/2 hours a day.

It's very interesting:



Contrast the video above (by the way, it's Canadian) with this CDC website (and practically nothing from the Feds on YouTube) and ask yourself two questions:

1) which approach to health promotion more likely to have a greater impact, and

2) is Washington D.C. really giving the taxpayers their money's worth when it comes to awareness of the benefits of prevention?