Population Health Management

As someone who lived in Wisconsin for 3 years, "population health management" sounds like issuing licenses to hunters to shoot deer. As applied to people, however, the hunt in population health management is for behavioral causes of disease within a population, and then identifying strategies to prevent the disease. Sounds harmless enough. To find the diseases and causes does require aggregating personal information of people within the population, raising some objections.

In Buck Consultants 2009 survey, several employers engaged in "managing wellness" announced 2 vision statements that are remarkable, especially since the trend has been for employers to shift away from defined benefits and towards defined contribution plans. Perhaps this shift is not indicative of an overall decrease in taking responsibility for employees' healthcare.

• “valuing employees’ health as much as their intellectual knowledge”
• “our vision is to make people feel better than they ever thought possible"

No doubt each of these statements will include qualifications and disclaimers of liability prior to becoming actual policy. Indeed, finalized corporate HR policies are often difficult to discern due to the care that is taken to avoid offense. If the above statements really do become the core of health benefits policy, it is worth considering the implications in the context of privacy and anti-discrimination regulations.  

Valuing employees' health seems to give the employer the right incentives during the course of the employment relationship, but the hiring and termination decisions are also implicitly impacted by valuing employees' health. With average employment tenure approaching just 3 years, firing and hiring happens a lot.  While various regulations protect employees in the hiring and firing process, those regulations cannot reach secret motives. In this case, the employer is making its preference for the health of its employees explicit. Careful wording of such preferences will be required to steer clear of privacy and ADA issues here.

California choice of law in class action on Canopy Financial

Given all the attention to Canopy Financial in the recent posting, and then the relative paucity of content, some explanation is in order. In the course of transferring my license from Wisconsin to California, I’ve been heads-down in the law of the Golden State. In case you were wondering about the rationale for those “choice of law” provisions in contracts, here’s a quick note on how class action lawsuit procedure would work differently under California rules for a case like Canopy. It appears that a class of California accountholders, rather than all the accountholders nationwide, would be the most likely approach to succeed. Under California class action law, the court can make the defendant pay for the notification that goes out to all the accountholders. Under other rules, a single representative class member must pay those fees (although they could be fronted by the law firm).

Primer on choice of class action procedure

Yes, that’s right, the specific court in which the trial is heard makes a difference. While California Courts will generally enforce a contractual choice of law if he state whose law was has an interest in the parties’ controversy, if the state’s interests are materially greater than the interests of the state whose law was contractually chosen by the parties, California State law applies. Brack v. Omni Loan Company, CA Court of Appeal - 4th District, No. D049198, July 16, 2008.

The best point of comparison for California class action procedures is Federal Rules of Civil Procedure (FRCP).  Corporate defendants tend to remove their cases to federal court, and federal courts apply Federal Civil Procedure. Assuming that the suit is brought in California on the basis of Canopy’s negligence, removal to federal court is possible if the amount in controversy (aggregated among class members) exceeds $75,000 and the corporation is not a citizen of California. Federal Courts are required to apply state law on substantive matters, but the FRCP trumps state procedural law, even on matters even as

It is worth noting that the Class Action Fairness Act of 2005 (CAFA) does provide an alternate means for obtaining or losing federal court jurisdiction. CAFA attempts to ensure that where class members are the primary defendants are citizens of the same state, the case will be heard in state court. The reverse is also true – if the class contains defendants from multiple states, they may sue in federal court.

FRCP requirements

1.      Numerosity

2.      Commonality

3.      Typicality

4.      Representative is adequate

Where money damages are being sought, it must also be shown that common questions of law and fact predominate over individual questions among class members, and that class action is the superior method for handling the dispute.

Once the class is “certified” on these bases, and counsel are appointed for the class, the court must send out a notice to all identifiable class members. This gets expensive, especially since it must be paid for by the representative.  Compare this with the California requirements, below, where the court may decide to make the defendant pay for the cost of the notice.

California class action

In California, a class action requires both an ascertainable class and a well-defined community of interest. At first, these sound like the same thing. A “community of interest”, however, is further defined as a determination of whether

1)      Common claims predominate

2)      Representative is adequate

3)      Class will result in substantial benefit to the parties and the court

Well, 2/3 of those factors are the same as under FRCP. The big difference, in my cynical view, is that individual notice to potential members of the class is not required, and the court can make the defendant pay for any notice that the court does require.

What was it about Canopy Financial that enamored VCs?

Canopy’s valuations implied a greater than 20-to-1 price to earnings (P/E) ratio, far exceeding your typical financial service provider’s P/E. What was it about the health spending accounts administered by Canopy that promised big scalable growth? What was it that promised to make Canopy the next Google or Microsoft? Two words: health data. While Google and Microsoft reinvent themselves with their own proprietary personal health record formats, Canopy had a stable system for collecting this data through the auspices of major financial institutions. The Google and Microsoft network has a great many nodes, many of which require costly business development efforts. Canopy, on the other hand, could touch hundreds of thousands of individuals healthcare transaction records through integration with a single bank. Those individuals’ millions of transactions are like the millions of individual decisions that Google’s search engine crawls to make the web meaningful. Canopy was going to out-Google Google on health, 1/7^th of the U.S. economy.

Consumers using the Canopy platform had the ability to tag specific health spending account transactions with a particular category of medical care for a particular beneficiary, and electronically attach receipts to support tax deductions. This metadata was laborious for the accountholders to compile, but very valuable once compiled. In aggregate, it was the beginning of a cost-of-care database that spanned a variety of health plan designs.

Health plans have long been in the business of collecting cost information, but according to Dr. John Langefeld of Claremont Partners, health plans typically confound price sensitivity data by making macro benefit design changes in response to macro utilization trends. The constantly permutating benefit designs make it tough to have valid longitudinal data about individuals’ decisions for a particular clinical procedure. Canopy’s data, on the other hand, cut across multiple major health plans and geographies, and typically with a focus on pre-deductible spending.

This morning I learned from a former Canopy executive that at least one of its forward-looking bank clients looked to process of HRA and FSA substantiation to feed price data back from its accountholders back to its accountholder community. For the most part, however, Canopy’s clients were wowed by Canopy’s health data in theory rather than in practice. While Canopy’s bank and insurance clients were enticed by health data capabilities during the sales process, these clients seldom took full advantage of the capabilities that existed.

One hindrance to full realization of the potential of aggregate health data from bank transactions is concerns about privacy. Even once HIPAA compliance is assured, advances in technology create an arms race of technology to secure the data. Recent research at the University of Texas showed that the data must be anonymized to near meaninglessness in order to prevent skilled deanonymization. Nonetheless, aggregate health cost and quality information provides tremendous utility to society and individuals, and the transactional side of health care is the most scalable way to collect it.

For readers interested in porting over Canopy’s health data and/or developing a road map for consumer-participatory healthcare, please contact me here. As mentioned in previous postings, I am putting together a remediation team with experience in addressing these issues.

Remediation Plan for Canopy Financial Omnibus Custodians

The meltdown at Canopy Financial is apparently placing client financial institutions' customers at risk sooner than expected. With most of the employees laid off, there are serious questions about how much longer the servers will keep running. I myself have an account at Sovereign Bank and am monitoring the situation. For those of you at Sovereign, Wachovia, Fifth Third, or CareMark, I am liaising with ABA HSA Council members to coordinate a response.

Some alternatives to Canopy are readily apparent. Lighthouse 1 has announced its interest in helping former canopy clients. Metavante would be another choice. Both of these vendors already assume an omnibus model, just like canopy. As a public company, Metavante might better smooth anxieties after the Canopy fiasco.

The project plan for the actual transition will be more complex. A couple suggested starting points:

1. Sourcing replacement: make/assemble/buy. As a vendor and a purchase, I've used RFPs for HSA services that ran to hundreds of pages.

2. Managing customer reaction: needs quick PR pieces re: security of data, with focus on HIPAA processes and how the PHI on the system will be retained

3. Managing transition to new CIF, if possible, so as not to fracture the back office.   

If you are interested in learning more, drop me a line. I have a number of executives from the credit union, community banking, and software services industries that are collaborating on a solution.

Tough Times for HSA Software Vendors

In a recent post, I cited HSA adoption figures from Canopy Financial. TechCrunch reports today that allegations of financial irregularities may be taking Canopy down. As pointed out by the TechCrunch, Canopy had been widely perceived as the market leader. At present, their website is down.


Could it be that the health spending account software model just doesn't work? Earlier this year, Members Health Network, the leading HSA software vendor to the credit union industry, liquidated its assets. According to an executive at HSA Bank, which has done custom work to modify a more standard core accounting system, there is just "not enough money in HSAs to split between the custodian and a service provider like them."


It will be interesting to see what other firms may attempt to take their banking clients, like Sovereign Bank, who had charged ahead touting a whitelabeled version of Canopy's product. Yet another joy for Santander, which acquired Sovereign last year. My bet would be http://www.hstechnology.com/, which has been around for a long time cautiously growing its network of insurance brokers that support its product. [EDIT 12/02/09 -- HST 's parent company, according an an executive there, has "made a decision to focus our human and financial capital on the Insurance Company that we acquired and reduce our emphasis on HST".]

Reid's Funding Mechanism for Senate Bill Rewards Cats

According to today's Washington Post, Senate Majority Leader Harry Reid will shortly be announcing a plan to pay for healthcare expansion by increasing the federal Medicare payroll tax on high-earning individuals. This would be a big win for those firms that coordinate consumer-driven healthcare mechanisms with employers to exclude income from the federal Medicare payroll tax. Current law sets the tax at 1.45 percent of income, an amount matched by employers. An increase to the tax would increase the substantial incentives already existing to coordinate health care around a site of employment (see earlier post on cats versus dogs for more on these incentives).

The increased tax would likely be restricted to individuals earning $250,000 or more. These individuals already have an incentive to contribute the maximum allowable amount to a family health savings account (HSA), $5950. While the employer's share of the Medicare tax avoided through those contributions is only $86, it pays for a lot of account administration fees. If the tax doubles, the incentives of employers to put HSA utilization programs in place roughly doubles as well. Using an ASP software service costing a few dollars per employee per month, the employer would net a substantial savings across a broad employee base.

Cats versus Dogs in Consumer Driven Care

In this multi-part series, I'll take a look at the cats and the dogs of electronic health records (EHRs), health savings accounts (HSAs), wellness companies, and the software companies that power them.

As a general rule, cats like places and dogs like people. In electronic health records (EHRs), the cats are the Epic Systems-type databases that capture everything relevant to a particular provider. The dog EHRs follow the patients around, attempting to interface with multiple provider settings, like Athena Health.

In the world of HSAs, the cats like to cuddle up to a place of employment. Credit unions are the most numerous cats of the HSA world. Several large credit unions sponsored by Fortune 500 businesses have more than 90% of the employer group as credit union members. While an employer-sponsored credit union will let you stay a customer if you lose your job, a big chunk of employer-sponsored credit union value proposition comes from the existence of a relationship with the select employer group (SEG). Credit unions with strong SEG relationships tailor their products the sponsoring employee group. Baxter Credit Union, for example, also has SEG relationships with CDW and Cardinal Health, each of which gets to see a differently designed website. The credit union makes sure that HR people at the employer are well-stocked with informational material about the credit union, may offer discounts to current employee within the SEG, and are sure to show up at benefits fairs to answer any questions about complex products like HSAs.

While complex, HSAs offer powerful advantages in a setting of high employee penetration. With more and more employers scaling back to high-deductible health plans (HDHP), we will likely to soon see more Fortune 500 employers go "full replacement" with HDHPs and HSAs. For those large employers with credit union relationships, suppose 90% of the employee group already treats the credit union as its preferred financial institution. That means the transaction accounts for medical spending are likely to be all in one place. This is a tremendous boon for the employer, who with appropriate technology can make sure that those transactions run through the HSA, boosting FICA savings. Using the latest figures from Canopy Financial, which provides HSA account management software for banks, the average family's HSA received $3528 in contributions. [UPDATE: need to double check these figure given recent news on Canopy] This generates FICA and Medicare tax savings of up to $269 for the employer per employee. At the same time, however, half of the people eligible for an HSA don't open one. For a 100,000 person employee group, that adds up to over $13 million in annual payroll tax savings forgone (50,000 * $269).

Technology solutions now exist that automate both creation of the account, and to “touch down” normal checking and credit card transactions for medical purchases in the HSA, so that the employer gets the full tax benefit. For those employees that otherwise would not contribute to the HSA, an automated routine can identify their actual medical expenditures from non-HSA accounts (or EHRs) and momentarily touch down a matching amount from their next paycheck into the HSA prior to moving the funds into their normal direct deposit account. Without any decrease in availability of paycheck funds to the employee, the employee has now accrued a tax advantage. With the average family's out of pocket medical expenses for HDHPs reaching $2620, an employee grossing $44,000 gets a $620 tax savings, equivalent to an $812 pre-tax raise.

In theory, this kind of technology could be deployed anywhere. In practice, it's a lot easier when the normal transaction accounts, direct deposit accounts, and HSA are all in one place. With the upwards of 90% penetration of the employee population select credit unions, the tax benefits are much easier to capture there.

In addition to maintaining stronger employer-sponsor ties, credit unions tend to be more technology savvy than their banking counterparts. Regulations prevented credit unions from running their own back-office systems until relatively recently, so they are much more likely to be real-time central information file (CIF) systems than their banking counterparts. Also, the capital structure of credit unions has necessitated acclimation to technology-powered growth strategies. While banks can raise funds by issuing equity, credit unions are restricted to the paid in capital of their members. With new branches costing $1 million, credit unions have traditionally focused on growing through non capex spending like internet banking.

How will all this fare in the next round of federal healthcare reform? Roy Ramthun, former advisor to the White House on Health Savings Accounts, predicts that credit unions are better poised than banks to influence consumer driven care in federal health reform. First, Roy pointed out the CUs are better positioned financially than banks. Second, Roy indicated that mobilization of grass-roots will be key. While "death panel" screamers at town hall meetings this summer didn't do much for the anti-reform crowd, credit unions have historically been very effective at mounting well-organized shows of support. With the opportunity to save their employer sponsors millions of dollars, my guess is that the credit unions will meow for retaining HSAs and related employer tax breaks.