Healthcare // Analytics
News
6/19/2014
04:36 PM
Connect Directly
RSS
E-Mail
50%
50%

Analytics Help Patients Follow Doctors' Orders

Community Health Plan of Washington's most vulnerable patients are doing a better job of sticking to their treatment plans -- thanks to big data, analytics, and hospital partner Health Integrated.

10 More Powerful Facts About Big Data
10 More Powerful Facts About Big Data
(Click image for larger view and slideshow.)

Ignoring doctors' orders often lands patients in the hospital, perhaps suffering from a new complaint or worsening medical problem that takes longer and costs more to treat.

Each year, about 125,000 people in the United States with treatable ailments die because they don't take their medication correctly, according to a New England Journal of Medicine report (registration required). More than half the 3.8 billion prescriptions written annually are taken incorrectly or not at all, the article said. Poor compliance is linked to between 33% and 69% of drug-related adverse effects that result in hospital admissions. And the Journal of General Internal Medicine reported (registration required) that poor compliance is associated with approximately one-fourth of nursing home admissions.

Noncompliant patients hurt Community Health Plan of Washington's quality of care and service, reducing its performance in objective measurements.

"Upper most, we wanted to improve the care experience for all of our members," Dr. Victor Collymore, chief medical officer at the 20-year-old not-for-profit health plan, told us. "We wanted to identify gaps in risk identification such that we could better target patients in chronic care management. We felt there were opportunities to improve our regulatory and compliance needs."

[Need advice on how much you should be budgeting for IT? Read How Much Should Healthcare Organizations Spend On IT?]

The organization needed to address gaps in scores and ratings, including Healthcare Effectiveness Data and Information Set (HEDIS), Consumer Assessment of Health Providers and Systems (CAHPS), and Healthcare Outcomes Surveys (HOS), he said. In addition, it wanted to raise its Star and Special Needs Plans (SNP) ratings. To accomplish this, the provider partnered with Health Integrated, which pulls data from multiple sources to determine the most at-risk patients and then recommends the best interventions.

Because noncompliance goes against healthcare providers' dual goals of improving care and reducing costs, many organizations want to tackle the matter head on. Increasingly, they meld insight garnered from big data and analytics with a new ability to deploy human resources -- such as case managers and social workers -- in better ways. By giving pinpointed care to the patients who need it most, healthcare providers say, they see patients' overall health improve and costs decrease.

Tracy Korman, executive vice president of market strategy and customer solutions at Health Integrated, told us the Tampa, Fla., firm targets vulnerable populations who often have multiple or chronic conditions and tend not to comply with doctors' orders. It works with Medicare, Medicaid, dual-eligible, and commercial plan members on care management strategies that incorporate technologies such as big data, analytics, and integration to improve health outcomes, enhance satisfaction and compliance, increase income, and lower costs.

These patients often cannot follow doctors' orders because they face bigger challenges such as homelessness, financial instability, joblessness, and emotional or mental issues, Korman said. "We focus on helping them comply with their care plan. For that population, it's really challenging. You really have two choices: You can either deploy an army of clinical managers, and that is not a good business model, or you can use data to solve the problem in a big data, logistics-driven way, the way FedEx would solve that problem."

InformationWeek's new Must Reads is a compendium of our best recent coverage of the Internet of Things. Find out the way in which an aging workforce will drive progress on the Internet of Things, why the IoT isn't as scary as some folks seem to think, how connected machines will change the supply chain, and more (free registration required).

Next Page

Alison Diana has written about technology and business for more than 20 years. She was editor, contributors, at Internet Evolution; editor-in-chief of 21st Century IT; and managing editor, sections, at CRN. She has also written for eWeek, Baseline Magazine, Redmond Channel ... View Full Bio

Previous
1 of 2
Next
Comment  | 
Print  | 
More Insights
Comments
Newest First  |  Oldest First  |  Threaded View
Alison_Diana
50%
50%
Alison_Diana,
User Rank: Author
6/25/2014 | 10:47:55 AM
Re: big data / analytics
Let me see if i can do a better job here in comments. Health Integrated uses data from muiltiple sources such as claims, pharmacies, plan membership forms, etc. So it's not big data in terms of teraflops, to paraphrase Korman; they view it as big data because the information comes from so many multiple sources. HI then uses its proprietary algorithms to identify patterns and potential weaknesses within the population, looking to find ways to intervene (such as case workers, housing, etc.). After identifying those weaknesses, HI moves the relevant population into the appropriate intervention programs -- case management, disease management, etc. They continue using analytics to determine whether these interventions are working; they're measuring results (is Patient X complying? is patient Y improving?).

HI uses apps for its analytics, believing this allows them to be faster and use real-time data. The company describes its approach as a big data hub: "We create very rich apps, front-end visualizations of data where we can show data in a lot of different ways. We put those tools in front of our clinicians and care coaches, and over time those apps develop in a lean and agile way. We very much have created a channel, a fast-track channel to the end user by creating a flexible big data hub in the center of these activities." 
YvonneAtKaiser
100%
0%
YvonneAtKaiser,
User Rank: Apprentice
6/24/2014 | 7:01:27 PM
big data / analytics
The article was not really clear on if they are using "big data", nor can I tell if they are using analytics. Yes, they are using data to identify non-compliant patients, but that's not really analytics. I would be interested in another article that addressed the analytics and the data mining aspects. thanks
Alison_Diana
50%
50%
Alison_Diana,
User Rank: Author
6/24/2014 | 9:53:48 AM
Re: Cutting edge analytics for healthcare institutions
@eCitizen: The numbers are horrifying, aren't they? And from what I recall from previous research, they have not gone down over the years. If you do your own research, you'll find other studies mirror these percentages, unfortunately, so it's not a question of bias. In other words, sadly the data is correct. 

Pilots suggest -- if not out-and-out prove -- that even rudimentary technologies such as text reminders boost compliance by reminding patients to take their medications or measure their insulin or whatever their maintenance plan calls for. Wearable tech -- smart watches or Fitbit-like devices -- could also remind patients via buzzes, vibrations, lights, or other pre-set alarm. If there was a way to tie in an alarm to the actual action of taking the pill, measuring insulin, etc., that could work well for some patients, perhaps...? Technology backed by people who monitor compliance seems, to me, to be the best approach and as we've seen it's most cost-effective since it prevents people needing to return to their (costly) hospital beds. In the case of the patients in this article, they were particularly vulnerable due to economic or other reasons. 

Pharmaceutical companies have to be careful, as they seem to realize, about stepping over privacy and other lines, here. Doctors don't have time to monitor all their patients. That said, you'd think there's room for another type of healthcare worker to take on that task, equipped with the right data and technology. This employee would escalate it to a doctor or nurse practioner if necessary. I think a mix of tech -- including text reminders, the type of analytics solution described here, apps that automatically remind patients that it's time to take medication, and other devices/software/other -- should improve these horrendous statistics.
eCitizens
50%
50%
eCitizens,
User Rank: Apprentice
6/24/2014 | 1:34:33 AM
Re: Cutting edge analytics for healthcare institutions
Alison, back to your original article. You said "Each year, about 125,000 people in the United States with treatable ailments die because they don't take their medication correctly... More than half the 3.8 billion prescriptions written annually are taken incorrectly or not at all, the article said. Poor compliance is linked to between 33% and 69% of drug-related adverse effects that result in hospital admissions. And... that poor compliance is associated with approximately one-fourth of nursing home admissions."

I'm really blown away by these numbers. Do you think a technological solution could be implemented to help patients take the right medications at the right time along with their doctors being able to monitor their doing so?

Or, perhaps this is a third-party solution where a tech solution is used by both patient and non-profit monitoring agency?

At the end of the day these are staggaring failures in our medical system. We have the ability with mobile to do much more careful distribution and monitoring. Why big pharma isn't stepping in is beyond me. Patients that live longer need more medication (my cynical perception of what could motivate them). More importantly of course, so many wouldn't die or need to be admitted in hospitals. 
Alison_Diana
100%
0%
Alison_Diana,
User Rank: Author
6/23/2014 | 5:24:51 PM
Re: Cutting edge analytics for healthcare institutions
That's the billion dollar question, isn't it, @ecitizen?! For one thing, deidentified data and anonymized data MUST be standardized. Anyone can say their data is deidentified -- but they may not be speaking the same language as someone else who says the same thing. It's like encryption; once that word had multiple meanings and was, therefore, valueless because it didn't necessarily mean data was secure. That's how i feel about deidentification and anonymity.
Alison_Diana
100%
0%
Alison_Diana,
User Rank: Author
6/23/2014 | 5:21:37 PM
Re: Cutting edge analytics for healthcare institutions
Thanks, @Anf. I'm of two minds here: There are many patients who see doctors, get prescriptions, but then don't get better because they can't fill their scripts (money or transportation, say); they don't have a real home and move around, perhaps losing medication or having it stolen; get sick because they don't eat well or regularly, etc. By using the type of tech described here to help these patients hook up with the right social services -- food stamps, county nurse, etc. -- at the right time, these organizations could be saving or vastly improving these patients' lives. Oh and, btw, even saving taxpayers some money because that pre-hospital care is less expensive than hospital, post-op, and rehab care.

OTOH, I too am leery of all this PHI data. As I said in my "deidentified data" story, I don't use those personal health devices or apps precisely because I don't like the idea of all my sleeping, moving (plus GPS) info, etc., being funneled who knows where. And I don't wear a tinfoil hat either! :)
Alison_Diana
50%
50%
Alison_Diana,
User Rank: Author
6/23/2014 | 5:16:21 PM
Re: Using existing prescription databases
I did not know about this -- although its existence doesn't surprise me, given the erosion we've seen of privacy, including personal health information privacy over the years. I am going to look into the link and story you provided. When my daughter needed confirmation of an innoculation for middle school, I hadn't got that info for one condition. The principal was able to look it up on a public database. Sure, it meant i didn't have to drive over to the pediatrician's office for my daughter's record -- but I'm not thrilled that anyone and their uncle can access this data. Last I heard, innoculations were health data too. And, you'd have thought, private.
eCitizens
50%
50%
eCitizens,
User Rank: Apprentice
6/23/2014 | 2:46:13 PM
Re: Cutting edge analytics for healthcare institutions
To be clear I'm not saying the information shouldn't be collected and shared. I am saying it should be highly protected and anonymized. This is a treasure trove but, gets sour really quick when it's tied to a person's identity. 

I do see massive value in this but, why not serialize every record without revealing identity? That way for a lifetime we can track a person's prescription history? The problem lies in who holds the final key? 

The opportunity is screaming for an encryption system but, with such an untrustworthy government who should we trust? Big pharma? Our tech giants? Other corporations? What's more important, privacy or knowledge? Unfortunately without privacy we have only a taste of the excesses abusers will go, what happens when those excesses turn in to the status quo? 
AnfH156
50%
50%
AnfH156,
User Rank: Apprentice
6/23/2014 | 2:37:25 PM
Cutting edge analytics for healthcare institutions
This is a good post on one of the hottest topics out right now. This new trend is revolutionizing the medical data industry. Ecitizens is correct but now it's going a step further. For instance this company I found while doing my research is changing the way things are done. Due North Analytics started their company knowing they could predict rates, future symptoms or disease, and determine the effectiveness of the medication we use to treat those diseases. Check out their site and see for yourself. http://www.duenorthanalytics.com/blog/
eCitizens
50%
50%
eCitizens,
User Rank: Apprentice
6/23/2014 | 1:43:09 PM
Using existing prescription databases
Most are unaware that all prescription data, tied to our identities, regardless if we pay cash, is available via ScriptCheck.

They collect all data, yep yours too, and use algorithms to predict what ailment or condition you have. They sell that to at the very least big insurers.

HIPPA laws? That's funny. We're a sieve of data with a smokescreen of laws proporting to protect our privacy.

Here's the story: http://articles.latimes.com/2013/oct/21/business/la-fi-lazarus-20131022

It's our privacy. Do we want it or not? Or, are you thinking if you haven't done anything wrong, with nothing to hide, it's not big deal? In either case, please post your email and banking logins and passwords here so we can share. It's not a big deal right? :)
Big Love for Big Data? The Remedy for Healthcare Quality Improvements
Big Love for Big Data? The Remedy for Healthcare Quality Improvements
Healthcare data is nothing new, but yet, why do healthcare improvements from quantifiable data seem almost rare today? Healthcare administrators have a wealth of data accessible to them but aren't sure how much of that data is usable or even correct.
Register for InformationWeek Newsletters
White Papers
Current Issue
InformationWeek Tech Digest September 23, 2014
Intrigued by the concept of a converged infrastructure but worry you lack the expertise to DIY? Dell, HP, IBM, VMware, and other vendors want to help.
Flash Poll
Video
Slideshows
Twitter Feed
InformationWeek Radio
Sponsored Live Streaming Video
Everything You've Been Told About Mobility Is Wrong
Attend this video symposium with Sean Wisdom, Global Director of Mobility Solutions, and learn about how you can harness powerful new products to mobilize your business potential.