Thursday, November 20, 2014

Using phone and fax to improve diabetes care

The Journal of Diabetes and Its Complications just published a study on "Diabetes self-management support using mHealth and enhanced informal caregiving" by a team of VA researchers from Michigan and Stanford. The team set up an IVR phone system for veterans with diabetes. The vets received phone calls every week to assess their status and generate automatic tailored messages about how to deal with their disease. The system was designed to also engage family members and other informal caregivers as well as the clinician. This was a fairly sophisticated system with multiple tree-structured algorithms that took 5 to 10 minutes per call to complete and covered a wide range of issues including hypoglycemia, hyperglycemia, medication adherence, foot self-care, self-monitoring of blood sugar and blood pressure, and so forth. Clinicians received a fax if the system detected a concern.

The paper reports very positive short-term outcomes.  72% of patients agreed to try it and 84% of the scheduled calls were completed. On average, the system detected about one problem for every 5 weeks of patient participation. Patients who had informal caregivers tended to have fewer problems. Of course, the Pareto Principle applies here as everywhere and almost half of the problems were concentrated in 15% of patients. The best news is that the number of problems seemed to decrease over time, suggesting that the patients were achieving better stability.

At this time, there are no data on whether the patients actually got better in terms of physiologic control, complications, utilization or satisfaction with care. However, it looks like the relatively old fashioned technologies of phone and fax still have great potential to enhance care by engaging patients, caregivers and clinicians.

Aikens JE, Zivin K, Trivedi R, Piette JD. Diabetes self-management support using mHealth and enhanced informal caregiving. J Diabetes Complications. 2014;28: 171-6.

Monday, September 8, 2014

Individualizing content for patient engagement efforts

I'm struck by the wise comments presented in this recent piece from mHealthNews. Editor Eric Wicklund wrote about the efforts of Kyra Bobinet, MD, MPH to increase the "emotional intelligence" of communications between provider and patient. "If it's personal and relevant," says Bobinet, "that should work." 

I couldn't agree more. Generic messaging of the "everybody ought to get a flu shot" variety has little resonance with the patient and is unlikely to impact behavior in a big way. Specific, tailored messages along the lines of "Because your last A1C diabetes test was high, you need to recheck again now" are more motivating. Of course, they are much harder to produce if you don't have an automated system to scan the information you have about the patient, identify the most important opportunities to improve care, and reach out to the patient in language they will understand via media they will accept.

This is not the sort of process that most practices can do on their own. It requires some infrastructure and more than a little staff time to do all that communicating and engaging! Patient Engagement Systems offers a software-as-a-service solution that allows you to connect to your patients, keep them engaged in their care, and has been proven to improve outcomes and reduce the total cost of care. As more and more practices, payers and Accountable Care Organizations realize they need to increase engagement on a large scale to improve quality and reduce costs, they are turning to automated systems to do the job.

Sunday, June 29, 2014

Diabetes prevalence goes up - again!

The good folks at the Centers for Disease Control recently put out the 2014 National Diabetes Statistics Report. They estimate that over 29 million Americans have diabetes. That puts the overall prevalence at 9.3% or more than one in 11 people. The vast majority of them are adults, with prevalence rising to over 25% after age 65. There are 1.7 million new cases per year. In addition to the huge burden of illness, disability and premature death, the current cost of the diabetes epidemic is $245 trillion per year.

But there is some good news. Patients with diabetes who are engaged in their care, especially with primary care services, have better outcomes and require a lot less time in the hospital. Their costs are thousands of dollars lower per patient. The tools to do this are available at remarkably low cost in just about any kind of primary care setting.

New payment structures mean that the incentives to reduce costs and improve quality are in place. Accountable Care Organizations (ACOs) can't just pass the costs of bad quality along; they need to capture the savings that have been proven to occur with this kind of Patient Engagement, and they are starting to do it.

If you know of an ACO (or other practice setting) that is motivated to reduce total utilization for diabetes and other chronic conditions without disrupting provider workflow, please let them know about Patient Engagement Systems. Call Toll-free: (855) 870-4337 or email

Thursday, April 17, 2014

Winning the war against kidney failure

Ed Gregg, the master of diabetes epidemiology, and his team from the CDC, just published their latest report on the state of the nation's epidemic of diabetes. The good news is that the rates of diabetic complications - heart attacks, amputations, kidney failure, fatal hyperglycemic crisis, stroke - have fallen quite a bit since 1990. This is almost certainly do to the widespread application of practice guidelines by specialists and especially primary care providers. Monitoring and managing A1C, blood pressure, lipids and early renal damage is paying off at the national scale.

What about adults without diabetes?
"Trends in the population of adults without diabetes were generally not as promising as those in the population with diabetes, with smaller reductions in the rate of acute myocardial infarction, no significant change in rates of stroke and lower-extremity amputation, and an increase in the rate of end-stage renal disease."
An increase in end-stage renal disease? At the same time that we were making significant inroads against renal failure in diabetes (-28% drop in the rate), we see a swelling of the rates in non-diabetic adults by a whopping 65%. Why is this?

First, early stage chronic kidney disease is often un-noticed by providers who may not understand the implications of apparently small increases in serum creatinine. Second, proteinuria, the hallmark of treatable early renal damage, may not even be sought. Third, blood pressure is hard to reduce, often requiring multiple medications and persistent, frequent monitoring. You can't prevent the progression of early chronic kidney disease with once-a-year visits.

These problems in the quality of care are very much like those that beset diabetes 20 years ago, but we have made great strides by becoming aware of the problem, deciding as a profession that we should and shall do something about it, and marshalling all our resources - guidelines, educators, non-physician providers, reminders, lifestyle changes, new medications - and building them into our ongoing processes of care. In many practices, the optimal care of diabetes is a stated goal and the office uses standing orders, automated reminders and all the tricks of modern systems management to reliably deliver state-of-the-art care for diabetes. It's time to do that for early kidney disease.

Changes in Diabetes-Related Complications in the United States, 1990–2010
Edward W. Gregg, Ph.D., Yanfeng Li, M.D., Jing Wang, M.D., Nilka Rios Burrows, M.P.H., Mohammed K. Ali, M.B., Ch.B., Deborah Rolka, M.S., Desmond E. Williams, M.D., Ph.D., and Linda Geiss, M.A.
N Engl J Med 2014; 370:1514-1523 April 17, 2014 DOI: 10.1056/NEJMoa1310799

Thursday, April 10, 2014

Patient Engagement for Suicide Prevention

You can't manage what you can't measure is a truism in quality improvement. In mental health, it goes even further: You can't manage what you won't even talk about. Although lots of barriers have fallen in the realm of acceptable topics for conversation between doctors and patients - we no longer avoid saying "cancer" like the word itself was the disease - suicide is still a particularly difficult topic for providers to broach. No doubt training the doctors to deal with this sticky area head on can help, but it takes two to tango, and physicians hate to initiate a difficult discussion as much as anyone else.

So, Ruby Shah, Richard Kravitz and the good folks at UC Davis went at the problem from the other end. They used patient engagement techniques - a tailored interactive media program shown in the waiting room - to activate the patient to discuss their symptoms (both emotional and physical) including thoughts of self-harm, and the possibility that those symptoms were due to depression. Among patients with mild depression, there was little impact, perhaps because there was little need. But among patients with moderate to severe depression, the probability that suicide would at least be discussed went from 40% to 58%. 

Is this enough to make a difference? Talking about the issue is only the first step. Did it result in appropriate action? Kravitz and company already reported in JAMA that the engagement approach increased treatment (or referral for treatment) from 16% to 26%. So, the intervention works in the sense that it sometimes stimulates the necessary discussion and appropriate treatment. However, the depression scores 12 weeks later were not better in the engagement group and were even a bit worse. Increased prescribing for non-depressed patients raises the possibility that the engagement program actually made things worse for some patients.

This study was excellent science, but not a clear cut success for patient care. The tailored messaging had an impact on behavior and resulted in treatment changes. However, some of those treatment changes could have made things worse, not better. Twelve weeks later, there was little evidence that the patients felt better - probably because depression is hard to treat, especially with medications.

Shah R, Franks P, Jerant A, Feldman M, Duberstein P, Y Garcia EF, Hinton L,
Strohecker L, Kravitz RL. The Effect of Targeted and Tailored Patient Depression 
Engagement Interventions on Patient-Physician Discussion of Suicidal Thoughts: A 
Randomized Control Trial. J Gen Intern Med. 2014 Apr 8. [Epub ahead of print]
PubMed PMID: 24710994.

Kravitz RL, Franks P, Feldman MD, et al. Patient Engagement Programs for Recognition and Initial Treatment of Depression in Primary Care: A Randomized Trial. JAMA. 2013;310(17):1818-1828. doi:10.1001/jama.2013.280038

Sunday, March 30, 2014

Is depression a link between diabetes and kidney disease?

A team of epidemiologists and nephrologists led by Margaret K. Yu from the University of Washington and the VA recently published strong evidence that among diabetics, depression raises the risk of End-Stage Renal Disease (ESRD or renal failure) by 85%. This is a very big deal. There are lots of diabetics and lots of them have major depression.

Although it is not yet clear if treatment for depression can avert or delay ESRD for these folks, there are plenty of other reasons to identify and treat them. Even if treating depression doesn't prevent ESRD, it probably saves money, improves productivity, averts suicides, reduces morbidity and makes life better for the patient, the family, the neighbors and the rest of us. Given that the most effective treatment for depression is non-pharmacologic (Cognitive Behavioral Therapy) and doesn't impair glucose metabolism or renal function, the main issue is how so we get these patients into therapy?

Endocrinologists have a role, but they only see a small fraction of diabetic patients, and they often aren't equipped to diagnose and manage depression. This is one of those things that Primary Care does - and is getting better at all the time. More and more PCPs are routinely screening their patients for depression (often with the PHQ-9 - the same instrument used in Yu's study) and are developing systems to provide therapy either in the practice or by referral.

However, PCPs are prone to the same tunnel vision as every other doctor. If the patient isn't in the office, often nothing happens - no diagnosis, no treatment, no prevention. So, here is one more reason to use systematic patient engagement strategies to make sure that the PCP and the patient are interacting. Because that's when the good stuff happens!

Yu MK, Weiss NSDing XKaton WJZhou XHYoung BAAssociations between Depressive Symptoms and Incident ESRD in a Diabetic Cohort.  2014 Mar 27. [Epub ahead of print]

Sunday, March 23, 2014

Stories from the Experts by Experience

Pat Salber at The Doctor Weighs In blog reviewed a publication called Experts by Experience which reports a number of fascinating first-person stories of care gone horribly wrong (or brilliantly right) because of the ways that physicians and patients communicated. Patient Engagement is a new(ish) way of thinking about how patients interact with their problems, their health care and, especially, their providers. This lovely collection of a dozen short stories is an excellent reminder that listening is the most fundamental engagement tool of them all.  Enjoy!

(Hat tip to Bob Beltran at the Latino Medical Journal for pointing this out to me!)

Wednesday, March 12, 2014

Why is it so hard to communicate with cancer patients?

Carolyn D. Prouty and a team of researchers from Seattle, WA, Worcester, MA and Atlanta, GA just published an analysis of provider-patient communication in cancer care. Using focus group methods, they sorted out the causes of communication breakdown as seen by 59 providers. Some of the issues seem fundamental to the setting - patients are overwhelmed and scared - but some seem amenable to systematic interventions. For instance, these include inadequate time to meet with patients, failure of all the providers to agree on what the message to patients ought to be, and payment systems that undervalue communication. The next steps after this good work, of course, are to design and test some systematic interventions. 

Prouty CD, Mazor KM, Greene SM, Roblin DW, Firneno CL, Lemay CA, Robinson BE, Gallagher TH. Providers' Perceptions of Communication Breakdowns in Cancer Care. J Gen Intern Med. 2014 Mar 6. [Epub ahead of print]

Monday, March 3, 2014

The cost of diabetes care goes...down!

In a very surprising report in the American Journal of Managed Care, Peter Cunningham and  Emily Carter presented an analysis of the out-of-pocket costs for patient with diabetes. Over the years between 2001 and 2009, per-patient spending on prescriptions fell from $1,095 to $763, accounting for the bulk of the savings. This is due, in large part, to the shift from brand-name to generic drugs. As life-saving products to control blood sugar, but also cholesterol, blood pressure, depression and other common co-occurring conditions go off patent, they become available to many more patients without destroying their families. During the period of the study, the percentage of patients with "high financial burdens" (spending more than 10% of the family income on health care) fell from 23.9 to 18.6. What a relief - for patients, families and payers!

P. Cunningham and E. Carrier, “Trends in the Financial Burden of Medical Care for Nonelderly Adults with Diabetes, 2001 to 2009,”American Journal of Managed Care, Feb. 2014 20(2):135–42.

Sunday, March 2, 2014

DPS Health's Virtual Lifestyle Management Program

DPS Health has some very impressive results out for their Virtual Lifestyle Management Program (VLM). This is the web-based lifestyle coaching program that a number of health plans, provider and employers have been using to engage their employees in personal healthcare and risk reduction. The program includes multi-channel outreach, recruiting, and support for eligible members, a year-long comprehensive digital intervention with online education and dynamic behavior goal-setting, planning and tracking, and digital coaching via secure messaging and moderated chat. They also use accelerometers, journaling and other engagement tools.

The Government Employees Health Association has used it for a few years in about 5,000 members. After a year, 65% had lost weight and user satisfaction was good. They did a comparison to a matched population that did not have access to VLM and found that expenditures were lower in the VLM group. The saving came to $2,637 per member per year. This is a huge accomplishment and demonstrates the power of active engagement of patients to improve health and lower costs. More details are available here.

We are so impressed by DPS that my company, Patient Engagement Systems, recently announced that it has added VLM to its suite of patient engagement and clinical decision support tools for patients with diabetes and chronic kidney disease (CKD). This partnership means that the two services with the strongest data proving efficacy and cost savings can be integrated into one package, making them even better.

Friday, February 28, 2014

Engaging patients in screening for colon cancer

Colon cancer runs in families. If you have a relative with colon cancer, your chances of getting the disease are markedly higher than if you don't. The good news is that if you get a screening test periodically, the disease can be found early, treated relatively easily, and you can go about your business. The bad news is that we don't do enough screening, especially of first-degree relatives and especially in rural areas where the long drive to colonoscopy is a barrier.
Anita Kinney, a nurse and epidemiologist in Utah, led a teem of researchers from five states in testing whether telephone outreach could get more high-risk patients screened. They found patients at elevated risk of cancer by recruiting the relatives of diagnosed patients. They either got a mailed brochure or a 40 minute telephone call with a genetic counselor plus a follow-up letter (with a copy to their primary provider). This was a carefully designed and executed randomized controlled trial with 481 subjects, so the improvement in colonoscopy from 16% to 25% is not only statistically significant (P<0.001), but unlikely due to bias or confounding. Interestingly, the effect of the telephone intervention was about the same in rural, urban, high income, and low income groups.
Although this is an impressive and clinically important effect, this intervention is a bit costly. The authors don't provide details, but these counselors are highly-trained and sophisticated providers and it takes a lot of their time to handle each case. I have to agree with the authors that this intervention should probably be part of a stepped strategy with the genetic counseling reserved for those who don't get screened after less expensive methods are tried first, especially since they had such a nice response from their brochure.
Now the challenge will be how to integrate this proven intervention into an ongoing delivery model, and that means, an ongoing business model. Who will pay for the outreach and the colonoscopy, especially if the index patient has different coverage than their relatives?

Kinney AY, Boonyasiriwat W, Walters ST, Pappas LM, Stroup AM, Schwartz MD, Edwards SL, Rogers A, Kohlmann WK, Boucher KM, Vernon SW, Simmons RG, Lowery JT, Flores K, Wiggins CL, Hill DA, Burt RW, Williams MS, Higginbotham JC. Telehealth Personalized Cancer Risk Communication to Motivate Colonoscopy in Relatives of Patients WithColorectal Cancer: The Family CARE Randomized Controlled Trial. J Clin Oncol. 2014 Mar 1;32(7):654-62. doi: 10.1200/JCO.2013.51.6765. Epub 2014 Jan 21.

Tuesday, February 25, 2014

The Power of Small Data

Big data is the uncharted territory of our time. Vast, multiple exabyte-sized data sets are like new continents that will, upon exploration, yield insights into all manner of human affairs. This is all well and good. But in the meantime, how about applying the latest tools and techniques of Big Data to much smaller data sets, say an insurer's chronic disease patients?

Today, with high speed data exchange and processing, it is possible to comb through millions of lab tests to find the results of eligible patients, monitor their timing and values, and provide nearly instantaneous feedback to patients and providers. This is the power of Small Data.

Small data connects people with timely, meaningful insights. It is organized and visually presented to be accessible, understandable, and actionable for everyday tasks.

Small data is the right data. You shouldn't need to be a data scientist to understand or apply it for everyday tasks.

Paraphrasing Joseph C. Kvedar, MD, the Founder and Director of the Center for Connected Health, Big Data will guide policy makers, but it's Small Data that will help patients get better.

Living with and managing a chronic disease, such as chronic kidney disease, is complicated. Small Data can deliver the right information to patients and providers at the right time so that they can do the right thing.

Wednesday, February 19, 2014

Using hand-held technologies to engage patients

Two recent articles take a look at how some small devices - smartphones and tablets - can be used to foster patient engagement.

The smartphone project tackled a tough problem  - supporting patients with chronic disease (in this case, HIV) and heavy alcohol use to reduce their drinking. Expanding on their previous approaches using Interactive Voice Response (IVR) telephone calls, Deborah S HasinEfrat Aharonovich and Eliana Greenstein from Columbia University enrolled 43 alcohol dependent HIV-infected patients and followed them for 60 days. The average number of drinks fell from 9.3 per day to 3.9. An impressive 25.6% were alcohol-free at the end of the two months. You can get all the details from their report in Addiction Science & Clinical Practice.

The second article provides the early experience with a tablet system that was given to 30 hospitalized patients at the University of California, San Francisco. The iPad2 tablets had an educational program about patient safety and a link to the hospital's patient web portal. The 30 patients were able to use the devices and liked having them. This preliminary report didn't address any changes in outcomes that wiring up inpatients might bring, so it is a little soon to endorse this approach, but they overcame the first hurdles (getting the machines in the hands of the patients and getting the patients to use them). The authors (S. Ryan Greysen, Raman R. Khanna, Ronald Jacolbia, Herman M. Lee, and Andrew D. Auerbach) do promise that they will investigate post-discharge outcomes in future work. You can find this in The Journal of Hospital Medicine.

These innovations have potential. I'm looking forward to some solid assessments (randomized clinical trials, anyone?) in the next few years to help us separate the effective approaches from the marketing chatter.