Patients Make More Medication Mistakes

An interesting review by the Wall Street Journal (WSJ). It turns out that the number of serious medication errors people make is on the rise, leaving about one-third of such people in the hospital.
“Every two minutes someone calls a U.S. poison control center about a medication error. They took the wrong dose, took their medication twice or accidentally took someone else’s, among other mistakes.”
The WSJ report quotes a recent publication by researchers from Ohio, entitled: Non-health care facility medication errors resulting in serious medical outcomes. The researchers examined data from the Poison Control Center.
The WSJ review and the original publication have more valuable data. To us, this comes as no surprise. The complexity of prescriptions, the long duration of most of the prescriptions and the ever-changing appearance of the generic medications makes it very difficult to fulfill doctors’ instructions even for the most motivated patients. And, needless to say, we believe that our mobile Pill Recognition technology can help to reduce patient medication errors which leads to improved health outcomes at a reduced cost.

Drugs Could Soon Come With a Money-Back Guarantee

The US healthcare system is undergoing through a major process of changing business models. The government and private insurers have been trying for years to move away from the fee-for-service system that pays doctors and hospitals based on the volume of treatments they perform. The accountable care model is all about contracts that reward quality and better outcomes. Medicare, the federal insurance program for seniors, plans to make half its payments through such arrangements by 2018.

Bloomberg Business has published a comprehensive review about this topic. To achieve a higher quality of care, drugs makers may be asked to link the prices of the drugs they sell to how well those products actually work. The shift in business models could help address a long-standing problem: The benefits observed in clinical trials don’t always materialize when a treatment is deployed in the real world.

According to Bloomberg’s article, linking payment to performance, while appealing in theory, is tricky. One challenge for drug companies: Patients need to take medications correctly and consistently for them to work. A drug’s benefit may also depend on other factors beyond the manufacturer’s control, like the patient’s diet.

At Irody, we are busy developing our novel ActivPrescription™ technology that is offering a new way to measure patient compliance and also assist patients in managing their therapy. We strongly support this statement from the Bloomberg’s article: “To create the proof of the outcome, you need to know what’s going on with that patient on a regular basis”.

Medical Privacy and Pill Tracking

The FDA has recently approved a first-of-its kind “digital pill” from Otsuka Pharma and Proteus Digital Health. That pill is for treating a psychiatric condition. Potentially, third parties could track patients’ use of that particular psychiatric medication and tell if they see that a patient is not doing what the doctor ordered.

NBC News has brought up an interesting angle for pill tracking technology: “The upside of a pill that signals doctors is that patient compliance can be enhanced — and it needs to be. Twenty percent of prescriptions are never filled. Half of all people who do get their medicines don’t take them in the way they are supposed to! So, figuring out ways to get you to take you medicine appropriate is important. But, developing snitch pills also has obvious dangers for patients.

The same concern is echoed in Russell Brandom’s article at The Verge: What your doctor can know others could know — or demand to know.”

Indeed, we believe that tracking pill use should be left to patients, and we encourage them to do so by offering a series of compelling services that provide value to them, in addition to the obvious value of just tracking.

Embedding a chip inside a pill, practically forces digital tracking. We believe this is a measure that is too hard. Instead, we suggest to use our smart phone technology that can identify pills, just before swallowing. Not only can the app detect pills that may be taken not on time (a.k.a. patient medication errors) but it also tracks the use of medications and keeps a digital validated record. The data is in the control of the patient. We do not keep any identifying details of any kind and only the patient can allow access to his or her data by any third party.

Real Solutions to Medication Error Problems

As we discussed last time (see Adverse Drug Events: A Pervasive Problem, below ), it was remarkable to see two major publications talk about the problem of medication errors and seniors without mentioning electronic solutions. The problem is complex, involving the fact that certain medications are more likely to cause adverse effects in the elderly as well as the general problem that 53% people in the US do not have the health literacy needed to correctly take or dispense prescription medicines.

This complex problem requires a sophisticated—but straightforward—and multifaceted solution. At the heart of the solution is electronic health records (EHR). While the end-to-end solution is not there yet, it’s coming, and it will potentially save this country billions of dollars in healthcare costs.

In a fully implemented solution, each patient’s EHR would list their prescriptions, with interconnectivity between providers so that the full list of prescriptions is available to all. (Vitamins and supplements should be part of this list, although that may provide more of a challenge, since there’s no scrip associated.) Additionally, there should be an electronic capacity to check for drug interactions—whether this occurs with the physician, at the pharmacy, or through the health insurer (whose motivation to reduce unnecessary hospitalizations is perhaps the greatest).

Finally, there must be a way to know that the patient is taking their prescriptions as directed, with checks to make sure that a) they have done so and b) they don’t take too much of any drug (i.e., the wrong medicine at the wrong time).

We’re the only company that has a solution to this last part in our medicine identification app. Its current functionality includes reminders to take medicines and warnings when the patient is about to take the wrong medication. Just as importantly, there is an accurate record of everything a patient has taken. Since we adhere to interchange standards, the information about what patients have really taken (including those vitamins and supplements!) can be shared electronically with pharmacies and physicians.

The result will be a more accurate record of what the patient has taken as well as the ability to red flag potentially dangerous errors or interactions.

Adverse Drug Events: A Pervasive Problem

Often we come across articles in major publications (The New York Times and The Atlantic) about elderly patients and adverse drug events (ADEs). Elderly patients are particularly susceptible to hospitalizations due to ADEs. The Atlantic article cites nearly 100,000 emergency hospitalizations caused by medicine events in older patients.

One of the causes is that doctors often don’t know what their patients are taking because their patients see multiple doctors. What surprised us about both these articles is that the authors recommend paper solutions to the problem—a list of medicines kept in the patient’s wallet.

While today’s seniors might not be as wired as the rest of the population (according to Neilsen, the 25–34 group led the way in smartphone usage, with a 62% adoption rate as of Q3 2011; only 18% of those over 65 had them), “soon-to-be” seniors are adopting smartphones at a rapidly increasing pace. Those ages 55–64 are adopting cell phones at a faster rate than any other age group. (Tablet ownership is also increasing rapidly: see this Pew report for the pre- and post-holiday numbers.) The percentage of smartphone users in the 55–64 age group was still relative low (30%), but mobile health app developers must pay attention to the aging population’s needs and abilities.

We certainly are. We provide solutions that can do so much more than a paper solution. Here is the form that the American Geriatrics Society recommends seniors fill out and carry with them. Can you find any shortcomings in this form? Do you think this form is going to help reduce medicine-related hospitalizations?

We’ll continue this conversation next time.

Mobile Health: What Do Patients Really Need?

Last time, we talked about what research suggests patients want in mobile health: to find health information and track progress. So far, people are tracking their progress in nutrition and fitness (see this article on the top health apps), which is helpful but only a beginning.

At Irody, we’re focused on the dual problems of medicine noncompliance and patient literacy, summarized by these two stats:

  • 50%–75%of US patients do not take their prescribed medications correctly.
  • Almost 90% of US adults have problems understanding everyday health information; only 53% in a recent study had the health literacy needed to correctly take or dispense prescription medications.*

The result? About 125,000 people in the US die of treatable conditions in the US each year because they did not take their medication correctly. Medicine noncompliance costs the US about $8.5 billion annually.

So what do patients need? They need to know that they’ve taken the right medicine at the right time, and they need to know when they haven’t. This need combines nicely with what patients want, which is to track their progress. Irody’s medicine identification app enables both, with potential to correlate medicine compliance with other health information through partners.

* For more stats on medicine compliance, see this study from the Journal of Applied Research. Some of these stats are truly alarming. For the study on health literacy, see the National Assessment of Adult Literacy results.

The National Library of Medicine and Pill Image Recognition

According to The National Library of Medicine (NLM), unidentified and misidentified prescription pills present challenges for individuals and professionals. Unidentified pills can be found by family members, health professionals, educators, and law enforcement. The nine out of 10 US citizens over age 65 who take more than one prescription pill can be prone to misidentifying those pills.

NLM is having two projects centered on the visual appearance of pills: The Pillbox project and the C3PI (Computational Photography Project for Pill Identification). The Pillbox project has created a library of about 3,500 pill images and the C3PI ~1,800. The primary difference between these projects is that C3PI aims at the creation of a library of images that should be suitable for machine interpretation while the pillbox project collects images as part of a manual identification tool.

We have found out that the larger library of the Pillbox project can be effectively used for our “pill visual search” feature that is included in our MyPillSense™ app, and we use this library in the app under a license from NLM.

To achieve the difficult challenge of high throughput and a highly reliable automatic recognition of pills by a smart phone, we are using our novel and proprietary computer vision algorithms, together with our own image database. The pill database that we are having is growing all the time, as new users scan their pills. Our technology enables inclusion of new pills the minute they are being used by patients, and this includes new generic medications, medications that look differently at different geographies and in general allow for rapid creation of a large database using modern “user generated content” methodology that has proven itself as efficient in other projects that aggregate real-world information, such as Wikipedia and Waze.

Why do pills look different each time I fill the same prescription?

Consumer Reports have noticed that each refill of the same generic medication may result with pills that are different in shape and color. This is fairly common. The same generic drug, made by a different manufacturer, can indeed look different. Due to certain patent laws that govern brand-name medications, generic drug manufacturers are not allowed to copy how a brand-name pill looks in terms of its shape, its color and its size. Additionally, generic drug manufacturers can also take it a step further by producing generic pills that look different from another manufacturer’s generics pills.

Pharmacies often change suppliers of generic drugs. The result is that you might receive different looking pills from month to month, even though it is for the same generic medication. A different looking drug can also be dangerous for those who are depending on color or shape to differentiate their pills.

This situation of same medication that looks different on refills, further emphasizes the need for people taking medications to have a capability of automatically identifying what they take, just before taking it, to prevent errors. At Irody we believe that by turning the phone’s camera into a “pill sensor” we can help patients to get the benefit of lower cost medications, while keeping them safe.

Fitbit confirms the need to track ‘inputs’ – but can it do it as good as it tracks activity?

Yesterday, Fitbit, known for its activity trackers, which are wearable devices that measure data such as the number of steps walked, quality of sleep, and weight, announced a series of new products. Of particular interest to us is their new app on Microsoft’s platform that now has a capability to track food. Fitbit is doing so by using natural language recognition, where a user can dictate what she just ate to the app.

This announcement comes right after Apple has included a caffeine intake tracking feature in its brand new iHealth app. Apple is offering users to manually input how many cups of coffee they consume in a day. Jawbone has a similar coffee tracking app announced earlier this year.

We believe that moving into tracking coffee and food is a clear validation that tracking health should include the ‘inputs’, which include food, beverage, and also vitamins and medications.

Before Fitbit and Jawbone came up with their activity trackers, people could track activity by writing down how many miles they walked or run in a notebook. Such data is more qualitative than quantitative. The introduction of the trackers has enabled a much more precise and ‘true’ activity tracking, not to mention the convenience (which comes at a price, though). But – when these companies come to tracking the inputs, isn’t their solution similar the old manual tracking of activity? The method they use is based on user input, and not on an objective measurement, leading to data that is not as good as the activity tracking data.

At Irody, we offer a quantitative way of tracking vitamins and medications, by scanning them. Because this is an objective measurement, the data that we produce is on par with current standards that are expected from a modern tracking device.