Category: e-health

Information is the answer

published date
October 16th, 2006 by

Information is the answer

ABC News is running Prescription for Change, a week-long series focusing on what ails the US health care system and what to do about it. As part of the event they asked me to write an opinion piece, which is published on the ABC News website under the headline More, Better Information Key in Fixing Healthcare. Matthew Holt of the Health Care Blog also contributed an article, entitled Why is Fixing American Healthcare So Difficult?

Here’s my original piece:

The US health care system is a mess by any measure. Costs are far higher than anywhere else in the world but life expectancy and infant mortality are worse than average. To add insult to injury, customer service is often abysmal.

It doesn’’t have to be this way. Other sectors of the US service economy –such as retail, financial services, and lodging– are the envy of the world. Companies like Vanguard, Target and Hampton Inns offer higher quality, better customer service and lower prices than you’ll find in other countries. The US health care system could be like that, too, but the archaic culture of medical providers, distortion of the market by the third-party payment system, and resistance to trading off costs and benefits complicate reform.

The best hope for reforming US health care lies in utilizing information. We need to generate more and better information, make better use of the information we have, and be willing to use information to make tough choices.

The good news is that it is starting to happen.

Generating more and better information

The kind of information we take for granted in other sectors is unavailable in health care:

  • How good is your doctor, really?
  • Which specialist is best for your condition?
  • Is that academic medical center with the great reputation better than the community hospital down the street?
  • How likely are you to die from a medical error?
  • How much does an operation cost at one hospital compared to another?
  • Does that drug you were prescribed work? What were the other options for treatment?
  • How much does the drug really cost (not just the co-pay)?
  • Is the drug any better than a generic?
  • How does your genetic makeup affect the treatment you should get?

These questions are beginning to be answerable. Consortia like Massachusetts Health Quality Partners and Integrated Healthcare Association along with private companies such as HealthGrades provide objective information on quality and patient experience. (They still can’t quite say ““customer satisfaction.”) Medicare’’s Hospital Compare website has information about key quality indicators. Various ““transparency”” initiatives are underway to provide information on costs. Consumer Reports rates drugs and Wal-Mart is showing just how inexpensive generics can be. Genetic profiling is farther off, but groups like the Brain Resource Company are starting to make personalized medicine a reality.

Making better use of information

It’s not enough to make information available. We have to become better users of information and better communicators. We should also analyze and act on data that are currently ignored. Empowered consumers are taking the lead: doing research on the internet, establishing personal health records, demanding a greater role in their care, even negotiating prices. Now we need medical professionals to step up.

Many physicians are lousy communicators. Doctors are hardworking and dedicated, but generally donՉ۪t use email, donՉ۪t keep records electronically, rely on treatments theyՉ۪re comfortable with rather than using evidence-based medicine, and are insufficiently skeptical of information from biased sources (like drug reps). Physicians rarely explain alternative treatments; many do a poor job communicating diagnoses, laboratory findings and information about drugs. Primary care physicians and specialists often drop the ball in communicating with each other about patients they are treating, leaving the patient or caregiver to try to close the loop on their own. And most physicians have no idea what anything costs.

Pharmacists are rarely better. They spend six years in school learning more about drugs than most doctors will ever know, then disappear to the back of the pharmacy to count pills. Pharmacies are convenient and accessible; their pharmacists are a vast untapped information source.

The introduction of electronic health records and electronic prescribing are helping, but are only part of the solution. We need health information exchanges, like the ones piloted by the Massachusetts eHealth Collaborative, to share patient data. We need secure, efficient platforms such as RelayHealth for doctor/patient messaging. We need decision support tools like SimulConsult to organize clinical findings to help diagnose rare diseases, sparing patients the misery of going from doctor to doctor in search of a name and treatment for their condition. We need more patient communities such as Patients Like Me and treatment planning tools such as NexCura.

We should make better use of information from existing diagnostic tools. Advanced analysis software from companies such as AMID mines digital data from inexpensive ultrasounds, reducing the need for expensive MRIs and PET scans. iCardiac Technologies analyzes ECGs already collected for clinical trials, identifying cardiac safety problems early in drug development, before the next Vioxx comes to market.

Using information to make tough choices

Better availability of information will improve quality and customer service, but while information technology cheerleaders promise that big cost savings will inevitably follow adoption, thatՉ۪s by no means certain. Some efficiencies will be gained by eliminating transcription costs and reducing duplication of tests, but better information may actually lead to higher spending. After all, with more information about their conditions, patients may seek additional treatment.

Information can be used to hold down health care costs, but that requires political will. In the UK, the ironically named NICE (National Institute for Health and Clinical Excellence) evaluates treatments based on data and decides if benefits outweigh costs. When the answer is no, as it often is, the National Health Service won’t cover the treatment. In Germany, whole classes of drugs are evaluated to determine whether on-patent products are more effective than generics. If not, branded products are reimbursed at the same (low) level as generics.

If we choose as a society to use information for hard decisions, costs will be contained. But the model will be different from the UK and Germany. WeՉ۪ll use incentives and cost sharing, not outright bans and price controls. Examples are emerging in the public and private sectors.

Within a decade there will be plenty of good information available and we’’ll have learned to use it well. Then we’ll finally be headed toward a health care system worthy of respect.

Help found

published date
September 20th, 2006 by

Help found

In Help wanted I described the difficulties in finding a replacement for National Coordinator for Health Information Technology, Dr. David Brailer. An interim replacement has now been named. It’s Dr. Robert Kolodner, the chief health informatics officer of the Veterans Health Administration, according to Modern Healthcare.

I don’t know Dr. Kolodner, but considering the VA’s track record on health care IT, I assume he’s a safe choice. It’s not clear to me why he’d want the job, though considering its lack of power and budget.

Here comes Healia

published date
September 18th, 2006 by

Here comes Healia

I devote a fair amount of space on this blog to whining about Google (1, 2, 3, 4, 5, 6) . The company retains too much information about searchers and fundamentally is more interested in segmenting searchers for advertisers than in serving the end user. However, I have to admit they are damn good at what they do, and like most people I know, I use Google all the time despite my reservations.

I’ve mentioned before a new health care focused search engine, Healia. It’s been in beta for awhile; it’s now being launched officially. I like Healia’s philosophy:

Generic search engines provide good results for many general topic areas, but they often provide questionable or “harmful” results for many health-related topics. This is especially true for health topics that are commonly the subject of scams and questionable activities. For example, most search engines produce links to questionable results promoting or selling unproven products and services in response to searches within the areas of alternative medicine, nutrition, diet and weight loss, prescription drugs, and cures. Healia’s innovative technology ensures you get high quality results from the most trusted sources on the Web.

One of the nice things about Healia is the ability to filter searches. Type in a search term like “diabetes” and Healia will give you the opportunity to filter the results by criteria such as male/female, kids/teens/seniors, and to look for content of different types, e.g., basic/advanced reading, easy to scan/fast loading/interactive tools. Then there are a number of tabs containing results by type, such as symptoms, diagnosis/tests, and treatment.

I tried a search on diabetes, which should be a good term since there is so much published about the topic and the filtering function should be helpful. I decided to filter for males and basic reading. Results weren’t good. I clicked the Symptoms tab and found the following:

  • The first listing was a dead link for Advocate Health. I’m not even sure from the description that the result was relevant.
  • The second listing was to the Polycystic Ovarian Syndrome Association discussion board, and contained a marginally relevant post related to gestational diabetes.

Moving on to the Diagnosis/Tests tab, results weren’t much better:

I did receive a warning that, “Selecting multiple filters may produce few results; deselect one or more filters to see more links.” But for a topic as broad as diabetes, that’s a fairly lame message. Plus the problem wasn’t the number of links but the quality of them. When I deselected the links I got more relevant results. Maybe that’s because there isn’t that much of a difference between men and women for diabetes, I don’t know. In any case it requires too much thinking and the results aren’t clearly better than Google’s.

I tried searching on Google for diabetes symptoms males and got better results. Google’s algorithm incorporated the “males” term but since it wasn’t terribly differentiating it didn’t throw off the results :

  • The first link was to MaleHealth: Diabetes, which talked about diabetes symptoms and had links to male health topics
  • There was a sponsored link to Diabetes Symptom Directory

I then went to Google Co-op, which I’ve maligned in the past, and typed in diabetes. It brought up a similar list of filters and topics as Healia, but it only lets you select one filter at a time (probably a good idea given the Healia experience). There’s also a long list of sponsored ads. I then selected symptoms and got relevant results and two relevant ads. Same deal when I selected tests/diagnosis.

I don’t doubt that there are other searches where Healia does a better job than Google. In particular, searches for topics like diet pills or Viagra –or any topic that generates a lot of spam– are likely to benefit from human moderation.

Unfortunately for Healia it will need to be not just at parity but clearly superior to Google to persuade people to use it. Considering that Google has a gaggle of unpaid consultants helping it improve its health care searches, that’s going to be tough. (You don’t see as many volunteers for Microsoft, helping them keep upstarts like Linux at bay!)

We really do need alternatives to Google for health care searches, and I hope Healia can hang in there. At a minimum, it will have to make the filters work better to win me over.

Time flies when you’re messaging your doctor

published date
September 12th, 2006 by

Time flies when you’re messaging your doctor

I received the following message from PatientSite, the secure messaging service my doctor uses:

Dear PatientSite Users:

We are sending out this email as a reminder to you about the Compose Message page having a 15 minute timeout. If you are composing a long email in PatientSite, please click the Save Draft button every 10 minutes to ensure that your text will be saved. An alternative to this method is to compose your email in a Word document. You can then copy (control + C) and Paste (control + P) all your text into your Compose window in PatientSite.

PatientSite times out after 15 minutes on the Compose Message page and the Records pages when there are no page flips, which represents no activity to the program.

PatientSite Tech Support

The message tell me two interesting things:

  • First, that PatientSite tech support doesn’t realize that Ctrl + P means “print” in Windows, not “paste,” which is Ctrl + V
  • More importantly, that PatientSite’s designers didn’t understand how patients actually interact with physicians online

The designers must have assumed 15 minutes would be plenty of time for a patient to compose a short note to their doctor. After all, unlike my favorite secure messaging platform, RelayHealth (now part of McKesson), PatientSite doesn’t let patients engage in structured messaging. It’s basically just secure email.

Most emails take only a few minutes to compose. But it’s different when patients are messaging their physicians. One of the interesting observations from early studies of RelayHealth users is that patients actually spent a lot of time composing their messages, yet the patients didn’t perceive that they were spending too much time. Instead of feeling rushed to present themselves intelligently in the few minutes or seconds of an office visit or phone call, patients would take the time to express themselves clearly, thinking through their messages and re-writing them as needed. After all, they weren’t wasting their physician’s precious time.

One of the reasons that asynchronous services like RelayHealth and PatientSite are so valuable is that patients can take all the time they want on their end to get their thoughts straight without making the physician (and other patients) wait around. The patient can even get help from a family member of friend. Once the physician receives the message he or she can view it and dash off a response in a few seconds or minutes, and everyone is happy.