Health information is much more than Wikipedia and WebMD, much more than The Lancet and Gray’s Anatomy. Health information also refers to patients’ health records – both individually and in aggregation.

Incredibly, according to a report published last month in the New England Journal of Medicine, fewer than one in five of doctors in the United States has started using electronic health records. That means every time you visit your doctor he or she probably still uses good ol’ pen and paper to make notes about and track your health history. When you change doctors, or move to another city, that folder of papers, x-rays, MRI scans, and other information related to your health history get put in the mail and are physically transferred from one clinic’s filing cabinet to another. Yet despite the fact that this manila folder will follow the patient around for most of her life, the patient herself will probably never see it. Which, according to William Tierney, Senior Research Scientist at the Regenstrief Institute and Director of Research at the IU-Kenya Partnership, is exactly how it should be.

A practicing doctor of internal medicine, Tierney says that health practitioners often use patients’ health records as a way to remember their thoughts from visit to visit. In the United States, for example, the average doctor has 390 patients at any one time. (Interesting aside: Cuba has the lowest patient:doctor ratio worldwide at 170:1.) There is no way that any doctor can remember every visit from each of his or her 390 patients. So the patient’s health record is a place for the doctor to take notes about the patient’s health, but could also be used to actually describe the patient’s personality and how he or she is best dealt with. If a patient has access to his own health record, the doctor will be more likely to withhold information, and that might negatively influence how the patient is cared for.

Visualization of patient to doctor ratios worldwide

Tierney also emphasizes that patients do not understand the information that is on their health record and that, if patients were given access, doctors would have to spend more time educating their patients about how to interpret data than actually caring for them.

According to a recent survey by Deloitte, however, 75% of patients want their doctors to provide online access to an integrated medical record and 25% would pay more for the service. John Halamka, Harvard Medical School’s Dean for Technology and a frequent blogger about health IT, says that “patients will ultimately be the stewards of their own information.” There is no greater example of this trend than Google’s entry into the health record market.

Eric Schmidt giving the keynote speech at the Healthcare Information and Management Systems Society Annual Conference in Orlando, where he introduced Google Health.

According to Marissa Mayer’s introductory post on the official Google blog, Google Health will allow users to “collect, store, and manage their own medical records online.” An impressive advisory council of doctors and health experts is helping Google negotiate the privacy and ethical concerns that are related to the storage and access of health care records. As the introductory tour explains, Google Health will allow internet users to start filling in information about their own medical history and then learn more information about the conditions and symptoms they enter. But, most importantly, it also allows you to import your health record and prescription history from hospitals, labs, and pharmacies that you have visited in the past.

As of May 20, “more than two dozen companies and institutions announced that they are partners with Google Health, including Walgreens, CVS, the American Heart Association, Quest Diagnostics, Beth Israel Deaconess Medical Center and the Cleveland Clinic.” John D. Halamka, who is also a member of the Google Health Advisory Council, describes the specifics of how Google Health is being implemented at Beth Israel Deaconess Medical Center.

Google Health isn’t without its skeptics and critics. Most complaints center around security concerns and the privacy of health information. ha.ckers.org, a blog focused on network security, cites Google’s past security vulnerabilities and that, unlike many other health record systems, is not a “covered entity” under the Health Insurance Portability and Accountability Act. Most worrying to the author is how common users’ passwords for Google’s services are either stolen or guessed.

Over!My!Med!Body!, the blog of a recently graduated medical student, lays out a long argument about why patients should not control their own medical records from the perspective of a doctor. Some of his points include:

• Patients don’t know, use, or understand medical terminology.
• Doctors will trust patient-controlled health records less than doctor-controlled health records.
• What happens when patients pick and choose what data they want to import into their record?

The negative aspects of Google Health from the perspective of a medical doctor providing treatment become apparent. But the staggering potential of Google Health from a public health and epidemiology perspective, also becomes readily apparent.

HealthMap is a project led by Children’s Hospital Informatics Program and the Harvard-MIT Division of Health Sciences & Technology, with support and funding from Google.org’s Predict and Prevent initiative. As reported in Wired, the website monitors content from Google News, the World Health Organization, and online discussion groups, and maps the latest health-related stories worldwide. Results can be filtered by news source, disease, and country. The Wired article and Patrick Philippe Meier describe how that data mapping can be used to track new disease outbreaks.

Now imagine that the same map all of a sudden has access to the most recent symptoms which are entered into Google Health both by patients sitting at home and by doctors treating them in clinics. In real time public health officials will see where similar symptoms are reported and how doctors are diagnosing those symptoms. Doctors will also have access to the information, allowing them to connect to colleagues and share observations about what they are seeing. Not only will disease outbreaks be mapped, but how they travel from place to place will be better understood. When combined with demographic information such as age and ethnicity, this too can help researchers better understand the nature of the disease they are studying. It also, of course, brings up more ethical issues about privacy control and what information is shared with whom.

The question at this conference is, how applicable is the Google Health model in the developing world? In terms of patients’ control of their own medical records, this is already the established norm throughout most of Sub-Saharan Africa. Most Kenyans, for example, carry their entire life’s health record with them in a single small blue book (similar to those used in college exams in the United States). They bring that blue book with their health record to the hospital, it is then updated by the doctor and/or nurse, and given back to the patient to take care of until his or her next visit.

It is interesting to note that Tierney, Director of Research at the IU-Kenya Partnership, is less worried about patient control of health records in Kenya than in the United States because of cultural differences. “In Kenya,” he says, “the doctor is god and you don’t question what he tells you. Most Kenyans do not look at their health records, they just take care of them.” In the United States, on the other hand, Tierney says many patients would scrutinize and even argue with their doctors about what is put into their health records, which would take away from the time the doctor could spend treating patients.

The main obstacle to implementing Google Health in Kenya then has to do with internet access. No connection, no health record. Partnerships would also need to be established with Kenyan hospitals and pharmacies and the user interface would have to be localized into the many languages of Kenya. Furthermore, information about symptoms and diagnoses would have to be mapped across languages, while keeping in mind cultural contexts of health, disease, and treatment. There is also the issue of reading and computer literacy.

In the next post I will focus on health information technologies for regions without reliable internet access.