Medical report cards will damage patient privacy

by Kip Sullivan
published Oct 6, 08

Last May Governor Tim Pawlenty signed a law authored by Sen. Linda Berglin that requires the Minnesota Department of Health (MDH) to publish medical report cards. The new law authorizes MDH to collect the medical records of all 5 million Minnesotans and use them to publish report cards on all 17,000 doctors and all 131 hospitals in Minnesota.

In an editorial published in the September edition of Southside Pride, Sen. Berglin denied that her new law damages patient privacy. She claimed that giving MDH authority to commandeer patient medical data is not a big deal because the insurance industry is already doing it. Specifically, she argued that Minnesota Community Measurement, a group founded by the state’s largest health insurance companies in 2001, has been collecting patient data for years and using it to publish medical report cards on their website. Therefore, she argued, there’s nothing wrong with letting the State of Minnesota do the same thing.

This argument is not merely unconvincing. It raises the question, Why has Sen. Berglin done nothing to stop this abuse of patient privacy by the insurance industry?

Having made the argument that a wrong committed by Blue Cross Blue Shield justifies another wrong committed by the government, Sen. Berglin then asserted that clinics and hospitals submitting data to MDH “must strictly adhere to all federal and state data privacy laws.” This is true but irrelevant. Those laws require that, in the course of submitting medical records to MDH, clinics and hospitals take steps to minimize the chances that unauthorized third parties will see them. Even if we make the wildly unrealistic assumption that compliance with these laws is 100 percent and no one ever loses computerized data through theft or negligence, those laws do not change the fact that MDH will still be getting its hands on the medical records of millions of Minnesotans.

These records will be linked to particular patients, if not by attaching the patient’s name to the record, then by a patient-identification code number (and MDH, of course, will hold the code). There are several reasons why MDH will need information that links medical records data to particular patients. One reason is that without patient-identifiable data MDH will have no way of “assigning” patients to particular clinics or hospitals. Many patients visit multiple health care providers during the course of a year. If John Doe visits Clinic A four times and Clinic B twice during the course of a year, how will MDH know which of these providers should get how much of the credit (or the blame) for Mr. Doe’s treatment if it can’t link data from both clinics to Mr. Doe?

A second reason why MDH will need to get patient-identifiable data is that Sen. Berglin’s new law requires that MDH take into account differences in patients, including differences in their health, when it grades clinics and hospitals. This process of adjusting scores to reflect differences in patients is known as “risk adjustment.” The need for risk adjustment is obvious. Doctors who treat sick and poor patients will tend to get lower grades than doctors who treat healthy upper-income patients if MDH fails to take differences in patient health and income into account.

(In some but not all cases clinics that treat sicker and poorer patients can push their grades up if they spend more time and money on those patients. But that will only mean that patients who go to those clinics whose care is not being graded will suffer, and the net effect will be the same—clinics with sicker and poorer patients will suffer from report cards that are not risk-adjusted.)

Take for example a report card that purports to measure the quality of diabetes care by measuring the percent of a clinic’s diabetics who have blood pressure under 130/80. It is well established that age is a major risk factor for high blood pressure (the condition is uncommon in kids, but two-thirds of Americans over 60 have it). It is also well established that people who have low incomes, or have insurance with drug co-payments or no insurance at all, are less likely to comply with their doctor’s recommendations to take prescription drugs, including drugs that lower blood pressure. Thus, if MDH publishes a report card that uses percent-of-diabetics-with-blood-pressure-under-control as the “quality” measure, and MDH fails to take into account patient age, income and insurance status, the report card will be inaccurate. It will measure differences in patients, not differences in the skill levels of doctors and nurses. Obviously, MDH cannot take these patient-related factors into account if it is not collecting information on those factors for particular patients.

So, to repeat, for at least two reasons, MDH is going to need to know the identities of all patients it collects data on: to “assign” patients to particular clinics and hospitals, and to risk-adjust grades.

In addition to having their medical records turned over to MDH, some Minnesotans will have their medical privacy violated three other ways: by MDH audits, by clinics and hospitals hiring outside “vendors” to prepare their reports for MDH, and by loss of medical records data by providers, vendors and MDH.

MDH will have to audit some of the data it gets from providers. (Not auditing at least some data would be the equivalent of just letting doctors and hospitals mail in their own grades.) Audits require MDH employees or vendors to visit clinics and hospitals and personally inspect patient medical records to see if there are real records corresponding to data MDH received from those providers, and to see if those records say what the providers claimed they said.

Minnesota Community Measurement uses vendors to do some of its audits. Its August 2007 online newsletter, Measurement Minute (available at mnhealthcare.org), announced that “chart review” is being done by a private firm called Q Mark. (“Chart” is another word for “medical record.”) “Acting as our agent, Q Mark, a chart review vendor, will be conducting chart review,” said the newsletter. “Q Mark will be calling medical groups to schedule chart reviews for August or September.” According to Q Mark’s website, “Services include … medical record pursuit….” BusinessWeek describes Q Mark this way: “Its products and services include … collecting information from medical records ...”
Many clinics and hospitals will find that it is cheaper to hire an outside company like Q Mark to respond to MDH’s demands for medical records. This means, obviously, that instead of your doctor or nurse going through your medical record to find the data MDH wants, an employee of Q Mark will do it.

Sen. Berglin asserted in her editorial that focus groups conducted by Minnesota Community Measurement indicated “consumers do want” report cards. In fact, according to that same August 2007 edition of Measurement Minute, the focus groups said just the opposite. The newsletter reported that 28 diabetics who participated in the focus groups “were suspicious of statistics in general and the motivation of groups that publish statistics specifically.” This stinging assessment of report cards, plus polls indicating massive majorities of Americans oppose government agencies seizing their medical records, indicate Sen. Berglin has little popular support for her report card project.