Some good advice I've given my children is to never put anything on the Internet they wouldn't put on a billboard. That advice applies to all Minnesotans, since our most private information is about to be made available on the Internet: our medical data. Minnesota is one of the first two states to mandate that all health care providers use an "interoperable" electronic health care record (EHR), allowing all doctors to view every other doctor's notes — supposedly privately and with a purpose. Since 2014 has been dubbed "The Year of the Hack," we can presume this medical data will be invaded.

It is including psychological records that concerns me most. With decades of experience as a clinical psychologist, a consultant and a teacher, I have to believe this mandate can easily mean the end of psychotherapy as a useful treatment.

Perhaps no other form of medical treatment is more based on privacy and confidentiality than psychotherapy. Patients come to speak the unspeakable. Much like the sacrament of confession, psychotherapy relies on a closed door and a belief that what is spoken stays behind that door. It's why we don't practice psychotherapy in public settings. Once notes are posted, any suggestion that our treatment is private is untrue.

What do people tell us behind our closed doors? About the affair they are having, or had, or thought about having. About abuse and tremendous vulnerabilities that may be deeply shameful. Even with privacy, it takes time to build trust enough to speak what has never been spoken before — the events that drive their symptoms, the anxiety and depression. For relief, patients expose themselves in tremendously vulnerable ways.

Proponents of the law suggest that we can limit what we post by keeping a separate record that details more delicate information. If being a forensic consultant has taught me anything, it's that there are no private medical records. For psychotherapists, posting any information about patients is wrong, because it violates our duty to protect privacy in this most vulnerable setting. As the U.S. Supreme Court's decision in Jaffee vs. Redmond said that without confidentiality, there cannot be psychotherapy. There aren't crises requiring immediate access to psychotherapy records; they can be accessed more securely, in plenty of time if needed.

The law as written apparently allows no one to opt out of the EHR. Anyone who wants out must opt out completely. No health care provider can opt out. Patients who pay cash for their medical care may be able to avoid having their records posted, circumventing a loss of confidentiality, but not most of us.

Imagine this: You're having sexual or marital problems? You had an abortion? You're depressed and feel it's a personal failing to need treatment? You're being treated for the effects of childhood sexual abuse? It's all available on the Internet — and to every other medical professional you may see.

There are two important ways this law must be shaped:

First, we should exclude any amount of a mental health record because of the catastrophic effect to psychotherapy.

Second, we must allow individual patients to decide what they do and do not want posted to their electronic record. If we don't want a particularly sensitive medical concern in the EHR, we should have the right to exclude it. If a physician believes it is important to include, let them counsel us — just as they would about a medicine we may not want to use.

My own patients range from the powerful and affluent to the middle class to those on Social Security, Medicare and public assistance. All of them deserve privacy at my office. Let's pause before throwing caution to the wind and irretrievably losing access to confidential and meaningful mental health care.

Talk to your legislators and the Minnesota Department of Health before it is too late. The rules for EHR are still being formed. Now is when we can be national leaders in forging a system in which the sharing of health records is both meaningful and reasonable.

Peter A. Zelles, of St. Paul, is a clinical psychologist.