“Are there any side effects I should know about?” asked the young lady.

“No, there aren’t,” answered the doctor.

It wasn’t the vaccine for influenza they were discussing. It was the vaccine being marketed by Merck under the trade name Gardasil. It is supposed to protect against the human papilloma virus, thought to be a common cause of cancer of the cervix.

I was hearing about this conversation after the fact. The patient was in her mid-teens, bright, well-educated, and well-read. The doctor was a family practitioner. Trying to be charitable, I told the young lady that the doctor was being paternalistic, although that is a behavior that should be passé among physicians.

She was saying there were no side effects that should dissuade the patient from accepting the vaccine, failing to understand that was the patient’s call to make, not the doctor’s. And she might have answered differently if the patient had been 10 years older.

From my perspective, having spent many years studying biomedical and professional ethics, I considered the doctor’s behavior to violate the ordering principle of medical ethics in Western societies: patient autonomy. The patient thought the doctor was just a liar.

Perhaps now that the patient is a college student, pre-med, who has learned a good bit about medical ethics already, she would be more charitable. But her reaction – which was to decide against accepting the vaccine and simultaneously to decide she would not be seeing that doctor again – illustrates the importance of honesty in the doctor-patient relationship.

That young patient was my daughter, and she knew that her very conservative lifestyle meant her risk of acquiring HPV was zero, that the risk of an adverse reaction to the vaccine was small but non-zero, and that therefore the risk-benefit analysis did not come down on the side of benefit.

For many years I have worked at hospitals that strongly encouraged employees, especially nurses, to get a flu shot every year.

The “pitch” was that it would protect the recipient from getting sick. Every time I read or heard the pitch, I bristled. They should be honest, I said. It is not at all clear that, for a young healthy person not in any high-risk group, the risk-benefit analysis favors benefit.

What is such a person’s chance of becoming seriously ill with seasonal flu? Very small. What about the risk of a serious adverse reaction to the vaccine? (Guillain-Barre syndrome, which causes paralysis, starting in the legs and moving up, sometimes temporary, sometimes not, is the most common serious reaction.) Also very small, but non-zero.

The risk-benefit analysis should be carefully considered by the person contemplating accepting the vaccine.

I thought the hospital management, employee health department, or infection control practitioner – whoever was making the pitch – should say the reason for being vaccinated was to protect our patients, to avoid acting as vectors of transmission of influenza from sick patients to those who were not sick with the flu but were susceptible.

As it turns out, the effectiveness of vaccinating health care workers as a way of protecting their patients is a matter of some controversy. It is intuitively appealing. It makes sense. From what we know about contagion and public health and health care associated infections, it just has to be true. But it’s not so easy to study, and the available evidence is not compelling.

And that brings us to the subject of this essay: mandatory vaccination.

The Centers for Disease Control and Prevention (CDC-P, or just CDC for short) currently recommends that everyone (past the age of 6 months) get the flu vaccine. I have to assume the rationale is that it will reduce the total societal burden of influenza illness during flu season.

But people don’t find that persuasive. They want to hear that it will keep them from getting sick. So that is the message. If the messengers are being responsible, they will include the caveat that the vaccine isn’t 100%, and it’s still important to pay attention to hygienic measures that reduce contagion.

But they aren’t as reserved in their recommendation as the facts suggest they should be. There are important limitations with this vaccine. The influenza virus is constantly changing. For those who have not studied virology, don’t worry, I’m not going to torture you with the details. But the proteins that make up the viruses mutate and rearrange, producing minor and major shifts, and occasionally new (“novel”) strains cross from other animal species to humans. The vaccine produced each year covers just a few strains, chosen based on a best guess of what strains will be in circulation. Those guesses aren’t always right.

With any vaccine, some people will be non-responders: the immune system just doesn’t make antibodies in response to the vaccine, at least not enough to be protective against illness.

And unfortunately the elderly, who are at greater risk of becoming seriously ill with the flu, are also at greater risk of being non-responders.

Finally, a substantial proportion of illness with flu-type symptoms – collectively called “influenza-like illness” –is caused by viruses other than the influenza virus. The vaccine affords no protection whatsoever against those. At the peak of flu season, about half of all influenza-like illness is caused by those other viruses.

So it’s easy to see why it’s not a certainty that vaccinating health care workers protects their patients and why it might be challenging to study it and try to prove it one way or the other.

At my hospital, employees are encouraged to get the vaccine. If they decline, they are asked why. If their reason doesn’t square with what is known about the vaccine – for example, some people are convinced you can get the flu from the vaccine, which simply isn’t true for the injectable version, because the virus has been inactivated – they are given information and asked to reconsider. But vaccination isn’t compulsory.

Recently ABC News reported on the firing of a nurse in Indiana who refused influenza vaccination. She thought she had a legitimate reason for refusing: She said her belief that she had a right to protect her body from this foreign substance could be likened to a religious belief, and religious exceptions were permitted. Some infection control specialists don’t believe in religious exceptions. They think people whose religious beliefs preclude vaccination shouldn’t be taking care of patients in a hospital.

Frankly, I don’t have an opinion on religious beliefs about vaccination. I don’t know enough about them to have an informed opinion, and I don’t think I should have uninformed opinions.

But this much I do know: the principle that vaccinating health care workers against influenza protects their patients is based on a consensus of expert opinion, not compelling scientific evidence. A consensus of expert opinion may be a sufficient basis for recommendations. It is not a sufficient basis for compulsory vaccination.

It is time to start being honest about the difference between what we think is a good idea and what is supported by a solid foundation of medical science. And it is time to start making policy decisions that may violate individual autonomy as though we understand that difference.

Dr. Solomon teaches emergency medicine to residents at Allegheny General Hospital in Pittsburgh and is Medical Editor in Chief of ACEP News. He is a social critic, political pundit, and blogs at www.bobsolomon.blogspot.com.