Without evidence of benefit, an intervention should not be presumed to be beneficial or safe.

- Rogue Medic

Blogging and Privacy and Competition for Dr. Deborah Peel

J Gen Intern Med. 2008 Jul 23. [Epub ahead of print]Click here to read Links
Content of Weblogs Written by Health Professionals.
They make it seem as if you get the free full text, but you don’t. On the other hand, you are not really missing anything.

The research cited in this article is pretty poor. They start by assuming that they can identify most of the bloggers, but do not demonstrate that. Here is their big chart of proof:

Blogs n (%)

Reviewer agreement kappa

Blog author identification

First name

139 (51.3)

0.8

Last name

95 (35.1)

0.9

Location

117 (43.2)

0.7

Identifiable photo

60 (22.1)

0.6

Sub-specialty

197 (72.7)

0.6

Patient privacy

Patient images

3 (1.1)

0.6

Patient radiographs

8 (3.0)

0.7

Patient laboratory studies

0 (0.0)

*

Other privacy information

1 (0.4)

*

Patient depictions

Any comments about patients

114 (42.1)

0.8

Negative comments about patients

48 (17.7)

0.7

Positive comments about patients

43 (15.9)

0.5

Commentary on health profession/heath care system

Comments about the health care system

137 (50.6)

0.5

Positive comments about health care profession

108 (39.9)

0.4

Negative comments about health care profession

86 (31.7)

0.6

Product endorsement

Blog posts promote health care products

31 (11.4)

0.6

What do they agree on the least? “Positive comments about health care profession.”

They make a lot of unreasonable assumptions about being able to identify the person writing the blog, based on the initial assumption, they look at the information presented about the “patients” mentioned in the blogs, presuming all of the information to be correct, they jump to the conclusion that there are HIPAA violations.

Over half of the 271 blogs had identifiable authors; 89 (32.8%) authors provided both first and last name and 43 (15.9%) gave enough information about their name, subspecialty, or location to be identifiable. A recognizable photo was displayed by 22% of authors. Of sampled blogs, 206 (76.0%) were written by physicians, including 23 interns and residents, and 59 had nurse authors (21.8%). The remaining six blogs (1.8%) were written by groups of health professionals including both doctors and nurses.

The fixation on HIPAA also ignores the patient’s right to privacy. Patient privacy did not begin with HIPAA. Dr. Deborah Peel and these clueless researchers do nothing to improve patient privacy. Claiming that you are accomplishing something and actually doing it are not at all connected.

There is much good that may come out of blogs. That is one of the reasons I read and write blogs. I can’t guarantee that no medical blogger is inappropriate with privacy information in a blog, in a book, in a radio interview, or at a cocktail party. Condemning medical blogs will not change any of that.

These researchers looked at only 19% of the medical blogs that they found. They cherry picked the ones they thought most likely to have violations of privacy issues. Out of those 19%, they found 17% that they claim had privacy issues.

This is only 3% of the medical blogs they found. But wait, where do they identify any actual patients based on the information in the blogs?

They do not. They mention the case of a doctor sued for malpractice, who blogged about the experience. The lawyer for the plaintiff recognized the information as being about his client and apparently used that to get a large settlement. This is not somebody with no knowledge of the patient, but someone intimately aware of the medical details of the patient’s medical records. Then you add the unusual identifier of the details of the malpractice case and it is not at all surprising that the lawyer was able to recognize the patient. This is the only case, of a patient being identified, that they mention. They draw some conclusions below, but do not demonstrate that they were actually able to identify anyone.

Forty-five blogs (16.6%) that described interactions with individual patients also included significant identifying information about the blog author. Three blogs showed recognizable photographic images of patients, and one provided an extensive description and links to pictures of a patient. Eight blogs showed patient radiographs without further identifiable information.

Yeah, that X ray, that looks about the same as every other X ray of that part of the body – that one is mine؟

They do not mention if these patients gave permission for their information to be used on these blogs. This is something that is relevant to the topic. Is this information able to be used to identify a patient? If so, did the patient give permission?

RM – Mr. Rectal Bleed, I’m going to make you a star.

Mr. RB – A star! Wow, wait until I tell the little hemorrhoids. What do I have to do?

RM – Just sign this standard disclosure form and smile for the camera. I’ll try to get your good side.

Mr. RB – Who would have thought that anything good would have come out of my sphincter?

There is also some concern that a patient might recognize him/herself in the description of a patient encounter posted on a blog. This is possible. The patient may be the person most familiar with their medical history. The patients described on blogs may be described with the permission of the patient, may be composites, or may be completely fictitious – used to demonstrate a particular concept (ethical, economic, political, educational, . . .).

A patient may read a blog and conclude that he/she is the patient described in a blog. What if the patient knows that the blog is by their doctor and they have a rare disease that is not the doctor’s specialty? Then it is likely that the patient has found information about their particular case. The doctor should have received some permission to write about the patient, first. What about the doctor who specializes in rare diseases? It is not unlikely that a doctor, specializing in some rare disease, has more than one patient with a similar presentation.

What if a patient believes that they have located personal information about themselves on a blog?

The person probably needs the services of a psychiatrist, rather than a lawyer.

Why do they believe that this is about them and not some of the hundreds, or thousands, of similar patients?

Why do they believe that the information presented is not a composite of several, or even dozens, of patients?

It’s not all about you!

Unless you are reading EE’s blog and call 911 for any one of many extremely stupid reasons. Then you need to contact me, because I certainly cannot identify you from the information she puts in her blog, you moron, and I will attempt to engage in some behavior modification for all of the system abuse you are guilty of. Of course, these are people who are having their stupidity violated – if you can ever identify them. This has nothing to do with medical privacy.

Another thing that they criticize is doctors making negative comments about patients, other doctors, or the profession of medicine. It is unprofessional in their opinion. Apparently all problems should be addressed with a false smile and more respect for social graces than for improving health care. Augh!

As far as product endorsement is concerned, they highlighted one blog entry. A response by the author of Medicine and Man, a seemingly respectable and not controversial blog, who points out that he was misrepresented and only wrote about an article in the New England Journal of Medicine. His response is On “Content of weblogs written by health professionals”. According to them I would be endorsing a lot of products, since I have written about many products described in peer reviewed journals. The post in Medicine and Man was not an entirely positive post, so their claim that it was advertising, rather than a literature review, is quite a bit exaggerated.

In the discussion they are a bit more balanced, but there are so many conclusions drawn from so little data – data that has been intentionally made misleading – that we should ignore the scary headline numbers completely. The discussion was only hinted at in the article. There are some people, who have probably been a bit careless with patient identifying information. I think that it is unlikely that they will avoid getting pointers from other bloggers about appropriate posting of patient interactions. Making this out to be more than that is just being silly.

Maybe Dr. Deborah Peel is using a pseudonym to publish her HIPAA hysteria.

Below is what I posted in the comments to the LA Times article. Imagine, limiting me to 650 characters. The nerve.

17% blogs included enough information to identify patients. They located 1,434 medical blogs, but narrowed it down to 271 “journal style” blogs. This cherry picking means that 17% of medical blogs is inaccurate. They only looked at 19% of the medical blogs they located, but selected the ones they felt would be most likely to be indiscreet. If they were good at cherry picking: 3%. They do not mention being able to identify any patients. They mention one case reported in the news. 1 case out of over 1,000 blogs! If only actual doctors, taking care of real patients, had such a good record. Rogue Medic http://roguemedic.blogspot.com/
Submitted by: rogue medic
10:55 PM PDT, August 5, 2008

This is also written about at On medical blogging codes on
DB’s Medical Rants.

.

Comments

  1. You assume that the call I write of is true? Or that I even went on the call. Maybe it was a friend who did. Maybe it was a story that an instructor told in class once. Maybe I’m not even a medic. Maybe I’m just a wanna-be. Maybe.

  2. EE,I was writing about Dr. Deborah Peel. I was using your identity as a pseudonym, to protect her innocence, or naivete, or cluelessness. :-)Send ’em anyway. I promise not to play nice.

  3. Nice write up. A lot of research papers are published every year but some of the them do not live up to the the scientific standard of the journal they are published in. I was surprised that this article was actually accepted in the Journal of General Internal Medicine without any corrections from the reviewers.

  4. Dr. Bansal,Thank you.One of the topics that gets me riled up is poor risk management. Another is poor science. They see something that concerns them, rather than look at all of the information, they look for things that they think confirm their bias. Not a good way to come up with policy recommendations, which seems to be their goal. They did make some comments about the limitations of their research, but they missed a lot. One was the way they completely misrepresented your post.

Speak Your Mind