Doctors using social media to talk to patients, but where’s the evidence?

The patient-centered model of care is predicated among other things on physicians factoring in knowledge of the “person behind the patient” into their treatment. That’s means understanding and, where practical, honoring the patient’s beliefs, values and preferences. In order for a communication between a physician and person (patient) to be “patient centered,” it must be congruent with patient preferences for how they want their physician to communicate with them.

So just how “patient centered” is social media?
Let’s consider test result reporting to patients. If you are among the 5% of patients who (in very recent large-scale studies) indicate they want to receive normal test results by e-mail for example, e-mail results reporting is very patient-centered. Only 1% of patients prefer receiving abnormal test results via e-mail. Social media, e.g., e-mail, is not very patient-centered however if you among the other 95% of patients that prefer to be notified of normal and abnormal test results by telephone, snail mail, or in person visits with your doctor.

I understand that e-mail is not necessarily considered “social media” like Twitter, Facebook, or blogs, but it is the only “indicator” we have to date in the research literature. I also acknowledge that non-physician blogs and social networking sites such as PatientsLikeMe show great promise in building self care management skills, confidence and support among people with similar chronic disease conditions.

Doctors using social media to talk to patients, but wheres the evidence?

This is not to say that physicians should avoid social media when communicating with patients. I am just saying that, according to the evidence, social media is not for everyone at this point. No doubt patient preferences involving social media will evolve with the development of new applications and privacy protections . . . but we are nowhere near that point yet.

From my vantage point, when it comes to communicating with patients, physicians’ time would be much better spent by:

  1. Learning what their patient preferences are (with regards to communications, medications, exercise, nutrition, etc.).
  2. Tailoring conversations with patients during office visits to their preferences and concerns. The evidence shows that by doing, physician can more effectively engage patients, increase patient adherence, reduce cost and improve outcomes and satisfaction.

I have yet to see large scale studies that shows how social media can do that.

Steve Wilkins is a former hospital executive and consumer health behavior researcher who blogs at Mind The Gap.

Submit a guest post and be heard.

Comments are moderated before they are published. Please read the comment policy.

  • Vox Rusticus

    Sorry, but social media will remain fro me something for my personal time, just like other socializing activities. Uncompensated online correspondence really does not offer much value to my practice nor does it have the value of other means I use and need to correspond with my patients. And I pay no mind to those prognosticators who think and say that doctors that don’t adopt a practice style that involves Twitter and other social media outlets will be eclipsed by those that do. That’s bunk. Unless you can show that you are being paid as well for every minute you spend on your Blackberry as you are in your office seeing patients, then those assertions are not true on their face.

  • Stephen Ferrara, NP

    I applaud your insights and rationale with the various methods of communication between providers and patients. We must constantly evaluate the current technologies and see if they improve communication, increase quality and lead to better outcomes.

    However, I would hardly call the “large-scale” study you cited (available at: as hard “evidence.” This was merely a survey of 728 patients from a particular health system in Texas. The study did provide some interesting insights and highlighted these patient’s preferences. We must be extremely accurate and thoughtful when referring to evidence which must be both valid and reliable. This referenced survey is a good start to ascertaining patients preferences in obtaining lab result notification but lets not make the stretch to social media and think that it is necessarily generalizable to the rest of the population.

  • Steve Wilkins

    Stephen…the findings from the “single” study you questioned are consistent with findings from other studies (see below) that looked at consumer preferences for how they wish to receive health information:

    - Patient preferences for notification of normal laboratory test results: A report from the ASIPS Collaborative,

    - Edelman’s 2009 Health Engagement Barometer –

    Steve Wilkins
    Mind the Gap

  • Matt Mealiffe, M.D.

    Both the quantity and quality of evidence evaluating the impact of social media and older tools like email on morbidity, mortality, and other outcomes of interest (like patient satisfaction) are less than ideal. That said, I am much more bullish on the future impact that some of these technologies will have on health.

    Clearly, this is going to be a non-starter for physicians in many settings unless we find better ways to reimburse them for these types of interaction. Nevertheless, integrated systems like Kaiser Permanente have forged ahead with these things because their presence as both payer and provider with relatively less migration of patients out of the system allows them to utilize these tools to their maximum effect. I think we will see some very interesting data come from their experiences with these tools.

    Sure, they’re not for everyone, but recent data show that familiarity with and frequent use of the internet is increasing substantially even for seniors. Where it’s feasible, we’d be foolish not to take advantage of new media and social media to optimally care for the population.

  • Stephen Ferrara, NP


    I appreciate the additional references. It was difficult to find the evidence you were referring to in your post.

    Nonetheless, the evidence remains insufficient to draw any real conclusions (with the additional survey consisting of a sample size of 20 and was conducted over 5 years ago and I’m not even sure what level of evidence I would give for the Edelman Report).

    Perhaps, I’m just sensitive to the word evidence as I always attempt to approach things from an evidence-based approach and seek the gold standard of studies: the randomized, double-blind type and systematic reviews.

    Thanks for the discussion.

  • Steve Krizman

    As of the end of 2009, there were 3 million Kaiser Permanente patients registered to use secure online features. In 2009 alone, there were 13 million patient views of test results and 8 million e-mails sent to doctors. Kaiser Permanente has over 8 million patients, so these numbers indicate that when online access to medical information is available, it is quite popular among patients.

    Perhaps a reimbursement system that is based on outcomes rather than on office visits would eliminate the financial barriers to patient-centered communication.

  • Steve Wilkins

    Good points. Note however that the studies cited above simply examined patient/consumer “preferences” for how they wished to be communicated with by their physician – in this care regarding test results. Since no interventions were actually tested and no claims made about efficacy, a double-blinded, controlled study was not necessary.

    The Edelman study I cited earlier consisted of a sample of randomly-selected, representative sample of 1,000 US residents. Admittedly, findings from the Texas study can be generalized only to the population of the health plan studied.

    The point remains the same. Just as it is important for providers to understand a patient’s health literacy or interest in receiving health information, it is equally important for providers (and health marketers) to understand how consumer/patients wish to receive such health information.

    Additionally, physicians (and patients) would be much better served if they used their time to address the well-documented challenges of face-to-face communications with the patients before they get distracted with Twitter or Facebook.

  • Ted Eytan

    Hi Steve,

    I’d to add to my colleague Steve Krizman’s comments above, and a helpful critique of the studies you cited.

    The ASIPS Collaborative study was performed in 2005, which is nearly an eternity in Internet time, and on my review, clearly was not done in the background of the secure My health manager personal health record that Kaiser Permanente operates in Colorado, among other places. The preference for telephone was perceived based on lack of a secure/private alternative, which now exists.

    The Edelman Trust Barometer is great, I don’t see where it says that patients don’t want to be connected to their doctors.

    As Steve Krizman, and the data that I’ve posted on my blog shows ( see This post for the most recent data), when patients actually have access to these services, and they are connected to their physician and care team, they are perceived as incredibly useful, and 13,000,000 results viewed on line is pretty large scale.

    If you want to reference peer-reviewed literature about the Kaiser Permanente experience, you can find it here.

    I found your post because of the discussion about evidence around social media, and I agree with you that, today, we just don’t know. However, this doesn’t mean we shouldn’t find out, maybe we shoudn’t comingle that discussion with the one about patients accessing their health information online.

    Should you need any additional data about the use of online services by patients when they were offered, let us know, we’re happy to help / inform.


    Ted Eytan, MD

  • Steve Wilkins

    Drs. Steve and Ted,

    Thanks for your comments. For the record, I never said that patients do not like or want to communicate with their physicians by e-mail. What I said and what the data I referenced showed is that not everyone shares the same preferences for how they wish to communicate or have their doctor communicate with them. Given the dearth of research on the subject, I used e-mail and test results reporting as a “straw man” to demonstrate this point.

    Let’s say I prefer getting abnormal test results in a face to face visit with my physician. Let’s say it’s more convenient (and efficient) for my doctor to e-mail me tests results – whether normal or abnormal..and in fact does so. By not respecting and honoring my stated desires, I as the patient am inclined to react in the following way:

    1) Feel like my opinion does not matter
    2) Be less inclined to share other information that my physician is not likely to remember or follow-up with.

    Paying attention to patient preferences are part of what patient-centered philosophy is all about. I am sure that there are a lot of Kaiser patients that appreciate having the ability to communicate with the physician by e-mail. However according to your own data, a fair number of them still do not avail themselves of this communication tool. Others may prefer to call the Kaiser Advice Nurse. The point is that people have preferences and are free to exercise their own preferences. I rue the day when technological determinism dictates how we must communicate with our physician.

    I appreciate your personal enthusiasm for the technology and your role as evangelists for the Kaiser. I hope you are just as enthusiastic when it comes to protecting the patient’s right to have their own communication preferences.

    Steve Wilkins

  • Ted Eytan

    Hi Steve,

    I understand the concern that you express, that the web, as an additional channel, might reduce access to the other channels for various reasons. It’s a realistic worry, and one we considered carefully 10 years ago when starting this work. I can assure you after almost 10 years of doing this, this is not happening; the opposite is. Our members have far better access to their physicians than their peers in other systems do.

    All right, thank you for allowing us to add to and fill in to the post you created, social media style. Patient preferences – very important. Multiple options to get needed care – very important too.

    And lastly….Steve, I enjoyed your wording in the last paragraph of your response; however, it seems a bit ad hominem to me and out of place for a collegial discussion. I don’t question your commitment to patients and my record reflects the same.

    Be well,


Most Popular