How would you like your own, personal, prostate cancer management dashboard on line?


A fascinating paper recently published on line in the Journal of the American Medical Informatics Association addresses issues related to the development and use of personalized dashboards conveying patient outcomes-reported data in prostate cancer care over time.

This paper by Hartzler et al. examines both the design and the feasibility of  online, personalized, patient-oriented dashboards as a way to help prostate cancer patients track information about their outcomes over time and compare their outcomes to those of other, similar patients. The proposed dashboards would be accessible by patients themselves and their healthcare providers and therefore needed to be able to meet the needs of both sets of potential users.

It would be impossible to get into all the details of this paper in a brief commentary like this, but where Hartzler et al.’s work differs from that of others exploring this field is that rather than focusing primarily on issues related to data capture and recording, Hartzler and her colleagues have also made a major, initial attempt to explore the best ways to output the data collected so that they are easily understood by the users.

The paper describes three elements to the research conducted:

  • An initial process of capturing the informational and data capture needs that would be informative for both the patients and their providers (through a series of focus groups and interviews)
  • The iterative creation and refinement of a prototype dashboard designed to play back information in ways that could be tested with respect to the formatting of information
  • Pilot testing of the refined dashboard to explore whether it helped to improve patient complieance with quality indicators for prostate cancer care over time.

Here are just some of the findings of this study as reported by Hartzler et al.:

  • In focus groups and interviews, patients had a high level of interest in being able to view their health-related quality of life (HRQoL) outcome data and share it with their providers (even though only two of the participants had prior knowledge about patient-reported outcomes data).
  • Both patients and providers generally preferred data on the dashboards to be presented as bar charts and line graphs as opposed to pictographs, but …
  • Patient advisors encouraged the developers to give users a range of options regarding exactly how data was presented to them so that the users would feel they had some degree of “ownership” of how they could review their data (“tailored design”).
  • In pilot prototype testing. both patients and providers found the prototype dashboard to be acceptable, and specifically that dashboard use led to:
    • Higher quality of discussions between patient and provider
    • Greater adherence to quality indicators in prostate cancer care

The “New” Prostate Cancer InfoLink has long believed that, when one looks at prostate cancer management as a process over time, in which initial treatment and high quality of recovery of function after such treatment is the core process for the majority of patients, the use of this type of on-line, information-rich dashboard system has a great deal of potential utility in helping patients to monitor their progress over time with respect to sexual and urinary tract function, as well as related issues such as frequency of Kegel exercises and quality of erections.

We sincerely hope that Hartzler and her colleagues will be able to make further progress with this research such that this type of on-line tool could become widely available for use by the patient and the provider community in the near future.

8 Responses

  1. Thanks once again for providing this interesting review.

    The closest thing I know to this is the outcome data provided in the continually updated freedom-from-recurrence data set managed by the Prostate Cancer Results Study Group. A critically important feature of that data set was the use of criteria that put surgery, radiation and other results on an apples-to-apples basis, which greatly reduced “clutter” from studies that did not meet the criteria (such as comparable Gleason score timing (clinical, instead of clinical for radiation vs. pathological for surgery), minimum number of patients, minimum follow-up, etc.). A disadvantage of that approach, in the fast-moving world of prostate cancer technology, is that some emerging technologies, such as TomoTherapy and more recently the awkwardly named stereotactic body radiation therapy (SBRT) tend not to be covered for the early years of their emergence. Also, results with older technology, such as EBRT with doses now known to be too low, tend to somewhat clutter the graphs. Moreover, side effect data, which is very important to patients, is addressed in narrative form rather than in the graphs.

    I’m glad the authors see the importance of tailoring the presentation to the patient. It’s very clear that results are best communicated in different ways to different patients.

  2. I, for one, would like it. I would feel more directly “involved” with my current status and outcome expectations.

  3. Dear Jim:

    Respectfully, the dashboard approach proposed by these authors isn’t anything like the data set provided by the Prostate Cancer Results Study Group.

    The approach proposed by Hartzler et al. is highly individualized and primarily dependent on data actually being input by patients on a regular basis and associated with their quality of life. By comparison the approach taken by the Prostate Cancer Results Study Group is simply a rather scientifically unsound meta-analysis based on highly selected data designed to “prove” a preconceived outcome and has nothing whatsoever of a personalized nature about it. The fact that it uses some unusual pictures to illustrate the results claimed is meaningless because the base data are questionable.

  4. I went to the Prostate Cancer Results Group and agree with the Sitemaster. The analysis and interpretation of findings ignored the many complexities of evaluating treatment outcomes in the absence of controlled trials. Despite the impressive panel of MDs, the conclusions were simplistic and seemed oddly biased (e.g., toward brachytherapy).

  5. I love the concept. We saw in a recent article what a powerful tool it can be to have patients all fill in a validated QOL questionnaires (in that case EPIC). Getting enough patients and doctors on board, and building the database so that it’s personalizable seem like a formidable tasks. I hope they can pull it off.

  6. Brian,

    Did you look at each of the criteria used by the Prostate Cancer Results Study Group? Do you see why each one is important?

  7. Dear Jim:

    If you want to get people to believe that all women with red hair are beautiful according to whatever standards meet your particular fancy, you go out and select pictures of dozens of beautiful women with red hair who meet your criteria and then you put up a web site called “All Redheads Are Beautiful” with pictures of all these women and keep stating how beautiful all women with red hair are. Of course this doesn’t prove the point you are trying to make (and it also doesn’t mean that all women with red hair are beautiful either) … but lots of people will start to agree with you.

    Similarly, if you put up a web site that says “All Brachytherapy is Wonderful” and select data that is designed to confirm that (regardless of the absence of any well-established data from clinical trials), then you may well get a lot of people to believe that too. You seem to be one of these people.

  8. Brian,

    The “impressive panel” is misleading as well. I asked one of the ROs listed what he did on the panel. He said that they asked if they could cite a study he did and use his name. They make it sound like it’s a review board that meets and discusses each study they decide to put in.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: