Dr. Jason Cafer, MD

Medication IconoGraphs are a patent pending set of visualization tools to assist mental health providers manage complex medication regimens. IconoGraphs display a patient’s medication list to the prescriber in a novel way, conveying a wealth of clinically relevant information in a data-dense, easily interpretable format.

A set of IconoGraphs creates an enriched medication list on a graphical user interface (GUI) for rapid interpretation by the prescriber.  For each medication in the patient’s regimen, an IconoGraph displays the pill image, relative dosage, interactions, duration of treatment, treatment compliance, patient-specific efficacy and side effects, and general tendency to cause weight gain or sedation.

Medication IconoGraphs address several significant problems in management of complex medication regimens.  In present clinical practice, in paper and electronic patient records, medication lists are presented to the provider in text format with very little detail or contextual information.  Prescribers’ incomplete understanding of this contextual information may lead to medication errors.  Existing tools to prevent medication errors have intrinsic limitations that Medication IconoGraphs have the potential to overcome.  Beyond preventing errors, IconoGraphs have clear potential to improve treatment outcomes and reduce costs of psychopharmacotherapy.

Figure 1 shows how Medication IconoGraphs represent the same two medications as it relates to a specific patient. One IconoGraph is displayed for each medication on a patient’s list.  Labeled are the types of information not displayed to providers by traditional medication lists.

Iconic Health, LLC is a Missouri-based company formed in 2005 (as IconoPsych Care, LLC) by three psychiatrists and a psychologist to improve the quality and efficiency of mental health care through innovations in provider-patient communication and novel data visualization. Their work has produced 5 pending patents.

Iconic Health is funded by Centennial Investors in Columbia, MO.  Anthony Allen & Quinn d.b.a.The Innovation Factory from St. Louis, Missouri owns an equity stake in Iconic Health in exchange for business development services.  Anthony Allen & Quinn’s original venture sold to WebMD for $260 million.

IconoPsych (founders’ company) received a $360,000 Phase II SBIR grant from the USDA for developing and marketing HomePsych, an online communication platform for mental health clinics. The product is currently being beta tested.  HomePsych provides secure patient and practitioner accounts for online processing of psychological surveys, integrated with online telepsychiatry infrastructure for use in rural communities.

Iconic Health is presently creating a novel medication management system called Medication IconoGraphs (MIGs), data visualization tools to help mental health care providers manage a patient’s complex medication regimen.

MIGs are designed to assist medical professionals in understanding an individual patient’s medication list. A set of MIGs displayed on a computer screen represent a patient’s medication regimen, including the patient’s use and experience with each medication, as well as drug-drug interactions and other reference information. At this time, MIGs are at an early stage of development, consisting of a paper-based blueprint (submitted as a patent application) and Flash-based prototype web page. Although MIGs may be useful for any medical specialty, the initial functional version will be developed with a focus on adult psychiatric medication management.

Iconic Health intends to license MIGs as a module or “widget” within well-developed third party products, e.g., electronic medical records or ePrescribing systems. As most EMR products lack innovative visuals, they appear homogenous to prospective customers. Inclusion of MIGs would distinguish a licensee’s product from competing EMRs, providing a potentially large competitive advantage. During the Phase I work plan, we will also investigate the feasibility of deploying MIGs as a stand-alone program or a module within HomePsych specialized for adult psychiatry.  Psychiatry is within our scope of expertise and an initial specialty product could be effective with a smaller set of medication reference data.

Health Information Technology (HIT), such as electronic medical records (EMRs) and ePrescribing software, has become the focal point for containing health care costs and improving current medical care (Murff, 2001). However, adoption of HIT has lagged significantly behind both projections of use and in the development of new models for the presentation of medical information to providers and patients. Nowhere has this lag been more evident than in the development of decision support systems (Murff, 2001). This lag has been precipitated by numerous factors including the lack of standards detailing how data is to be collected and stored and the inability of software vendors to bypass proprietary considerations in favor of cooperation and collaboration. A casualty of this fragmented development method is that few, if any, software products have been developed with the direct input of large numbers of potential physician users. This lack of feedback has produced software products, including decision support systems, which are plagued by design faults that render the products ineffective. Primary among these faults is that most developers do not understand that most physicians (and the population in general) are visual learners, and that, if EMRs and ePrescribing are to be readily absorbed and utilized, information should be presented in a graphical format (Alafaireet, 2006). Given the growing amount and complexity of drug information available to physicians, it is imperative that a more functional solution be found to deliver this vital information to physicians in a timely, user friendly manner. This is the challenge that Iconic Health has undertaken.

Medication IconoGraphs (MIGs) were created (in paper design) by a practicing psychiatrist in 2005 as a vehicle to improve patient care by enabling the accurate identification of prescription medication, and the various interactions of these prescription drugs within a patient’s care regimen. The concept of MIGs was developed in direct response to a perceived need by physicians, perhaps even more so than most individuals, to receive information in visual format for fast and complete processing.

A set of MIGs creates an enriched medication list designed for rapid interpretation by the prescriber. For each medication in the patient’s regimen, a MIG displays the pill image, relative dosage, interactions, duration of treatment, treatment compliance, patient impression of efficacy and side effects, tendency to cause weight gain, and tendency to cause sedation.  MIGs display a wealth of clinically relevant information in a data-dense, easily interpretable format.

Medication IconoGraphs represent a paradigm shift in conceptualizing medication lists  In present clinical practice, in paper and electronic charts, medication lists are presented to the provider with very little detail or contextual information.  For example, a two medication regimen is often displayed to providers as: “Lamictal 200mg PO qAM, Carbamazepine 400mg PO TID” whether in written or electronic format.

The concept of visualizing the convergence of general medical reference data with patient-specific data is, to the best of our knowledge, unique. Three years after the original conception of MIGs, there appears to be no products on the market for visualizing a patient’s medication regimen.  Products do exist, however, that enable the identification of pharmaceuticals by sight and some also link the medication to documentation regarding the drug, its dosage and potential side effects. None of these known products, however, display the drug treatment regimen of a single person.

Furthermore, the basic IconoGraph design of perpendicular bar graphs encasing a set of icons is novel for any field, medical or otherwise. Iconic Health’s assertion that the IconoGraph design is entirely unique is based on:

Since the IconoGraph design and the concept of detailed visualization of medication regimens is new, very little directly relatable research can be utilized to form the experimental basis required for this work to proceed.  This revised work plan is intended to advance knowledge in this area by validating an optimized set of designs so that an improved, comprehensive blueprint and adequate information can be presented to support software development and the required database structure necessary for implementation of this much needed concept.

MIGs were designed by Jason Cafer M.D. when he was a Psychiatry resident at University of Missouri-Columbia to address the problem, “I could provide better care if I understood more about each patient’s experience with their medication regimen, and much more information is available than I can assemble and interpret in the time allowed.” MIGs are to be viewed by psychiatrists or other mental health providers responsible for managing the patient’s medications, presumably on a computer screen, although much of their value could be preserved on a black and white paper printout.  At this time, MIGs are at an early stage of development, consisting of a paper blueprint (submitted as a patent application) and Flash-based prototype web page.

The intended user of MIGs is the medication prescriber and potentially other mental health professionals who need to understand the patient’s medications, e.g., psychotherapists, nurses and caseworkers.  MIGs offer considerable potential advantages over current technologies used to enhance prescribers’ understanding of a patient’s medication regimen.

With ePrescribing or Computerized Physician Order Entry (CPOE), prescribers are currently alerted to potentially dangerous interactions by “pop-up” alerts when they are attempting to prescribe an interacting medication.  It is difficult to set these alerts at appropriate sensitivity or specificity.  If alerts are triggered too easily, prescribers develop “alert fatigue” and tend to ignore them altogether (Weingart, 2003). When alerts are only triggered by the most dangerous interactions, doctors may remain unaware of additive effects or more subtle clinically relevant interactions.  In the context of ePrescription or CPOE, the patient’s MIG display would re-render and incorporate the new medication prior to finalizing the medication order.

There are numerous potential methods to collect the patient-specific data needed drive the MIG visuals. Thoughtful analysis is necessary to decide which method(s) are feasible for the initial versions of the MIG module. Patient-specific information could be entered manually by health care providers, pharmacists or, potentially, patients themselves. Information on medication dosage, duration or treatment, compliance could be imported to the MIG application from EMRs or pharmacy databases.

Generally, drug interactions are of three types, those that: 1) increase or decrease the blood concentration of a medication by affecting absorption or clearance of that medication from the body, 2) cause problems such as excessive sedation through combined additive effects, 3) interact through molecular mechanism to increase the probability of a relatively rare risk of a medication.  With MIGs, blood level adjustments are represented intuitively by 1 to 3 triangles pointing upward or downward from the upper border of the Dosage Bar, visually raising or lowering the height of the Dosage Bar to correct for the interaction.  As proposed, interactions other than adjustments for blood levels are displayed as a dark star icon of various shades, black representing the most dangerous potential interactions.  Callouts appear when the cursor hovers over the IconoGraph, describing the interaction specifically with text in the manner of traditional medication checking programs.  Also, IconoGraph(s) of the medication(s) responsible for the interaction will be illuminated when the cursor hovers over the IconoGraph of the affected medication (not shown).  When possible, Iconic Health intends to include medical conditions and genetic test results into the visualization scheme, e.g., hepatic insufficiency resulting in upward adjustment of blood levels of medications cleared from the bloodstream by the liver.  Figure 5 provides concrete examples of how Interaction Indicators in conjunction with Dosage Bars can improve quality of medication management at reduced cost.

Several programs currently allow prescribers to manually check if potential interactions exist between 2 or more drugs.  Prescribers do not routinely run interaction checks, as considerable time is spent entering medications and interpreting the results.  No visuals are presented.  For example, an interaction check with ePocrates, the most used physician medical reference software, yields the following results.

Compare this to the MIG display for these medications in Figure 5, which shows the Paxil IconoGraph. (The actual MIG display would show 4 IconoGraphs, one for each medication). Note that St. John’s Wort, commonly used for treatment of depression, cannot be queried through ePocrates Multi-Check.  Since the initial version of MIGs will be tailored for psychiatric practice, St. John’s Wort will certainly be included.

The Impression Bar is a work in progress, and this study will analyze precisely what information will be displayed and how it will be collected.  As currently proposed, information needed to construct the Impression Bar would need to be entered manually as this data is not formatted in current Health IT systems.  The general concept is to use the space along the lower border of the MIG to indicate some formulation of positive and negative aspects of each medication as experienced by the patient.  Conceptually, this could be any combination of perceived therapeutic benefit, perceived negative effect, affordability, patient’s desire to continue a medication, a score on a psychometric scale, etc. The Impression Bar could reflect the answer to a single question, e.g., “Are you satisfied with this medication?”  It could reflect the prescriber’s impression of the medication’s benefit to that patient. It does not necessarily need to be a bar, nor does the black versus white contrast need to be used.  Possibly, this space could be used for additional text or an indicator of some entirely different data.  The research team will resolve these questions and present rationale in the Phase 1 report.

This  Phase I study will also investigate the feasibility of displaying additional medical reference information (of particular relevance to psychiatric medications) without obtrusive clutter of the MIG interface.  The carbamazepine IconoGraph in Figure 1, for instance, displays general tendency of that medication to cause weight gain and sedation.  The Weight Gain Indicators conceptually make the IconoGraph “fatter”, and the Sedation Indicators conceptually drag the IconoGraph downward.  The research team will analyze and debate the optimal appearance and placement of each IconoGraph element to optimize usefulness to the clinician.

Medication IconoGraphs address several significant problems in management of complex medication regimens. In present clinical practice, in paper and electronic patient records, medication lists are presented to the provider in text format with very little detail or contextual information. Prescribers’ incomplete understanding of this contextual information may lead to medication errors. Existing tools to prevent medication errors have intrinsic limitations that Medication IconoGraphs have the potential to overcome. Beyond preventing errors, IconoGraphs have clear potential to improve treatment outcomes and reduce costs of psychopharmacotherapy.

Related Information:

Loss of enzyme induction: Ups and downs of a hidden drug-drug interaction by Jonathan M. Meyer, MD and Susan G. Leckband, RPh, BCPP.  This article from Current Psychiatry explains the complexity of multiple drug-drug interactions that affect blood levels of psychotropic medication.  Medication IconoGraphs are specifically designed to address this complicated problem.