Difference between revisions of "Reviewing a clinical decision aid for the selection of anticoagulation treatment in patients with nonvalvular atrial fibrillation: applications in a US managed care health plan database"

From Clinfowiki
Jump to: navigation, search
(Methods)
 
(8 intermediate revisions by 3 users not shown)
Line 3: Line 3:
 
== Background ==
 
== Background ==
  
While newly-available computer-based physician order entry (CPOE) can provide many benefits for the medical organizations and providers that use them, there are certain patient safety risks that might be presented after a CPOE implementation. With the [[Meaningful use]] incentive promoting EHR and CPOE adoption in both the inpatient and outpatient settings, the accelerated implementations have increased reports of negative effects of these systems’ use. As a result of these negative reports, there has been a push to create a method of assessing CPOE-specific risks for healthcare organizations.
+
There have been many [http://www.texasheart.org/HIC/Topics/Meds/blodmeds.cfm/'''anticoagulant'''], besides [[Reducing warfarin medication interactions | Warfarin (which has many negative interactions)]] such as apixaban, dabigatran, and rivaroxaban that have been introduced and accepted as effective treatment options to prevent and treat stroke and systemic embolism patients. All of these therapy alternatives have their own risks as well as unique benefits and it can be difficult for clinicians to pick the best one for each specific patient's circumstance. In order to help mitigate this issue, a clinical decision aid was created to help prescribing clinicians choose the best type of coagulation therapy by comparing the available treatment options with a particular patient's  individual factors such as risk values and bleeding ratio.
  
== Materials and Methods ==
+
== Methods ==
  
The authors performed analysis on relevant literature based on EHRs/CPOE systems and patient safety and interviewed subject matter experts in order to create a base of 250 items to include in the assessment. These items were then preliminary tested at site visits with clinicians. After this first round of tests, the items had been compressed to 22 concepts to be assessed.
+
The authors gathered national medical claims data for patients diagnosed with AF, who had two or more healthcare encounters that were at least thirty days apart between 2005 and June 2010. These patients were divided into two subgroups - those on commercial health insurance plans, and those on [http://www.medicare.gov/part-d/'''with medicare Advantage part D coverage''']e. All of the patients' HAS-BLED and CHA2DS2-VASc stroke risk score was calculated with the information from the claims and the percentage distribution of each possible combination of HAS-BLED and CHA2DS2-VASc scores was created and each combination's clinical decision aid recommendation was recorded using a baseline bleeding ratio of 2:1. The percentage of the patients that would be recommended to use each choice of anticoagulant was calculated.
 
+
They presented this 22 item assessment to a group of chief medical informatics officers (CMIO) and asked to them to respond with how much deliberation the CPOE system presented for each of the items. The responses to the assessment were assembled and analyzed.  
+
  
 
== Results ==
 
== Results ==
  
There were nine CMIOs that returned the completed assessment, from various hospitals across the country, with a variety of EMRs in use. The CMIOs indicated that they were mostly able to complete the assessment by themselves and thought the items were concise and some suggested a few extra items to be assessed (a common theme in the respondents' interviews were the concerns of over-use of alerts). The respondents all agreed that the assessment was useful and perceived the purpose of the guide as a review for making sure the widely accepted practices in CPOE are implemented.
+
The study's findings suggested that there was a strong positive correlation between HAS-BLED and CHA2DS2-VASc scores. Both the mean HAS-BLED and CHA2DS2-VASc stroke risk scores were higher in the sample of patients with Medicare Advantage with part D coverage than the sample with commercial insurance plans.  
  
== Discussion ==
+
If the Clinical Decision Aid chose the treatment for the total sample population the distribution of recommended therapies would be:
  
The authors successfully created and field tested a SAFER assessment. Some of the takeaways from this project included:
+
70.50% apixaban
* The opinions of what practices make up a safe and effective CPOE integration differed greatly from one CMIO to another.  
+
25.86% no treatment
* Some respondents thought that there were important items left out of the assessment.
+
3.62% ASA
* Wording of the assessment guide was unclear and needed to be more specific.
+
0.01% dabigatran 150
* Some of the recommended items in the assessment were not feasible with the various practices’ current software.
+
  
== Commentary ==
+
ASA + clopidogrel, dabigatran 110, and rivaroxaban would not have been recommended by the clinical decision aid for any of these patients.
  
In this article, the authors tried to help come up with a solution to a nationally escalated need. Taking in input from various resources to create a guideline for an assessment was a great way to start tackling this project. They tested this and have tweaked their original assessment guide to mirror what they had learned in this project which is great, but they could definitely be overloaded with feedback if they continue to test. Since they only had nine respondents in the test, they were able to go through and thoroughly include all feedback, but had they received a response from every hospital using an EMR, a more analytical approach to feedback patterns would most likely need to be taken. 
+
== Discussion ==
  
 +
The clinical decision aid generates recommends therapies based on risks for stroke and major bleeding (which together create the net clinical outcome). The clinical decision aid uses the findings of several anticoagulation medication trial studies to help calculate which therapy would be best for people with various risk factors. This and other evidence-based tools can help physicians when many treatment options are available but many risk and benefit factors go into selecting the best option for a specific patient. With the availability of a relative abundance of new antithrombotic agents this has created a knowledge gap between reasearch and clinical practice. There still exists a great deal of indecision among clinical experts and between organizations and jurisidictions.<ref name="clinical decision aid for antithrombotic therapy">A clinical decision aid for the selection of antithrombotic therapy for the prevention of stroke due to atrial fibrillation.http://eurheartj.oxfordjournals.org/content/33/17/2163</ref>
 +
 +
== Commentary ==
  
 +
In this article, the authors point out that this also might be helpful to not only physicians and healthcare providers, but also to healthcare plan payers for population-level patient care optimization which I think is a very interesting point. Since, such a high majority of patients would have been recommended to try apixaban, I'm not sure how this aid would be so important. This seems to me that the majority of patients would be prescribed apixaban and special cases would be more fully analyzed by a patient's care team to determine the be st solution.
  
 
== References ==
 
== References ==
Line 33: Line 34:
  
 
[[Category: Reviews]]
 
[[Category: Reviews]]
[[Category: CDE]]
+
[[Category: CDS]]
 
[[Category: EHR]]
 
[[Category: EHR]]

Latest revision as of 04:43, 12 March 2015

This is a review for Steven B. Deitelzweig, MD, MMM, Yonghua Jing, PhD, Jason P. Swindle, PhD, MPH, and Dinara Makenbaeva's, MD, MBA Reviewing a Clinical Decision Aid for the Selection of Anticoagulation Treatment in Patients With Nonvalvular Atrial Fibrillation: Applications in a US Managed Care Health Plan Database.[1]

Background

There have been many anticoagulant, besides Warfarin (which has many negative interactions) such as apixaban, dabigatran, and rivaroxaban that have been introduced and accepted as effective treatment options to prevent and treat stroke and systemic embolism patients. All of these therapy alternatives have their own risks as well as unique benefits and it can be difficult for clinicians to pick the best one for each specific patient's circumstance. In order to help mitigate this issue, a clinical decision aid was created to help prescribing clinicians choose the best type of coagulation therapy by comparing the available treatment options with a particular patient's individual factors such as risk values and bleeding ratio.

Methods

The authors gathered national medical claims data for patients diagnosed with AF, who had two or more healthcare encounters that were at least thirty days apart between 2005 and June 2010. These patients were divided into two subgroups - those on commercial health insurance plans, and those on with medicare Advantage part D coveragee. All of the patients' HAS-BLED and CHA2DS2-VASc stroke risk score was calculated with the information from the claims and the percentage distribution of each possible combination of HAS-BLED and CHA2DS2-VASc scores was created and each combination's clinical decision aid recommendation was recorded using a baseline bleeding ratio of 2:1. The percentage of the patients that would be recommended to use each choice of anticoagulant was calculated.

Results

The study's findings suggested that there was a strong positive correlation between HAS-BLED and CHA2DS2-VASc scores. Both the mean HAS-BLED and CHA2DS2-VASc stroke risk scores were higher in the sample of patients with Medicare Advantage with part D coverage than the sample with commercial insurance plans.

If the Clinical Decision Aid chose the treatment for the total sample population the distribution of recommended therapies would be:

70.50% apixaban 25.86% no treatment 3.62% ASA 0.01% dabigatran 150

ASA + clopidogrel, dabigatran 110, and rivaroxaban would not have been recommended by the clinical decision aid for any of these patients.

Discussion

The clinical decision aid generates recommends therapies based on risks for stroke and major bleeding (which together create the net clinical outcome). The clinical decision aid uses the findings of several anticoagulation medication trial studies to help calculate which therapy would be best for people with various risk factors. This and other evidence-based tools can help physicians when many treatment options are available but many risk and benefit factors go into selecting the best option for a specific patient. With the availability of a relative abundance of new antithrombotic agents this has created a knowledge gap between reasearch and clinical practice. There still exists a great deal of indecision among clinical experts and between organizations and jurisidictions.[2]

Commentary

In this article, the authors point out that this also might be helpful to not only physicians and healthcare providers, but also to healthcare plan payers for population-level patient care optimization which I think is a very interesting point. Since, such a high majority of patients would have been recommended to try apixaban, I'm not sure how this aid would be so important. This seems to me that the majority of patients would be prescribed apixaban and special cases would be more fully analyzed by a patient's care team to determine the be st solution.

References

  1. Steven B. Deitelzweig, MD, MMM, Yonghua Jing, PhD, Jason P. Swindle, PhD, MPH, and Dinara Makenbaeva's, MD, MBA Reviewing a Clinical Decision Aid for the Selection of Anticoagulation Treatment in Patients With Nonvalvular Atrial Fibrillation: Applications in a US Managed Care Health Plan Database. Clin Ther. 2014 Nov 1;36(11):1566-1573.e3. doi: 10.1016/j.clinthera.2014.09.016. Epub 2014 Oct 23. http://www.ncbi.nlm.nih.gov/pubmed/25438725
  2. A clinical decision aid for the selection of antithrombotic therapy for the prevention of stroke due to atrial fibrillation.http://eurheartj.oxfordjournals.org/content/33/17/2163