Review of Darvocet and Potential Design to Improve Product Safety
History
of Darvocet & Market Failure
Darvocet a powerful pain
killer which contains an active ingredient propoxyphene napsylate and
acetaminophen, and became one of the most commonly prescribed pain medications
in the United States until its removal in 2010. The drug was manufactured by
Xanodyne Pharmaceuticals Inc. of Newport, Ky. The active ingredient in Darvocet
propoxyphene was introduced to the US market in the 50’s and was officially
removed in 2010 after studies proved that this ingredient caused adverse
effects in heart rhythms. Immediately after Darvocet was pulled from the
shelves in the US, the FDA called other manufacturers to pull generic drugs
with the propoxyphene ingredient from the market (Allen, 2010).
Darvocet was introduced to
the market in 1976, up until its removal from the market in 2010. Darvocet
along with related drug Darvon where under scrutiny during its early release to
the market due to known heart risks and related adverse effects and deaths.
After the Public Citizen group filed petitions during the earlier years,
stricter warnings where placed to warn patients and providers of usage and side
effects if inappropriate use occurred. In 2005 the British market became the
first to pull Darvocet from the market due to its known side effects. Not long
after in 2006 the Public Citizen’s group reconvened and petitioned for the
recall of Darvocet for potential health issues primarily related 76% of deaths
related to cardiac toxicity. In 2008 the group filed a suit against the FDA for
disregarding their petition of 2006. Then in 2009 the FDA convened an advisory
panel to review Darvocet voting 14 - 12 in favor of pulling Darvocet from the
market. However, despite the vote the FDA allowed Darvocet to remain in the
market with stricter warning meanwhile, requesting a study to review the active
agent in the drug and adverse effects. In 2010 the FDA as a result of the study
pulled the drug from the market (Saiontz, & Kirk, n.d.).
Intended
Clinical Goal & Side Effects
Darvocet was used with the
clinical reason to treat mild to moderate pain. The two active agents in the
drug acetaminophen which increases the effects of propoxyphene (FDA, 2009).
Propoxyphene is a narcotic pain reliever used to replace the more addictive
prone related drugs morphine, and codeine, supposed to be used as a safer less
addictive narcotic. Darvocet must be used appropriately and is a high risk
prescription. Certain conditions must be noted in order to avoid complications.
Patients must advised physicians of certain conditions that include breathing
disorders, lung of kidney disease, history of brain injury, stemming or
intestinal disorders, suicidal behavior, or mental illnesses, drug or alcohol
addiction. Most common side effects include respiratory problems, fainting,
chest pain, unusual though behavior, seizure, or nausea and vomiting. More
severe side effect is death (FDA, 2009). Certain prescriptions that can cause
adverse effects if taken concurrently include blood thinners, birth control
pills, diuretics, antidepressants, antifungal’s, heart or blood
pressure medications, and amongst many other prescriptions. The interactions
with these drugs can result in the aforementioned side effects, death, or
overdose (drugs.com, 2018).
Solution
Description
In order to manage and
track drug performance, utilization, and outcomes. I would implement a solution
that tracks specific measures and outcomes that result from patients who
receive the prescription. According to Lougheed et al. (2012), the capability
to monitor affiliated healthcare providers can ensure that they are
appropriately providing a prescription in accordance to its designated
protocol. With regard to prescriptions management, having this extra layer of
validation in tracking provider prescribing trends, usage, and monitoring can
result in improved performance and reduce the risk of adverse events. In reviewing
the multiple resources I would pursue an infrastructure as a service (IAAS)
platform on a community cloud, which pulls specific data from a provider EHR’s
to include patient prescription data, patient diagnosis data, office visit
information, prescriber information, and ordering trends. In addition we would
need to monitor order history amongst other qualifying data that would allow
our organization to streamline best practice processes and improve accuracy of
dosage and drug usage, as well as additional side effect warnings as
needed.
With the plethora of data
being requested this would require a substantial amount of data storage and
processing technology. With this in mind, I would pursue an IAAS model,
although cost related to infrastructure including thin clients, apps, routers,
and other hardware will be at the organizations expense. The cost for high
performance data centers are accessible on a pay as you need basis and overtime
may incur our organization less cost. Moreover, the benefit includes less
security risk that is related to allowing a company to manage entire platform
as seen in a software as a service model (SaaS). Another possibility would be
to pursue a platform as a service model (PaaS). However, with this model the
infrastructure is included in a rental fee rather than owned and over time can
incur additional unwanted cost (Stokes, 2013). The benefit to IaaS is the
scalability and ability to pay for what you need. Investments in IaaS allow end
users to access servers over the cloud in real time and manage data. The
downside is the only service being provide is the storage and server space. The
organization will need to rely on internal IT resources to develop and
transform data in into scalable data that can be reviewed in an understandable
format.
In order to transform the
big data into understandable metrics we would leverage our IT resources. The
issue with big data is it comes in unstructured format. In order to transform
this data into formatted data that is relevant to our goal we must incorporate
a Hadoop Distributed File System (HDFS). This will allow us to bring in
unstructured data and apply a schema on read methodology to apply rules to
manipulate the data once we are ready to review the data. The unstructured data
is compressed across multiple servers and allows us to access this data at any
given time. The difference being compared to its counterpart SQL, the data from
Hadoop is accessible regardless if a server is down. The Hadoop model allows
for the information from the server that is down to re-route to another server
rendering the data accessible. The caveat is the Java based query which
requires IT resources who are familiar with this language to be able to write
queries that can provide the results required (Borthakur, 2008). Finally, the
data needs to be transmitted in accessible format, and once this is completed
by means of Hadoops extract transform load (ETL) job then we can load the data
into the client server on a graphical user interface that can be accessed over
a virtual private network.
Diagram
Required
Data
As mentioned, the clinical
data important for this program to succeed is information regarding patient
prescription data, patient diagnosis data, office visit information, prescriber
information, and ordering trends. This data would be used to manage usage and
track any unintentional and intentional issues that arise to include side
effects and relevant data necessary to create new usage requirements.
|
Data Type
|
Data Transport Mechanism
|
Vocabulary Type
|
Specific Code
|
Reasoning
|
|
Patient
Visit Data
|
Data will be extracted
via a TXT file
|
CPT
|
99211, 99212, 99213,
99214,and 99215
|
Codes used for office
visits or subsequent visits
|
|
Patient
Diagnosis
|
Data will be extracted via a TXT file
|
ICD-10
|
R52
|
Code is used for pain unspecified. This will allow me to track
inefficiencies in diagnosis related orders. I would use ICD-10 to also
monitor other history of diagnosis information that can affect the use of the
drug.
|
Conclusion
My system would allow us to
manage a large set of data using a Hadoop cluster to extract unstructured data
in the form of plain text files. The files will then be loaded into the servers
then transformed into formatted data that allow us to measure quality data that
will be loaded into a GUI to be analyzed and reviewed further. Our technology
will run over a community cloud that allows us to share metrics across multiple
parties that require access to the data and accessible over a VPN. This process
will work given the data size and allow us to leverage big data technology to
fully manage our data over a cloud platform.
References:
Allen, J.
(2010). Manufacturer Pulls Darlin, Davocet; FDA Wants Generic Makers to Do the
Same. Retrieved from: https://abcnews.go.com/Health/PainArthritis/painkillers-darvon-darvocet-coming-off-us-market/story?id=12194165
Borthakur, D. (2008). HDFS Architecture Guide. Retrieved from: https://hadoop.apache.org/docs/r1.2.1/hdfs_design.html
FDA.
(2009). MEDICATION GUIDE DARVOCET-N® 50 [dar-vo-cet-N] (C-IV (propoxyphene
napsylate and acetaminophen) tablets DARVOCET-N® 100 [dar-vo-cet-N] (C-IV)
(propoxyphene napsylate and acetaminophen) tablets. Retrieved from:https://www.fda.gov/downloads/Drugs/DrugSafety/UCM187067.pdf
Lougheed, C., Jain, A.,
Meil, D., Jarrell, B. (2014). U.S. Patent No. 2014/0032240 A1.
Washington, DC: U.S. Patent and Trademark Office. Retrieved from: https://docs.google.com/viewer?url=patentimages.storage.googleapis.com/pdfs/US20140032240.pdf
Saiontz,
D,. & Kirk, H. (n.d.) Darvon and Darvocet Problems Timeline; Darvon and
Darvocet heart problems that led to the drugs being pulled from the market in
2010. Retrieved from: https://www.youhavealawyer.com/darvocet/problems-darvon/
Stokes,
D. (2013). Compliant Cloud Computing Managing the Risks. Pharmaceutical
Engineering, 33 (44). 1-11. Retrieved from:http://www.percipient.co.uk/wp-content/uploads/2015/08/compliant_cloud_computing.pdf
