Transcript sfinxjub2013

Kari Laine, MD, PhD
CEO, medbase Ltd
www.medbase.fi
Adverse drug reactions as a clinical problem
• Lazarou et al., JAMA 1998 (meta-analysis)
– 39 studies in hospitalized patients in the U.S.
• 6.7% of all patients (n=2 216 000) are exposed every year
to a serious adverse drug reaction
• 106 000 die every year due to adverse drug reactions
• 4-6. leading cause of death in the U.S.
• Yearly cost to the hospitals 1.56 – 4 billion USD
Hospitalization in the elderly due to
commonly known drug interactions
Glibenclamide
+
Digoxin
+
ACE-inhibitor
+
Sulphamethoxazole
Clarithromycin
Spironolactone
Hypoglycemia
Digitalis toxicity
Hyperkalemia
6.6 x risk
12 x risk
20 x risk
Amoxycillin – no
elevated risk
Cefuroxim – no
elevated risk
Indapamide – no
elevated risk
Interaction could have been avoided !
1000 to 1500 new publications on drug interactions
evidence-based information is poorly
penetrated to clinical practice
Warfarin - metronidazole
Kazmier FJ. A significant interaction
between metronidazole and warfarin.
Mayo Clin Proc. 1976;51:782-4.
O’Reilly RA. The stereoselective interaction of warfarin and metronidazole
in man. N Engl J Med. 1976;295:354-7
Why
?
• Adverse drug reactions are a great health hazard
• Those based on drug interactions could be most
often avoided by right choice of medication
• Information overflow – evidence does not benefit
doctor/patient
N Engl J Med 2010;363:2060-7.
“Empirical evidence suggests that comprehensive
EHR systems can improve adherence to clinical
guidelines and reduce rates of medication errors.”
Challenges
• Alert fatigue due unacceptable signal-to-noise
- ratio
• Unclear clinical significance – classification
• Allows thresholds
• Inconsistencies in database contents
• Lack of patient/end-user oriented system
development
• Selective alerting by Sfinx in EBMeDS
• Evidence of benefit from prospective
intervention studies
General requirements
– Short and concise warnings & recommendations
– Clear classification system for clinical relevance
– Fully based on evidence – fully referenced
• Clinical documentation needed
– Be selective - Avoid over-alerting
• Alerts need to customized
– Internationally acknowledged attributes
– INN generic drug names and ATC-codes by WHO
– RxNorm, UNII, Snowmed
– Simple technical structure / user interface
– Quarterly updates
Classification – clinical relevance
A
Minor interaction of no clinical relevance
B
Clinical outcome of the interaction is uncertain
and/or may vary
C
Clinically relevant interaction that can be
handled by for example dose adjustments
D
Clinically relevant interaction that is best
avoided
Classification – level of documentation
0
1
2
3
4
Data derived from extrapolation on the basis of
studies with similar drugs.
Data derived from incomplete case reports
and/or in vitro studies.
Data derived from well-documented case reports.
Data derived from studies among healthy
volunteers and/or on pilot studies among
patients.
Data derived from controlled studies in relevant
patient population.
1. ENTERAL peroral
2. ENTERAL non peroral
intraoral including sublingual and buccal administration
rectal
3. PARENTERAL
i.v.
i.a.
i.m.
s.c.
intrathecal
intraperitoneal
intra-articular
intraventricular (cerebral)
preparations for dialysis
preparations for nasal mucosa with systemic absorption (a.g. hormones)
transdermal praparations
implants
4. TOPICAL
eye preparations including intravitreal preparations
ear preparations
preparations for inhalation
preparations for nasal mucosa with topical effect
per cutaneous preparations including shampoo
vaginal preparations
endocervical preparations
intrauterine preparations
urethral preparations
intravesical (urinary bladder) preparations
stents
5. OTHER
future inventions for drug administration
QUALITY OF CARE – not only ADRs
• Number of generic drugs covered 1400
– All have been checked regarding interactions
• Number of interactions 17.000 (release
4300)
• Localisations: Sweden, Finland, Italy,
Austria/Germany (ongoing)
What is
lacking ?
• Interactions obvious based on the PD of drugs
• Additive sedative effects
• Agonist-antagonist –interactions (e.g. metoprolol-salbutamol)
• Activated charcoal
• Interactions where clinical relevance has not been
established in clinical studies
• There are drugs that do not have interactions
• One cannot warn on everything, the signal has to be clear
Data should be modular!
in Finnish health care
Health care system in Finland
• Primary care
• 172 primary healthcare centers
• Occupational health care by the employer
• Specialized care
• Five University hospitals
• 20 hospital districts
– Central hospitals (23)
– District/City hospitals (about 45)
– Special hospitals (15)
• Strong private sector
15000 actively
working doctors
Pharmacies
• About 800 private pharmacies
• Helsinki university hospital pharmacy
chain (16 pharmacies)
• Hospital pharmacies
6000 actively
working
pharmacists
• Distribution of SFINX in Finland:
– SFINX/Pharao portal in Healthgate
– Integration to Electronic Patient Record systems
• EBMeDS
– Integration to pharmacy IT-systems
– End-user pays principal, mostly organisational
licenses
portal in the Healthgate
• Search for interactions/ADRs of a single
drug
• Analysis of patient’s drug list
• As an encyclopedia to find information of
drug interactions and ADR-profiles
• Mechanism and background texts
Penetration of
&
health care is 99.97%
portal to public
• Integration
– EPRs
•
•
•
•
•
Tieto – Effica
Logica – Pegasos
Logica – Uranus
Acute FDS
MediConsults- Mediatri
Cover 90% of the doctors
- alert response 2%  15-20%
– Pharmacy IT-systems
• Receptum
• Pharmadata
Cover more than 95% of pharmacies
Feedback
Concluding remarks
• Experiences from the penetration of
the healthcare systems:
to
• Strong idea marketing guarantees the rapid spreading –
attitudes need correction
• Strong and respected distributors
• Both portal and integrated solutions are needed!!!
• Data is considered reliable since it is fully evidence-based
• Drug interactions are considered a difficult area to handle
without proper decision support
• Short clinically holistic warning and recommendation texts
have been the key to success
• Saves time, costs and reduces the need for consultation
will never be ready
but we are getting closer