Discontinuation

Download Report

Transcript Discontinuation

LINKING MICRO- AND MACROUTILIZATION DATA
Euro-DURG Meeting, Prague, June 8, 2001
John Urquhart, MD, FRCP(Edin)
Pharmaco-epidemiology Group
Maastricht University,
Maastricht, Netherlands
MAIN SOURCES OF
PHARMACO-EPIDEMIOLOGIC DATA
MARKET
PHARMACY
PATIENT
MACRO
SEMI-MICRO
MICRO
Definitions at the MICRO level of
Ambulatory Pharmacotherapy - 1
• Acceptance -- does the patient accept or not
the recommended treatment?
• Execution -- how well does the patient
execute the recommended regimen?
• Discontinuation -- when does the patient
stop taking the medicine?
Definitions at the MICRO level of
Ambulatory Pharmacotherapy -2
• Acceptance -- dichotomous
• Execution -- continuous
• Discontinuation -- dichotomous
Definitions at the MICRO level of
Ambulatory Pharmacotherapy -3
• ‘Adherence’ is a useful overall term, subsuming
– Acceptance,
– Execution, the quality of which is called
compliance
• the extent to which the actual dosing history corresponds
to the prescribed drug regimen
– Discontinuation
• The length of time between acceptance and
discontinuation is called
persistence
PRESENT VIEWS ABOUT
COMPLIANCE ARISE FROM A
METHODOLOGIC REVOLUTION
• Electronic monitoring methods, introduced in
the late 1980’s, revealed that pre-electronic
methods grossly overestimated compliance
–
–
–
–
histories
diaries
returned tablet counts, canister weights
drug levels in plasma (white-coat compliance)
• These methods allow patients easily to censor
evidence for delayed or omitted doses.
25 YEARS OF
DEVELOPMENTS IN
ELECTRONIC MONITORING
quick overview
1976
$500 each
($1200 today)
1986
$400 each
1991
$250 each
2000
$80 each
Transition from punctual to
erratic
holiday
WHITE-COAT COMPLIANCE-taking the medicine just before
the scheduled visit
RELATIONSHIP BETWEEN ADHERENCE AND
VIROLOGIC FAILURE IN HIV INFECTION
100
90
80
70
60
50
40
30
20
10
0
94
75
50
36
19
<70
1/3 <70%
1/3 >95%
70-80
80-90
90-95
>95
% PRESCRIBED DOSES TAKEN
Paterson, , et al., Ann Int
Med 133: 21-30, 2000
Patterns of compliance in medically
unselected patients: 1 patient in ca. 6...
• punctual dosing
• all doses taken but
slightly erratic timing
• 5-20% of doses omitted,
but never more than 1 at
a time
“rule of sixes”
• many doses missed;
has a drug holiday 3-4
times /yr
• many doses missed;
has at least 1 drug
holiday/mo.
• takes few or no doses,
but presents as
compliant
Similarity of dosing patterns between fields
• Compliance appears to be more or less
independent of
–
–
–
–
drug
disease
prognosis
symptoms
• The usual patterns of poor & partial compliance
prevail in asymptomatic diseases, treated with
convenient drugs without attributable side-effects
• Compliance is mostly a patient-attribute, rather
than a disease- or drug-attribute.
BJR 35:60-5,
1996
% of prescriptions in category
Effect of typical compliance patterns on
refill intervals: 60-day prescription
40
35
Median: 71days
30
25
Mean: 78 days--4.7 refills/yr, not 6
20
15
10
5
0
60
63
71
80
92
109
133
171
240
400
Days needed to empty a 60-day prescription
100
80-89
60-69
40-49
20-29
90-99
70-79
50-59
30-39
10-19
% of prescribed doses taken
Role of Prior Selection - 1
• 30-40% of medically unselected patients
will substantially underdose, per the ‘rule
of sixes’
• ‘medically unselected’ means that the
patients have had
– no prior medical treatment
– ‘the usual’ interactions with a physicianprescriber and a pharmacy-dispenser
Role of Prior Selection - 2
• Patients who have failed to respond to a
prior course of rationally-prescribed
drug treatment will include more than
the usual proportion of poor/partial
compliers
• EXAMPLE: Burnier & Brunner found
that 53% of hypertensives who had been
escalated to triple therapy without
responding, were clinically unrecognized
poor compliers
Role of Prior Selection - 3
• Patients who have responded well to a
prior course of rationally-prescribed drug
treatment will include more than the usual
proportion of good compliers
• EXAMPLE: de Geest found poor
compliance in 5% of patients with severe
heart failure who had endured a long wait
for a suitable heart and then had
successfully received a cardiac transplant
Role of Selection by Compliance in a
Prior Course of Medical Treatment
good-enough
not goodenough
unselected
Role of Selection by Compliance in a
Prior Course of Medical Treatment
not goodenough
good-enough
unselected
‘drug
refractory’
hypertension
cardiac
transplant
recipients
SHIFTING GEARS
We turn to
PERSISTENCE
...
WHAT DO WE KNOW ABOUT
PERSISTENCE?
• Example: statins in Quebec*
– median persistence 173 days
– 33% persistence at 1 year
– despite full reimbursement & optimal convenience,
side-effect profile, efficacy
• Similar stories in hypertension and other
chronic-use medicines
• Major public health disappointment - but not
widely recognized...
*Catalan & LeLorier, Val Hlth, 3: 417, 00
HUGE CHASM BETWEEN PUBLIC
HEALTH OBJECTIVES AND REALITY
• A fortnight ago the US National Heart, Lung &
Blood Institute issued revised guidelines on
treatment of lipid disorders
• The guidelines would result in statin treatment
of ca. 20% of people over age 40
• To realize the goals of this program, median
persistence should be 20 years, not 6 months
• Closing that huge chasm has huge economic
implications
COMMON-SENSE BASICS ABOUT
PERSISTENCE
• Persistence increases one dose at a time
• The link between compliance and persistence is
the patient’s confidence in the values of ...
– the treatment program
– correct compliance with the dosing regimen
– continuation
• Regular review of the patient’s dosing history is
a natural way to achieve two desirable goals...
– assure satisfactory compliance
– reinforce the importance of correct, ongoing dosing
WHO PAYS?
• Lengthening persistence proportionally
increases the manufacturer’s revenues
• Gross margins on premium-priced medicines
($2-4/day or more) are >80%
– One added year of persistence with a $3/day product
returns $800 in gross margins
• In commercial quantities, electronic monitors
with a 2-year lifetime now cost $50.
• A good economic return is possible, IF indeed
compliance is a main link to persistence.
SOME CONSEQUENCES OF
SHORT PERSISTENCE WITH DRUGS
INDICATED FOR LONG-TERM USE
• high ratio of new to refill prescriptions but slow
or no growth in the total market
• heavy promotional expenses
• poor cost-effectiveness
• resources wasted on both diagnostics and drugs
• short persistence is in no-one’s interest
PERSISTENCE
High priority topic linking
micro- and macro-aspects of
drug utilization research
COMMENTS : [email protected]
500+ paper bibliography of published studies
based on electronic monitoring: www.aardex.ch