Transcript Magnesium Sulfate for the Management of

Magnesium Sulfate for the Management of
Eclampsia and Pre-eclampsia:
Some Economic and Cost Reflections
Andrew Farlow
Research Fellow in Economics, Oriel College
University of Oxford
Maternal Mortality day-conference, Oxford, June 2007
1
A note
• This presentation is a requested response to the
EngenderHealth/MacArthur Foundation background
document ‘The Utilization of Magnesium Sulfate for
the Management of Pre-eclampsia and Eclampsia’,
June 2007.
• Prepared, as requested, with an eye to research
issues from an economics/social science
perspective. Hence, a series of economics/social
science lenses… and not a rounded approach to the
issue of maternal mortality.
• Many of the themes have resonance in many other
areas: An opportunity to address cross-cutting
issues?
2
Key themes identified in MacArthur
background paper
•
•
•
•
•
•
•
•
Risk
Coordination
Health systems & health service providers
The need for context-sensitive solutions
Cost and supply
Provision and use of cost effectiveness evidence
Diagnostics/information/monitoring
Political processes (especially when they overlook some
solutions in preference for other solutions)
• Some of these may overlap in interesting ways:
– risk/coordination with health system issues
– diagnostics with cost effectiveness and risk
– risk with cost and supply issues, etc.
3
Maternal Mortality per 100,000 live
births in 2000 (2005 source)
Source: WHO “The World Health Report 2005 – make every mother and child count” (2005)
http://www.who.int/whr/2005/chap1-en.pdf
4
Causes of maternal mortalitya
5
Some background
• Approximately 63,000 pregnant women die every year
because of eclampsia and severe pre-eclampsia, which
are also associated with a higher risk of newborn deaths.*
• That is nearly 200 women every day.
• Pre-eclampsia/eclampsia ranks second only to
hemorrhage as a specific, direct cause of maternal death.
• The risk that a woman in a developing country will die of
pre-eclampsia or eclampsia is about 300 times that for a
woman in a developed country.
• Magnesium sulfate is the mainstay of treatment of preeclampsia and eclampsia in most developed countries. In
other parts of the world diazepam and phenytoin (used for
other types of seizures, including epilepsy) are more
widely used.
•
* ‘Balancing the Scales: Expanding Treatment for Pregnant Women With Life-Threatening
Hypertensive Conditions in Developing Countries, a Report on Barriers and Solutions to Treat Preeclampsia & Eclampsia’ EngenderHealth 2007 http://www.engenderhealth.org/files/pubs/maternal-6
health/EngenderHealth-Eclampsia-Report.pdf
Risk 1
• “There is no proof that evidence, no matter how clearly it is
formulated and spoon-fed to clinicians, will change
practice. Society would clearly benefit from better
understanding of what drives physicians’ behavior and
decision making.”(P4)*
• “Clinicians’ perceptions of the dangers of magnesium
sulphate may have contributed to the drug’s non-use.
Respondents acknowledged that the international trials in
which Zimbabwe collaborated showed clearly that the drug
saves lives. They also noted, however, that the belief of
many Zimbabwean clinicians in the drug’s effectiveness is
tempered by their perception of its dangers to
women.”(p7)**
*Referring to Thorp J. O’, Evidence-based medicine – where is your effectiveness? BJOG 2007; 114:1-2)
**Sevene E, et al. System and market failures: the unavailability of magnesium sulphate for the treatment of
eclampsia and pre-eclampsia in Mozambique and Zimbabwe. BMJ 2005; 331:765-769.)
7
Risk 2
•
•
•
•
•
•
“Physicians in Zimbabwe who participated in the
Magpie Study expressed reservations about safety of
use of magnesium sulfate in low-resource settings.”
(p14)
“If the utilization of magnesium sulfate were expanded,
in isolation from…other aspects of clinical practice, the
safety of magnesium sulfate and the overall implications
for lowering mortality are not known.” (p13)
These are clearly risk/risk perception issues.
“Some senior nurses feared that the intervention would
increase the demand for out-patient and inpatient care.”
(p12)
So ‘risk’ can also refer to risk of impact on local health
budgets too?
This properly needs a risk (and incentive) lens.
8
Risk 3
• If magnesium sulfate is given to women already with
preeclampsia…the clinician is already facing a high-risk
situation.
• If not being recommended for usage for mild to moderate
preeclampsia, this means it is being recommended for
sever cases only (p3).
• What is the personal cost-benefit of those administering
in terms of the risk they bear? What are their perceptions
of that risk?
• What is the global risk versus individual risk situation?
• Is the globally efficient solution bearable by those at the
local level?
• How are they ‘insured’? Can their risk be better handled?
• “Clinicians’ long use of other drugs to manage
eclampsia” (p6). Is this part of a risk averse strategy, or
the sign of some other failure?
9
Risk 4
•
What is distribution of risk ‘across the players’ of a
strategy that emphasizes the use of magnesium sulfate
particularly for eclampsia, and that places less emphasis
on its use for preeclampsia? (p14)
What about distribution of risk ‘across the players’ using a
strategy of shorter (targeted?) courses of magnesium
sulfate? (p14)*
The appropriate risk-based strategy is: For each possible
impact, minimize risks; then find ‘optimal’ solution in
‘impact space’ (this may be context sensitive [see below])
It may be a very ‘second best’-looking solution.
References to very limited drug budgets (p6), and hence
priority given to first-line drugs:
•
•
•
•
–
Hints also at perceptions of personal (clinician/hospital manager)
risk and trade-off of personal benefit v social benefit (latter could
be high, even if clinician/manager benefit not high).
*Weeks, AD, et al. Correspondence. The Lancet. Vol 360; October 26, 2002, p1329-1331.
10
Coordination issues
• Magnesium sulfate absent in Nigeria.
• Present in only 5%, 12%, and 25% of facilities in Burkina Faso,
Tanzania, and Rwanda respectively (p4).
• “Even after formal approval of the drug, difficulties with distribution and
management gave the impression to clinicians that the drug was still
unavailable. As a result, they continued to use alternative treatments
and did not request magnesium sulphate from the Central Medical
Stores or the pharmacy in their own health unit.” (p5)
• Referring to case of Mozambique: “Central Medical Stores composed
a list of purchases that included both the medicines listed in the
formulary and other drugs that clinicians regarded as necessary.
Magnesium sulphate had not been requested by clinicians, however,
and was therefore not included.” (p5)
• A coordination problem/prisoners’ dilemma?
– ‘A’ does not do since has to rely on ‘B’, who does not do since has
to rely on ‘C’ who does not do because does not think ‘A’ is going
to do, etc. Magnesium sulphate not used because magnesium
sulphate not used!
• Also practical differences since an emergency drug, and distribution
across a health system has (maybe?) all or nothing/coordination
11
features?
Health system issues 1
• “There are several reasons to be circumspect when
estimating degree to which mortality might be decreased
by increasing availability of magnesium sulfate…A very
significant portion of the maternal deaths from eclampsia
reported from many developing countries are among
women who had multiple seizures outside the hospital
and those without prenatal care* Improvements in
facility-based care are not likely to affect these women
nor prevent their deaths….It is the standard practice in
many countries to discharge women soon after
childbirth.” (p13)
*Sibai BM. Diagnosis, Prevention, and Management of Eclampsia. Am J Obstet Gynecol
2005;105(2):402-410 and Katz VL, et al. Preeclampsia into eclampsia: toward a new
paradigm. Am J Obstet Gynecol 2000; 182:1389-96).
12
Health system issues 2
• “One of the most significant barriers to improving care of
women with PE&E is the fact that fewer than 60% of
women in some countries have access to services where
preeclampsia could likely be diagnosed and fewer than
40% have access to professionals who could administer
magnesium sulfate.”(p8)
13
Countries with a critical shortage of
health service providers (doctors,
nurses and midwives)
Source: WHO World Health Report (2006)
14
Distribution of health workers by
level of health expenditure and
burden of disease
Source: WHO World Health Report (2006)
15
Maternal mortality and skilled birth
attendants
Source: WHO “The World Health Report 2005 – make every mother and child count” (2005)
http://www.who.int/whr/2005/chap1-en.pdf
16
Health systems issues 3
• Global shortage of about four million health-care
workers.
• The richer world sucking in these workers:
– UK and Europe – Failures in domestic provision?
– US – Lack of long-term human resource planning with an aging
population?
– Not to deny workers right to relocate to better themselves: Also
have to make more attractive to stay in home country.
• In addition, some recent studies have shown that
efficiency of health workers in some resource-poor
settings is heavily impaired by their need/incentive to
work outside health sector to supplement income,
absenteeism, and own ill-health (figures of efficiency as
low as 25% in some settings).
17
Estimated deaths
prevented by
vaccination
(deaths prevented in
blue, lives not saved in
grey)
The point of inserting this in
this presentation: The
diseases at the top have
cheap effective solutions*, in
an area we have spent
billions on, where all the
issues are completely
downstream. Yet coverage
is highly imperfect, and
many lives are still not
saved.
*In case of TB though, current BCG
vaccine is not good enough even if cheap.
18
Health systems issues 4
• Two-thirds of all African children who die under the age
of five could be saved by low-cost treatments such as:
– Vitamin A supplements
– Oral rehydration salts
– Existing combination therapy drugs against malaria
– Insecticide-treated bed-nets to combat malaria
• A tenth of all the diseases suffered by African children
are caused by intestinal worms:
– These can be treated for 25 US cents per child
• Again, the point is that there are many other areas with
low-cost solutions where delivery is very imperfect.
• There must therefore be common research themes.
20
Context sensitive solutions
• Magnesium sulfate might be the ‘drug of choice’ (p2) and the “use of
magnesium sulfate is now recommended worldwide…” (p15) But
what about the local context?
• “…before undertaking any intervention to improve the management
of PE&E, a thorough understanding of the local situation is needed.”
(p10)
• We must avoid “applying a solution that is unnecessarily complex,
expensive, fragile, or inconvenient.” (p9)
• However, “Aasserud et al. conclude that: “The difficulties in
obtaining information, combined with the wide and differing range of
barriers between settings, makes it difficult to envisage any single
intervention strategy…”(p13)
• Trying to get our heads around the argument that there is “a lack of
‘commercial’ incentive.” How true is this?
• Where is really at the heart of the problem?
21
Cost and supply issues 1
• One hypothesis for under-use is that magnesium sulfate may be too
inexpensive to motivate mass manufacturing, licensing, production
and distribution. (p4)
• At the same time: “While magnesium sulfate is an inexpensive drug,
the cost of this drug is ultimately a small factor in the overall cost of
management of PE&E.” (p11)
• What are the exact scale effects in manufacturing magnesium sulfate
(say, if large bulk purchases were possible)?
• Many cross-cutting examples where, to the contrary, lower COGS
(Cost of Goods Sold) is requisite for success. What key differences?
– Case of Hep B where importance of scale and appropriate
technology (and appropriate holders of the technology) and good
regulatory systems were key. [Hypothesis (p6), “complex
mechanisms of drug approval” act as barrier… There are lessons
from Hep B].
– TB vaccine investment case…. Driving down COGS is key to
uptake of booster and prime-boost vaccine combinations.
– Ditto, new generation malaria combination therapies.
– Ditto pneumococcal vaccine (where poor cost pressure and
technology decisions are harming potential impact).
22
Cost and supply issues 2
• “The cost of magnesium sulfate and the hospital care
involved with providing it were seen as barriers in some
countries. This problem of cost is reflected in the
discrepancies between private and public facilities in the
availability of treatment with magnesium sulfate.” (p8)
– What comparative work has been done on public v private
facilities, to draw out the key drivers of provision/delivery?
• Role of product price mark-ups: Recent studies (GFHR)
show that mark-ups on drugs are seen as a source of
revenue for cash-strapped health systems.
• In comparison, is magnesium sulfate sold in ways making
this difficult? If so, what are the implications for revenue
and incentives to use?
– What pricing power is there? Does it depend on the sector (public
or private) accounting for provision of magnesium sulfate or way it
is sold, etc.?
– Comparisons and contrasts with, say, way malaria drugs are sold?
– Nature of it as only an ‘emergency’ drug?
23
Cost effectiveness issues 1
• Targeting to maximize cost effectiveness (p11):
– 4 randomized trials comparing the use of magnesium
sulfate versus no treatment (placebo) to prevent
eclamptic seizures: 71 women with severe
preeclampsia needed to be treated to prevent one
case of eclampsia.
– In a subset of these patients – those women with
signs of imminent eclampsia (severe headaches,
blurred vision, or upper abdominal pain) – 36 women
required treatment to prevent one case of eclampsia.*
*Sibai BM. Diagnosis, Prevention, and Management of Eclampsia. Am J Obstet Gynecol
2005;105(2):402-410).
24
Cost effectiveness issues 2
• Referring to Simon et al., “...it was calculated that the
additional hospital care costs per woman receiving
magnesium sulfate in high, middle, and low GNI
countries were $65, $13, and $11, respectively. Many
women with preeclampsia need to be treated to prevent
one case of eclampsia.
• If treatments were reserved for only women with severe
preeclampsia, the incremental cost of preventing one
case of eclampsia in high, middle, and low GNI countries
were $12,942, $1179, and $263, respectively.
• While the authors did not calculate the cost of preventing
deaths, those costs would be considerably higher given
that it is known that only up to about 14% of women who
experience eclamptic seizures actually die.” (p11)
• So, cost effectiveness evidence seems to need some
more analysis?
• Although magnesium sulfate is cheap, there are
25
significant other cost hurdles. How/when do these bite?
Cost effectiveness issues 3
• “In low-resource settings, to what extent does magnesium
sulfate ‘compete’ with other drugs, such as diazepam
(which is also on the WHO’s Essential Drugs List)?
Diazepam has many other clinical applications, including
use as a pre-operative medication and for the treatment for
epileptic convulsions. If staff at some hospitals believe that
stocking Diazepam is easier to justify, how might this
attitude be changed? (p14)
• Diazepam and phenytoin have multiple uses, so while
magnesium sulfate might dominate on simple cost
effectiveness comparison basis, what happens if there is a
more complex cost effectiveness /organisational
comparison?
• Similar issues in, e.g., malaria vaccine cost effectiveness
measures, since tackling malaria involves a package of
measures:
– Wrong only to do cost effectiveness narrowly related to vaccine use;
26
– Ditto for TB interventions.
Cost effectiveness issues 4
• “It should be noted, however, that if only eclampsia
were treated with magnesium sulfate (and not
preeclampsia) the use of magnesium sulfate would be
an extraordinarily cost-effective intervention. Only two
eclamptic women would need to be treated to save one
life because maternal deaths are almost halved with
the use of magnesium sulfate.”
• “There are potentially additional factors in costeffectiveness equations that are not factored into the
above calculations. The Jamaican experience, cited
below, suggests overall cost savings because of a
reduction in bed-days.” (p11)
• C.f. malaria/TB/HIV calculations where the economic
costs to society are orders of magnitude higher when
these sort of costs are measured.
• Worth getting a better grasp on ALL avenues of cost
27
effectiveness?
Cost effectiveness issues 5
• Second-best thinking on cost effectiveness.
• Unrealistic to imagine in all cases that we do not have to
accept compromise and trade-off (even deaths versus
other things).
• In the vaccine world, it has recently become an almost
rule of law that advocacy needs better cost effectiveness
evidence and huge effort has gone into gathering it.
• Lessons from other initiatives to develop and deliver
health products?
28
Cost effectiveness issues 6
•
“What opportunities exist to collaborate or piggy-back
with related efforts such as overall maternal mortality
reduction initiatives or projects aimed at reducing
maternal-to-fetal transmission of HIV?” (p14):
–
–
–
–
Again recent GFHR work
Plenty of practical examples in other areas. Any transferable
lessons?
What is cost effectiveness evidence of this piggy-backing?
What past examples of piggy-backing worked? And why?
29
Monitoring/diagnosis issues 1
• “RCOG recommends monitoring patients receiving
magnesium sulfate by regular assessment of the
urine output, maternal reflexes, respiratory rate and
oxygen saturation.”* (p3)
• “Careful monitoring of blood pressure and
measurement of urine protein are required.
Laboratory studies of blood count, liver function, and
kidney function should be obtained…Furthermore,
monitoring and treatment should continue
postpartum as appropriate.” (p3)
• How costly is all this monitoring? How does it impact
cost effectiveness and practicability of this
intervention?
*Re to RCOG, Royal College of Obstetricians and Gynaecologists,
The Management of Severe Pre-Eclampsia. Guideline 2006 recommendations.
30
Monitoring/diagnosis issues 2
• Monitoring of condition that is relatively infrequent, and this
causes resistance to doing so (p4).
– How easy is it to get compliance with monitoring when patient
and practitioner are less informed about risk/costs/benefits?
• Patient and practitioner perceptions of what is of value to
them:
– MMV rice story
– Patients asked why they did not turn up for malaria-related
appointments. They explained they sat around for hours (at cost
to them), only to be sent away almost immediately when seen
by nurse/doctor. This generated a rumor of wasted time.
– Better compliance when attendees were told a story they
understood – that is was like searching rice grains for bad ones.
All the good grains need nothing. Only the bad need attention.
– Lesson: Manage expectations and educate even if it seems31
obvious to the ‘experts’ what is happening.
Monitoring/diagnosis issues 3
• How is cost effectiveness evidence impacted by diagnostics?
• Besides, there is “An insufficient number of qualified clinicians
to monitor the use of magnesium sulfate or even to prescribe
the drug in peripheral hospitals”.
• Again, lessons from other product fields. Those working on
TB, malaria, and dengue (examples known to the author)
realized at some point that cost and speed of product
development and cost effectiveness of intervention and
uptake were heavily impacted by state of diagnostics:
– Gates funded Foundation for Innovative New Diagnostics, FIND,
to tackle some of these issues (for TB and malaria and others);
• Are there diagnostic issues in the case of pre-eclampsia and
eclampsia that affect cost effectiveness, that need addressing
and may also have been overlooked?
– Mindful that this may not be comparable with the above cases.
• Re to RCOG, Royal College of Obstetricians and Gynaecologists,
The Management of Severe Pre-Eclampsia. Guideline 2006 recommendations.
32
Political/international organization
processes?
• What really drives flows of resources and priority-setting
globally?
– Pneumococcal vaccine (GAVI/G8), combined $1.8bn to solve
about 1%-2% of the total problem between now and 2030?
– Evidence from distribution of resources across areas of R&D?
– Intellectual Property debates (e.g. recent WHO initiatives like
CIPIH): Maybe IP-based debates are easier for advocacy groups
to push for, than delivery debates pushed for by delivery groups?
– Sustainability of funding flows/initiatives after the first big-hit.
Many recent initiatives have, or shortly will come up against,
financial sustainability issues. Does the need to sustain funding
have impact on advocacy and delivery issues?
– Crowding out of lower profile initiatives?
– How does advocacy and the politics of provision really work?
33
THANK YOU
Comments and feedback
always welcome:
[email protected]
34