Evidencebased consumer health information the need for unbiased risk communication

1
Evidence-based consumer health information the
need for unbiased risk communication

• University of Hamburg

Höldke B., M.P.H. and Mühlhauser I., Univ.-Prof.
Dr. med. University Hamburg, Unit of Health
Science and Education, Martin-Luther-King-Platz
6, 20146 Hamburg, FON 0049-40-428.38.3528 FAX
0049-40-428.38.3732
E-mail bhoeldke_at_uni-hamburg.de
ingrid_muehlhauser_at_uni-hamburg.dehttp//www.chem
ie.uni-hamburg.de/igtw/lehre/
2
What means unbiased consumer health information?

• Online consumer health information is rapidly
  growing and an active part of patients and
  consumers in decision making about preventive or
  therapeutic interventions is increasingly
  demanded.

Patient- Charta 1
Gesundheits-reform 2000 65b 3
GMC 2
The basis for informed consumer choice is the
communication of evidence-based scientific data
in a format that is clearly understood by most
lay-persons 4.
The way study results are presented (framing of
data) influence decisions by health care
providers and patients or consumers alike 5.
3
Informed consumer choice needs

• Evidence based data
• Outcome data should be reported as absolute
  numbers, absolute risk reductions or numbers
  needed to treat

Beyond the question of whether relative or
absolute differences are used, outcome data can
be framed byeither emphasizing achievable
benefits or the lack of such benefits
This means that a presentation of data as the
proportion of patients who remain free of a
target outcome rather than the proportion of
patients who benefit from a certain intervention
could substantially influence decision making.
4

• Risk communication
• Studies evaluating the communication of treatment
  results to patients were often focussed on the
  benefits of the respective interventions.

Such an approach is incompatible with unbiased
informed decision making by the patient, client
or consumer.
Therefore, instruments to judge the quality of
printed or online consumer health information
have to include rating the framing of outcome
data.
5
Towards an evidence-based and unbiased consumer
health information system

• In order to establish an online system of
  evidence-based consumer health information that
  provides unbiased evidence-based communication of
  outcome data mammography screening for breast
  cancer was used as a model.

After searching the literature according to
evidence-based medicine criteria the information
on benefits and risks of mammography screening
has been compiled.
Results are communicated as simple self
explaining illustrations as well as original
numbers equally emphasising the various aspects
of the outcome.
The described mammography screening consumer
information system is being evaluated with
experts and the target consumer population with
the final goal of an online evidence-based
consumer health information system.
6
An example for an unbiased presentation of
outcome data of mammography screening
In the following mammography screening is used as
an example to present identical data in various
usual and unusual ways. The data are based upon
an overview of the 4 Swedish mammography studies
(6). The article was selected because it reported
the necessary raw data on breast cancer mortality
and total mortality. According to this study,
mammography screening programms resulted in a
relative risk reduction of death due to breast
cancer of RR 0.80 0.70-0.92.
7
Unbiased health information - presentation for
experts
8
Unbiased health information - presentation
forlay people / patients / consumers
9
Sabine Fischer Grafik, Halstenbek, 1999
10
Sabine Fischer Grafik, Halstenbek, 1999
11
Sabine Fischer Grafik, Halstenbek, 1999
12
Sabine Fischer Grafik, Halstenbek, 1999
13

• Without mammography screening of 1000 women over
  10 years 4 die of breast cancer.
• With mammography screening of 1000 women over 10
  years 3 die of breast cancer.
• Without mammography screening of 1000 women over
  10 years 996 womendo not die of breast cancer.
• With mammography screening of 1000 women over 10
  years 997 women do not die of breast cancer.
• Out of 1000 women with mammography screening
  over 10 years one woman has a benefit as she does
  not die of breast cancer.
• Out of 1000 women with mammography screening
  over 10 years 999 have no benefit since they
  would not have died of breast cancer without
  mammography screening (996 women) or they died
  despite mammography screening (3 women).

Focus on the benefit
Focus on the lack of benefit
Or
14
Unbiased health information - presentation for
lay people / patients / consumers
Sabine Fischer Grafik, Halstenbek, 1999
15
. . . In conclusion
Future studies evaluating different formats of
communicating outcome data of preventive or
therapeutic interventions should include the
presentation of results both by emphasising
benefits and lack of benefits. Otherwise
patients or consumers do not have a fair chance
for informed decision making unbiased by framing
of data.
16
1 NHS The Patients Charta 1992 2 General
Medical Council Protecting patients, guiding
doctors. Seeking patients consent The ethical
considerations. London W1N6JE, 1999 3
http//www.bmgesundheit.de/vorhaben/reform/sgb/sgb
-g.htm Entwurf eines Gesetzes zur Reform der
gesetzlichen Krankenversicherung ab dem Jahr
2000 4 Coulter A. Evidence based patient
information. BMJ 1998317225-6 5 Fahey T,
Griffiths S, Peters TJ. Evidence based
purchasing understanding results of
clinicalTrials and systematic reviews. BMJ
19953111056-60 6 Nyström L, Larsson LG, Wall
S, Rutqvist LE, Andersson I, Bjurstam N,
Fagerberg G, Frisell J, Tabár L.An overview of
the Swedish randomised mammography trials total
mortality pattern and the representivity of the
study cohorts. J Med Screening 1996385-7