We are pleased to respond to the questions raised by Fahey et
al[1][2] in connection with our use of the Markov model to estimate the
years of life gained without a heart attack or stroke among people aged 55
years or older who take the polypill each day.[3] They ask for input
parameters. These are specified in the paper, though not in a simple list.
For clarity, we set these out in table 1 on our website - www.smd.qmul.ac.uk/wolfson/polypillpapers/markovmodelling (tables cannot
be submitted in rapid responses). The Markov model had two states: (i)
alive not having had an IHD event or stroke, and (ii) dead from any cause,
or alive having had an IHD event or stroke.
Their second query is whether a sensitivity analysis was carried out.
Such an analysis shows that our estimates were reasonably precise, as
shown in table 2 on our website, where we estimate the benefits of the
Polypill based on the upper and lower 95% confidence intervals of the
relative risk reductions, and show these together with the original point
estimates. In 100 men we estimated that 37 would benefit with an everage
of 12 event free years of life, the estimates using the confidence
intervals are 34 to 39 and 11-12 respectively. In 100 women we estimated
that 35 would benefit with an everage of 11 event free years of life, the
estimates using the confidence intervals are 32 to 37 and 11-11
respectively.
Nicholas J Wald, Malcolm R Law, Joan K Morris
Wolfson Institute of Preventive Medicine
Barts and the London School of Medicine and Dentistry
Charterhouse Square
London EC1M 6BQ
(n.j.wald@qmul.ac.uk)
References
1. Fahey T, Montgomery A, Ben Shlomo Y. Modelling and assumptions
need clarification. Bmj.bmjjournals.com/cgi/eletters/326/7404/1419
(20/02/2004)
2. Fahey T, Montgomery AA, Ben Shlomo Y. Methodological issues remain
unanswered. Bmj.bmjjournals.com/cgi/eletters/326/7404/1419 (20/02/2004)
3. Wald NJ, Law MR. A strategy to reduce cardiovascular disease by
more than 80%. BMJ 2003;326:1419
Competing interests:
See original paper submitted
Competing interests:
No competing interests
08 March 2004
Nicholas J Wald
Professor
Malcolm R Law, Joan K Morris
Wolfson Institute of Preventive Medicine, Barts and the London, Charterhouse Sq, London EC1M 6BQ
Rapid Response:
Methodological issues remain unanswered - Authors' response
We are pleased to respond to the questions raised by Fahey et
al[1][2] in connection with our use of the Markov model to estimate the
years of life gained without a heart attack or stroke among people aged 55
years or older who take the polypill each day.[3] They ask for input
parameters. These are specified in the paper, though not in a simple list.
For clarity, we set these out in table 1 on our website -
www.smd.qmul.ac.uk/wolfson/polypillpapers/markovmodelling (tables cannot
be submitted in rapid responses). The Markov model had two states: (i)
alive not having had an IHD event or stroke, and (ii) dead from any cause,
or alive having had an IHD event or stroke.
Their second query is whether a sensitivity analysis was carried out.
Such an analysis shows that our estimates were reasonably precise, as
shown in table 2 on our website, where we estimate the benefits of the
Polypill based on the upper and lower 95% confidence intervals of the
relative risk reductions, and show these together with the original point
estimates. In 100 men we estimated that 37 would benefit with an everage
of 12 event free years of life, the estimates using the confidence
intervals are 34 to 39 and 11-12 respectively. In 100 women we estimated
that 35 would benefit with an everage of 11 event free years of life, the
estimates using the confidence intervals are 32 to 37 and 11-11
respectively.
Nicholas J Wald, Malcolm R Law, Joan K Morris
Wolfson Institute of Preventive Medicine
Barts and the London School of Medicine and Dentistry
Charterhouse Square
London EC1M 6BQ
(n.j.wald@qmul.ac.uk)
References
1. Fahey T, Montgomery A, Ben Shlomo Y. Modelling and assumptions
need clarification. Bmj.bmjjournals.com/cgi/eletters/326/7404/1419
(20/02/2004)
2. Fahey T, Montgomery AA, Ben Shlomo Y. Methodological issues remain
unanswered. Bmj.bmjjournals.com/cgi/eletters/326/7404/1419 (20/02/2004)
3. Wald NJ, Law MR. A strategy to reduce cardiovascular disease by
more than 80%. BMJ 2003;326:1419
Competing interests:
See original paper submitted
Competing interests: No competing interests