Health technology assessment in osteoporosis

Abstract

We review the various aspects of health technology assessment in osteoporosis, including epidemiology and burden of disease, and assessment of the cost-effectiveness of recent advances in the treatment of osteoporosis and the prevention of fracture, in the context of the allocation of health-care resources by decision makers in osteoporosis. This article was prepared on the basis of a symposium held by the Belgian Bone Club and the discussions surrounding that meeting and is based on a review and critical appraisal of the literature. Epidemiological studies confirm the immense burden of osteoporotic fractures for patients and society, with lifetime risks of any fracture of the hip, spine, and forearm of around 40 % for women and 13 % for men. The economic impact is also large; for example, Europe's six largest countries spent €31 billion on osteoporotic fractures in 2010. Moreover, the burden is expected to increase in the future with demographic changes and increasing life expectancy. Recent advances in the management of osteoporosis include novel treatments, better fracture-risk assessment notably via fracture risk algorithms, and improved adherence to medication. Economic evaluation can inform decision makers in health care on the cost-effectiveness of the various interventions. Cost-effectiveness analyses suggest that the recent advances in the prevention and treatment of osteoporosis may constitute an efficient basis for the allocation of scarce health-care resources. In summary, health technology assessment is increasingly used in the field of osteoporosis and could be very useful to help decision makers efficiently allocate health-care resources.

Fig. 1

Cost-effectiveness plane. The difference in quality-adjusted life-years between intervention A and comparator O is represented on the horizontal axis and the difference in cost, on the vertical axis. The slope of the line between intervention A and comparator O is the incremental cost-effectiveness ratio. If A is located in quadrant II or IV, it is dominant (more effective and less costly than comparator O); in quadrant IV, intervention A is less effective and more costly than O. In quadrant I, A is more effective but more costly; and in III, it is less effective and less costly. The choice will depend on the cost-effectiveness threshold that represents the maximum amount the decision maker is willing to pay for a unit of effectiveness. Interventions that fall below the cost-effective threshold would be deemed cost-effective

Fig. 2

Example of a cost-effectiveness acceptability curve. This graph shows the probability of an osteoporotic treatment being cost-effective compared with no treatment in patients aged 70 years with prevalent vertebral fractures as a function of the decision maker’s willingness-to-pay per one quality-adjusted life-year (QALY) [108]. The curve was estimated from probabilistic sensitivity analyses where most parameters (such as therapeutic effect, fracture risk, cost, and disutility) were assigned a probability distribution (e.g., normal or uniform distribution) and values from each distribution were randomly selected during a predefined number of simulations

Fig. 3

Intervention thresholds in Belgium [119] (copyright permission from Springer)

Fig. 4

Maximum yearly cost (in euros) for an adherence-enhancing intervention to be considered cost-effective [data from 36, 37, 64]. For Sweden, improvement in medication adherence at 25 % should be read at 30 %. In Ireland, a longer refill gap period (9 weeks) was selected to define persistence resulting in higher base-case adherence levels