Dear Editor,
We would like to thank Dr Thornton for his comments and welcome this opportunity to expand on our original manuscript. In response to the pricing queries raised and in addition to the sensitivity analyses performed around unit costs of insulin and presented in our original manuscript, we are happy to provide here the results of a budget impact analysis performed using the input parameters suggested by Dr Thornton. In our original analysis, the cost-saving associated with stepwise addition of bolus insulin was found to be USD $1304 per intensifying patient over the trial period. If the impact hypoglycemia is ignored (by setting the cost per event to USD $0), then stepwise addition is associated with a cost-saving of USD $1013. Reducing the price of bolus insulin to USD $0.095 per unit, the price quoted by Dr Thornton, then stepwise addition is associated with a cost-saving of USD $936 per intensifying patient or USD $652 if hypoglycemia is ignored. In reference to the other points raised for discussion:
With regard to transparency of reporting, we are unsure as to the issue that Dr Thornton raises. Although the event rates used are taken directly from a clinical trial and were not previously published, we provided all rates in Table 1 of the original manuscript.
Dr Thornton highlights that the average wholesale price (AWP) of insulin is lower than the wholesale acquisition cost (WAC) that we utilized in the model. This is, of course, expected. However, given the range of health plans available and the differential rebates by area, one AWP was not representative of the US situation. For this reason, the WAC was used as indicated in the ISPOR guidelines [1]. The model set-up, of course, allows for a health plan payer to input their own price for all costs included in the model. We would also note that pricing was provided by Novo Nordisk for September–October 2013, the model and results presented being in 2013 USD and the AWP for October 2014 provided in the correspondence may not be representative of this time period.
The cost of hypoglycemia was always included in the original analysis. We feel that this is in line with the ISPOR guidelines, which state that the ‘costs of managing any side effects or complications should also be included in the cost of the current and new intervention’ (P.11) [1]. Sensitivity analyses were performed around the cost estimates assigned to hypoglycemia, with actual costs taken from the Canadian, German, Spanish and Swedish settings, again in line with Sullivan et al.Citation1. Our understanding of the ISPOR guidelines indicates that options to include or exclude cost items from the analysis should be included in the computer program, and we can confirm that this is the case in our model.
The comment made regarding market share is an interesting one. As the analysis focuses on the way in which a therapy is implemented and not on the therapy itself, the concept of market share is opaque. As with pricing, the ratio of stepwise to full-basal bolus use in the experimental scenario is fully customizable by the health plan payer; however, for the analysis presented there were no data available on which to base any assumptions about ‘market share’. For this reason, we undertook a direct comparison between the two implementations.
The results of the Full STEP trial provided further evidence that optimal clinical decision-making in diabetes involves consideration of both the best insulin product for the patient and the most beneficial method of insulin delivery. Under current economic conditions, however, the cost burden is also likely to be of substantial importance. Our manuscript was the first evaluation of this area, and we hope that the above data and responses prove beneficial to Dr Thornton and other readers that are interested in this important area of diabetes clinical practice.
Transparency
Declaration of financial/other relationships
The original study was funded by Novo Nordisk, from whom Ossian received consultancy fees.
References
- Sean D. Sullivan, Josephine A. Mauskopf, Federico Augustovski. Budget Impact Analysis—Principles of good practice: report of the ISPOR 2012 Budget Impact Analysis Good Practice II Task Force. Value Health 2014;2:5–14.