Mature drugs (i.e. the ones that have lost patent protection) are an ideal domain for generating savings in public health expenditures. Generic/biosimilar competition and gradual obsolescence justify the need of purchasing mature medicines at low prices. To this end, public procurement is a key function for establishing competitive markets and granting sustainable and reliable supplies to the NHS through a variety of possible practices in structuring tenders. The objective of this study is to assess the relationship between of a vast array of procurement practices on three dimensions of public procurement’s performance: i) degree of competition, ii) ability to obtain discounts and generate savings and iii) lowering the number of non-awarded tenders.
We analyzed historical data (2010-2020) on public tenders in Italy referred to five mature pharmaceutical markets (defined at the ATC-4 level): anti-TNF alpha, heparin group, opioid anesthetics, proton pump inhibitors and oncological chemotherapies (ATC-4 of taxanes and platinum compounds). The three dependent variables were defined at the level of the individual lot or sub-lot: number of offers (proxy for the degree of competition), discount awarded and probability of not awarding the focal lot. The statistical analysis was performed with different models, namely negative binomial, two-steps fractional response model and logit, according to the dependent variable. The explanatory variables representing procurement practices were: size of the lot (demanded quantity), minimum discount required, type of lot (simple -only products with same active ingredient, formulation and dosage- or unified -more formulations and doses demanded for the same active ingredient), type of procedure (open, restricted, negotiated or framework agreement) and the level of centralization (hospital, region/network). For biologics, we also added a variable indicating whether the focal lot was dedicated to originators, biosimilars or both. Robustness checks through alternative models and specifications were applied.
Competition: centralization and the use of unified lots are in most cases associated with a lower number of offers received, while open procedures are associated with a higher number of offers. Mixed evidence was found with respect to the minimum required discount. For biologics, the inclusion of both originators and biosimilars in the same lot is associated with a higher number of offers.
Discount: in most cases, the minimum required discount crowds-out the extra discount, disincentivizing competition. However, it is positively associated with the number of offers. Centralization, type of procedure and type of lot do not show significant relationships with the total discount awarded.
Probability of not awarding a lot due to the lack of offers: the most significant association is with the minimum required discount which is associated with higher probability of not awarding the lot (however the magnitude of the effect varies across markets). In the case of heparin group and of chemotherapies, the use of unified lots is associated with higher risk of not awarding the lot.
The paper shows that: i) the impact of procurement practices varies considerably across markets, ii) there is a trade-off in the tactic use of the minimum required discount because it can stimulate more offers (in some markets) but it also reduces the space for actual competition on the extra-discount, with opposite consequences on the total discount awarded; however, iii) a higher minimum discount can also increase the probability of not receiving offers for the lot, iv) centralization does not show positive impacts on competition (where it sometimes show a negative effect), discounts and probability of not awarding the lot, thus casting doubts on its usefulness as a cost-containment tool (perhaps it should be read as a more convenient organizational arrangement than an efficiency-generating factor), v) for mature biologics, the best performance (more competition, higher discounts) is reached when both originators and biosimilars are allowed to bid. Universal recipes do not exist to grant efficient and sustainable procurement of mature drugs, and each market has peculiarities that should be taken into account by procurement agencies. Some practices, moreover, do not show the expected effects on competition and prices, while some may actually obtain perverse unintended effects is used without anticipating the complex set of consequences (minimum required discounts)
Authors: Patrizio Armeni*, Francesco Costa*, Claudio Jommi^, Monica Otto#
* Associate Professor of Practice, CERGAS - SDA Bocconi School of Management
^ Professor of Practice, CERGAS - SDA Bocconi School of Management
# Lecturer - Department of Social and Political Sciences, Bocconi University
Speaker: Patrizio Armeni
Link zoom: https://unibocconi-it.zoom.us/j/94296222653
For those willing to participate in person, please write to email@example.com before Friday, April 15th.