Immunotherapy for Hymenoptera venom allergy: too expensive for European health care?


We read with interest the commentary by Ruëff et al. [1] regarding our health economic appraisal of Pharmalgen® wasp and bee venom immunotherapy (VIT) for the National Institute for Health and Care Excellence (NICE) [2]. We are grateful to the authors for highlighting the surprising results of our analysis to the European allergy community and would like to respond to the critique of our approach.

Our main finding was that VIT is not considered to be cost-effective, when only the direct costs of fatal and nonfatal anaphylaxis outcomes are considered. The reasons for this are that for most people, wasp and bee stings are uncommon; nonfatal anaphylaxis is not an expensive outcome; and even in those with a prior systemic sting reaction, subsequent fatal sting anaphylaxis is rare [2]. So although VIT is effective for preventing systemic reactions to stings [3], one needs to apply the treatment to a very large number of individuals to create any significant cost savings via anaphylaxis prevention. In contrast, if one takes quality of life (QoL) improvement associated with VIT into consideration, VIT becomes cost-effective for even modest improvements in QoL. This is because the QoL benefit affects most patients, whether or not they have any further stings [4]. The practice implication is that when deciding which patient should have VIT, we should not simply assess for risk of anaphylaxis but also consider patients' anxiety about future reactions and the impact of this on their daily life. In patients who are not anxious about future stings, the cost-effectiveness of the treatment is in doubt [5]. This message from our recent assessments of the clinical- [3] and cost-effectiveness [2] of VIT has been expressed and interpreted with clarity in the resulting NICE guidance [6], which is similar to existing European guidance (Table 1).

Table 1. Indications for VIT
SourceIndications for initiating VIT

1. Positive skin prick or specific IgE test to insect venom allergen(s) and a systemic reaction to an insect sting, with respiratory/cardiovascular symptoms.

2. Positive skin prick or specific IgE test to insect venom allergen(s) and a systemic reaction with urticaria, plus other risk factors (e.g. high exposure risk, mastocytosis) or quality of life impairment.

NICE [6]

1. Presence of IgE-mediated bee or wasp allergy and a severe systemic reaction to a bee or wasp sting.

2. A moderate systemic reaction to a bee or wasp sting, together with one or more of: a raised baseline serum tryptase, a high risk of future stings or anxiety about future stings.

We respectfully disagree with the criticisms raised by Ruëff et al. [1]. They suggest the outcomes of our health economic analysis were predictable – this certainly seems so with the benefit of hindsight, but the outcomes could not have been so clearly quantified or relied upon without undertaking this rigorous piece of work. The authors suggest we should have assumed a QoL benefit for health economic assessment in the ‘base case scenario’. The NICE appraisal process, for which this research was commissioned, recommends use of a preference-based QoL measure that can be mapped to generic health states in the general population (e.g EQ-5D) [7]. Because we did not identify studies evaluating the effect of VIT on such measures, we could not quantify the expected QoL benefit. We did however take into consideration the likelihood of a QoL benefit in our analyses and conclusions, based on the evidence that disease-specific quality of life improves with treatment [4], and this is reflected in the NICE guidance – that VIT can be considered cost-effective in patients with anxiety related to future stings [6]. Finally the authors suggest that we should have made assumptions around the possibility of long-term neuro-disability after anaphylaxis, and cite data related to cardiac arrest outcomes to support this assumption. While we acknowledge that insect sting anaphylaxis may rarely lead to long-term morbidity, reports of such events in the literature are remarkably rare. To make assumptions about the frequency of such events, and the effect of VIT on their prevalence, would therefore be speculative and not acceptable within the rigorous framework of NICE health economic assessments.

We again thank the authors for highlighting these important data, and believe that we have consensus on the most important implication of this work – that in keeping with other areas of clinical medicine, we should have a patient-centred approach to decision-making when initiating VIT.

Conflict of interest

The authors declare no conflict of interest.