Is there any relationship between actual benefit and added value of drugs and pharmacovigilance alerts?

For each new medicine in France, after drug approval by the French Drug Agency (Agence Française de Sécurité Sanitaire des Produits de Santé), the French National Authority for Health [Haute Autorité de Santé (HAS)] issues an opinion for or against including it in the basket of products qualifying for reimbursement [1]. On one hand, HAS advises public decision makers on reimbursement and pricing. On the other, it informs practitioners about the real value of a new drug to individuals and society. The two main criteria used are the ‘Actual Benefit’ (AB) and the ‘Improvement in Actual Benefit’, also called ‘added value’ (IAB) of every drug.

The AB describes the value of a medicine in terms of the severity of the disease, its efficacy/safety ratio in real-life situations, and its public health benefit. The AB may be rated as ‘major’, ‘significant’, ‘moderate’, ‘low’ or ‘insufficient’ to justify reimbursement. The IAB describes the actual therapeutic advance (added value) of a medicine over other available forms of treatment. The assessment must take into account the contribution of the medicine to routine practice. The HAS committees may ask for additional studies to be carried out under real-life conditions. There are five levels of improvement: level I, major progress; level II, significant progress; level III, moderate progress; level IV, minor progress; level V, no progress. The IAB is used in economic assessments of healthcare products. It is taken into account by the Committee for Pricing and Reimbursement of Healthcare Products, which sets the price of reimbursable products. Medicines with no IAB are reimbursed only if they reduce treatment costs.

The aim of the present study was to investigate putative relationships between therapeutic advances (evaluated by using HAS criteria) and risk of adverse drug reactions [evaluated through pharmacovigilance (PV) alerts diffused by the French Drug Agency].

PV alerts concerning serious adverse drug reactions, warnings and inefficacy studies, diffused by the French Drug Agency, were analysed from 1 January 2006 until 24 November 2008. AB classes and IAB levels of drugs involved in these alerts were found using HAS documents and The Hospital National Centre of Information on the Drug site [1, 2]. First, we compared the frequency of PV alerts in each AB class with other AB classes and in each IAB level with other IAB levels. Then, we compared frequency of PV alerts between the drugs with low IAB levels (I–III) and those with high levels (IV–V). Finally, drugs with ‘insufficient’, ‘low’ or ‘moderate’ AB were compared with those with ‘significant’ or ‘major’ AB. Different classes were compared using the χ² test. Adjusted odds ratio (OR) and 95% confidence intervals (CI) were also calculated. The significance threshold was 5%.

Comparison of IAB levels showed that PV alerts were significantly more frequent at IAB II (OR 2.34, 95% CI 1.63, 3.35; P < 0.001) and IAB III (OR 2.12, 95% CI 1.51, 2.96; P < 0.001) levels and significantly less frequent at IAB V level (OR 0.46, 95% CI 0.35, 0.59; P < 0.001). No difference was found for IAB I and IV. Comparison between low (I–III) and high (IV–V) IAB levels showed that PV alerts were significantly more frequent in drugs belonging to I–III IAB levels (OR 2.13, 95% CI 1.64, 2.77; P < 0.001).

Comparison of AB classes showed (Table 1) significantly more frequent PV alerts in the ‘moderate’ class (OR 1.41, 95% CI 1.07, 1.85; P= 0.011) and less frequent PV alerts in the ‘insufficient’ class (OR 0.40, 95% CI 0.26, 0.60; P < 0.001). No difference was found between drugs with ‘insufficient’, ‘weak’ or ‘moderate’ AB and those with ‘important’ or ‘major’ AB.

Table 1.

Comparison of different drugs AB classes and IAB levels

Cases	Number	PV alerts	Controls	Number	PV alerts	P	OR (95% CI)
AB major	34	0	AB significant, moderate, low, insufficient	8472	448	0.168	NA
AB significant	5886	327	AB major, moderate, low, insufficient	2620	121	0.074	1.21 (0.98, 1.51)
AB moderate	992	69	AB major, significant, low, insufficient	7514	379	0.011*	1.41 (1.07, 1.85)*
AB low	509	26	AB major, significant, moderate, insufficient	7997	422	0.869	0.97 (0.63, 1.47)
AB insufficient	1085	26	AB major, significant, moderate, low	7421	422	<0.001*	0.40 (0.26, 0.60)*
IAB level I	136	11	IAB levels II–V	2313	277	0.171	0.65 (0.33, 1.25)
IAB level II	223	49	IAB levels I and III–V	2226	239	<0.001*	2.34 (1.63, 3.35)*
IAB level III	277	56	IAB levels I–II and IV–V	2172	232	<0.001*	2.12 (1.51, 2.96)*
IAB level IV	340	48	IAB levels I–III and V	2109	240	0.145	1.28 (0.90, 1.81)
IAB level V	1473	124	IAB levels I–IV	976	164	<0.001*	0.46 (0.35, 0.59)*
AB major & significant	5920	327	AB moderate, low, insufficient	2586	121	0.109	0.84 (0.67, 1.05)
IAB levels I-III	636	116	IAB levels IV–V	1813	172	<0.001*	2.13 (1.64, 2.77)*

^*Significant values

AB, actual benefit; CI, confidence interval; IAB, improvement of actual benefit; NA, not accurate; OR, odds ratio; PV, pharmacovigilance.

Our data (Table 1) show that PV alerts are more frequent with innovating drugs, i.e. belonging to I–III IAB. Several factors could explain this association. These drugs show a therapeutic advance over available forms of treatment, which leads them to be quickly integrated in routine practice, especially in serious diseases, while their rare adverse drug reactions are still unknown to the general population. Moreover, in serious diseases, the patients are weaker and at higher risk of adverse drug reactions. Such innovating drugs also benefit from a significant publicity programme from their manufacturers, which increases significantly their consumption during the first months of marketing. These data show that the more innovatory a drug is, the higher the risk of PV alerts. They underline the interest of reinforced PV follow-up for new innovatory drugs.

Competing interests

None declared.