Table 5. Summary results of severity assessment by expert panel.

Animal model	Summary of scorers’ assessments	Final summary
Antifibrinolytics to promote blood clotting and control bleeding	The study of the shortest duration (2-6h) was generally assessed as moderate, with two scoring it as mild. The study lasting 24 hours had a spread of scores from mild, through moderate to severe, two scoring moderate on condition that bleeding was limited and analgesia given. Studies lasting 5–10 days: the first involved cutting the tail under anaesthesia and again produced a spread of scores from mild through moderate to severe. The study that involved rabbits having renal injuries inflicted under anaesthesia with 5–10 day follow up was mostly scored as severe.	Studies scored from mild to severe
Bisphosphonates to slow down /prevent bone loss	Most scorers felt the general model was of moderate severity, but rising to the severe category as the duration of the study increased and if no analgesia was used. One scorer consistently categorised procedures as less severe than other scorers for this study.	Mostly moderate, some severe
Corticosteroids to reduce intracranial pressure after traumatic brain injury	In general this model was scored as severe. In cases where animals were not anaesthetised or only lightly anaesthetised all scorers categorised procedures as severe. Endpoints were from 2h up to 30 days. Most endpoints were categorised as severe except for studies of the shortest duration which two scorers categorised as moderate as long as anaesthesia was used. For studies lasting 2–4 weeks all but one scorer categorised harms as severe.	Mostly severe
Tirilazad to protect brain tissue after stroke	The overall model was categorised as severe. Variations of the model were all categorised as severe, except for one study that reported post-operative analgesia; this was categorised as generally severe still, but potentially as less severe.	Severe
Antenatal corticosteroids to reduce neonatal mortality and morbidity in preterm babies	For the overall model scorers categorised harms to the mother as moderate to severe, depending on use of analgesia/ anaesthesia. For the overall model the scorers categorised harms to the foetus as generally severe, particularly if the foetus survived beyond birth and had no anaesthesia/ analgesia. For the administration of drugs to the mother scorers generally categorised this as moderate to mild. For the administration of drug to neonates after delivery, scorers categorised this as severe, particularly if neonates not anaesthetised. For administration of the drug to the foetus in utero scorers categorised the procedures to the foetus as severe if no anaesthesia was used (otherwise mild), and moderate to severe for the mother. For administration of the drug via the mother this was categorised as moderate to severe for the mother and moderate to severe for the foetus. For neonates having mechanical ventilation after delivery scorers categorised this as severe, or moderate if anaesthesia was used. In terms of endpoints, scorers categorised harms for those killed in utero and at delivery as moderate; as moderate to severe for deaths post-delivery (30 mins to 6 days), with increasing severity scores as time post-delivery increased. Harms to neonates left to die with no endpoint were categorised as severe by 5/6 scorers (animals found dead should automatically score as ‘severe’). Neonates having ventilation were scored as experiencing moderate to severe harms unless they were anaesthetised, with severity scores increasing with the amount of time spent on the ventilator.	Foetuses and neonates: mostly moderate to severe Mothers: mostly moderate to severe. Some maternal procedures scored as mild (e.g. maternal drug administration) but these were within overall models scored as moderate to severe.
Thrombolytics (tissue plasminogen activator, or tPA) to dissolve clot/ improve blood flow after stroke	In general scorers categorised the stroke model as severe. Some commented that if anaesthesia/ analgesia were used and duration of study brief, then studies might score less severely. Scorers categorised studies that induced stroke while animals were awake as severe. Most categorised relatively short term studies (up to 33h) that induced stroke under anaesthesia as severe. Most categorised studies that induced stroke under anaesthesia and then involved repeated MRI scans as severe.	Mostly severe