Despite the plethora of studies on the impacts of access to runs on chickens, there is a paucity of information on the welfare and behavioural repertoire of hens raised in the deep litter houses with or without access to legume- or grass-based pasture. Therefore, this study aimed to evaluate the impact of access to grass or legume pastures by laying hens on behaviour, physiological responses and bacterial load. The study was conducted to evaluate the influence of exposure of egg-type chickens to runs on grass or legume pastures on their welfare and behaviours. The study involved the use of 240 ISA brown pullets from 12 weeks of age and and lasted for 48 weeks. The treatments were deep litter housing with grass-based pasture run (PG), deep litter housing with legume-based pasture run (PL) and deep litter housing without runs (LD) having 80 pullets with four replicates of twenty birds each. Behavioural observations of the hens in each pen were made at 52 weeks of age and tonic immobility was assessed by making the birds lie on their back with their head resting in a U-shaped wooden cradle. The measurements of the respiratory rate and rectal temperature of the hens were assessed at 1:00 p.m. at different laying phases. The gastrointestinal and egg bacterial counts were conducted at 60 weeks of age. Results revealed that the proportion of time spent eating was highest (p < 0.05) in the deep litter housing system, while the legume and grass pasture were similar. The hens spent most of their time standing and eating in the three treatments. However, the time spent standing in PL and PG was similar but significantly higher (p < 0.05) than in LD. Results on tonic immobility duration showed that the time spent by the hens in LD during the reaction was significantly longer than those of the PL and PG in the first, second and third phases of the experiments. However, the time spent by the hens in PL and PG was similar. The rectal temperatures of PL and PG birds were comparable and higher than those of LD during the second phase. On the other hand, there was no difference in the respiratory rate. Plasma triiodothyronine (T3) of the hens did not follow a consistent pattern. The bacterial count in the large intestine in LD and PL was similar but significantly (P < 0.05) higher than that of the PG. It was concluded that access to pasture influenced the behaviours of hens and that tonic immobility duration was shorter in the hens on the pasture, suggesting that access to pasture favoured hens’ welfare.