Rearing enrichment affects perching behaviour and bone strength in pullets and in free-range hens varying in range use patterns

Highlights

• •Pullets in floor pens were reared with different types of enrichment.
• •Perching structures increased pullet perching behaviour but not bone strength.
• •Novel objects during rearing increased perching in adult hens and tibia bone strength.
• •Individual variation in range use did not correlate with bone breaking strength.

Abstract

The high laying rates of domestic hens places significant strain on the calcium metabolism in their bones and shell glands. Behavioural and physical impacts of housing systems begin during pullet development, but exercise is still important throughout lay for maintaining bone strength. Free-range systems may improve bone strength for those hens that range more. This study assessed the effects of rearing enrichment for pullets housed in experimental floor pens. Perching and wing-flapping behaviour, and bone breaking load at the end of rearing were measured. Perching behaviour was observed throughout the flock cycle and bone breaking load assessed at the end of lay in hens that differed in range use. As predicted, pullets perched more when they were given perching structure enrichments compared with pullets exposed to varying novel objects or control pullets. However, this did not affect the breaking load of the pullets’ tibias. The novelty treatment pullets had the lowest humeral breaking load, but wing-flapping behaviour was observed equally across treatments. During lay, the novelty treatment hens perched the most and showed the highest tibial breaking load at the end of lay. Differing range use patterns were not correlated with the breaking load of the tibias or humeri. This study indicates that enriched rearing environments can have long-term impacts on hen behaviour and subsequent skeletal health, but that more time spent outside on the range does not impact bone strength. Further research should continue to explore the behavioural and physical effects of enrichments provided in floor-based laying hen rearing systems.

1. Introduction

Domestic laying hens have been selectively bred to lay an egg almost every day throughout their adult production life, which places significant strain on the calcium metabolism in their bones and shell gland (Whitehead & Fleming, 2000; Whitehead, 2004). When hens are housed in conventional cages with restricted movement, they can suffer from osteoporosis, lowering their bone strength and increasing the risk of broken bones (Whitehead & Fleming, 2000). In contrast, loose-housing systems such as tiered aviaries allow the hens more movement opportunities through jumping and perching on elevated structures (Campbell et al., 2016a; Campbell et al., 2016b). This movement increases mechanical loading on the bones, thus increasing bone mass and strength (Casey-Trott et al., 2017a; Regmi et al., 2015). Exercise is important throughout the laying period for maintaining bone and muscle strength, but both behavioural and physical impacts of housing design begin during pullet development (Casey-Trott et al., 2017b; Regmi et al., 2016).

Bone strength develops throughout the rearing period, as does the behavioural development of perching and movement on elevated surfaces. Perching behaviour in chicks can be seen after one week of age (Kozak et al., 2016) and pullets will be motivated to perch on elevated surfaces during the rearing period if they are provided the opportunity to do so (DePaoli et al., 2024; Heikkilä et al., 2006; Rentsch et al., 2023). This simultaneous development of physical strength and behavioural capability during rearing is important for the birds to then make use of elevated systems and/or perches when they are transferred to laying housing (Ali et al., 2019; Pullin et al., 2020, 2024). Providing elevated surfaces during rearing can increase the use of elevated structures during laying, increase use of the highest tiers in the system, and reduce potential injury risk associated with system collisions (Ali et al., 2019; Pullin et al., 2020, 2024). Studies have also shown bone parameters will be improved when pullets are reared in more complex environments where they can exercise. Bone measures such as density, mineral content, cross-sectional area and breaking strength have all been shown to be higher in pullets reared in aviaries compared with conventional cages (Casey-Trott et al., 2017a; Regmi et al., 2015). Improvements in bone mineral density have also been shown in loose-housed pullets with increasing degrees of elevation to navigate ranging from perches above the floor, to two-tiered aviaries (Makagon et al., 2024). In floor litter rearing barns, objects added as enrichments, including stationary items that pullets can jump onto may also result in increased perching behaviour and greater bone strength.

Another type of alternative housing system for laying hens that offers opportunities for exercise is the free-range system. This system provides outdoor access with variation in whether the indoor shed portion is floor-based or a tiered aviary system. The outdoor areas can be vast, where many studies have shown that individual hens in free-range systems will vary in the extent to which they use the outdoor range. Some hens will venture outside for many hours each day, whereas others will only visit the range occasionally, or not at all (e.g., Campbell et al., 2017; Larsen et al., 2017; Wurtz et al., 2023). Several studies have reported both positive and negative correlations between daily range usage and health and welfare parameters such as feather coverage, foot health, body weight, internal parasites, and gastrointestinal organ development (Rodriguez-Aurrekoetxea & Estevez, 2016; Sibanda et al., 2020a; Wurtz et al., 2023). This includes correlations in the subset of adult hens that are reported on in the current study such as greater feather coverage and lower body weight in the hens that ranged more (Bari et al., 2020a). In terms of bone health, free-range hens have shown higher tibial breaking strength and bone ash percentage compared to hens in conventional or enriched cages (Yilmaz Dikmen et al., 2016; Sharma et al., 2021). However, at the individual level no significant relationship was identified between range use and tibial parameters in hens at the end of lay (Kolakshyapati et al., 2019; Sibanda et al., 2020b). To date, there is still a limited number of studies with assessment between range use and bone health, including in hens from different rearing environments.

The aims of this study were to: (1) observe the perching behaviours of pullets reared with different types of environmental enrichments and (2) assess their tibia and humerus bone strength at the end of rearing. Then to (3) observe the perching behaviour of adult hens from the different enriched rearing environments and (4) assess bone strength of hens from both different rearing environments as well as different range use patterns. It was predicted that the pullets reared with perching structures would exhibit more perching behaviour and have stronger bones at the end of rearing. These birds were also predicted to perch more as adults with both those birds reared with perching structures, and those that ranged more predicted to have stronger tibia and humerus bones.

2. Materials and methods

2.1. Ethical statement

All research was approved by the University of New England Animal Ethics Committee (AEC17–092).

2.2. Pullets and housing (0–16 weeks of age)

The housing details for this study have been reported elsewhere in accompanying research on the same flock of birds (Bari, Cohen-Barnhouse and Campbell, 2020c, Bari et al., 2020b, Bari et al., 2020a; Campbell et al., 2020a; Campbell, Belson, Dyall, Lea, & Lee, 2022). The study was conducted using a total of 1700 Hy-Line® Brown day-old layer chicks that were housed within 9 floor pens (6.2 m L x 3.2 m W) across three separate rooms in the Rob Cumming Poultry Innovation Centre of the University of New England, Armidale, Australia. Each pen was visually isolated using shade cloth, with rice hulls as a litter substrate, one water nipple line and 4 round feeders providing ad libitum water and commercially mixed mash, respectively. The mash was formulated for different life stages of the bird where Table 1 presents a summary of its composition. Resources per bird were provided in accordance with the Australian Model Code of Practice for the Welfare of Animals – Domestic Poultry that was in effect at the time of the study (Primary Industries Standing Committee, 2002). Across 16 weeks, the pullets were reared under three separate enrichment treatments (see Fig. 1 for examples of the pens) with one treatment replicate per room, balanced for location. The treatments were: 1) ‘control’ - with no additional enrichments beyond the floor litter, 2) ‘novelty’ - with various novel objects added/removed approximately weekly (e.g., balls, bottles, brooms, buckets, disks, ropes, chain, cinder blocks, containers, dog toys, milk jugs, plastic children's toys, pipes, strings) and 3) ‘structural’ - five custom-designed black-coloured, metal perching apparatuses coated in a non-slip paint/sand covering (L, W, H = 0.60 m). They had two solid sides and one open-framed side forming an H-shaped structure that could be placed in different orientations with square perches (3.0 cm width) and a flat surface (Fig. 1). By 16 weeks of age the bird density was approximately 15 kg/m² or 9 pullets/m² with a total of 174–190 birds per pen. This variation was a result of both chick mortality across rearing (ranged from 0 – 5 chicks per pen; total 7, 8 and 3 for the treatments of control, novelty, and structural, respectively) and errors in counting the number of chicks that initially went into each pen. The Hy-Line® Brown recommendations for alternative system management (Hy-Line, 2016) were used for temperature and lighting guidance. The temperature, which was controlled via heating and mechanical ventilation, started at 35 °C for the day-old chicks and gradually reduced to 21 °C by 5 weeks of age. The lighting schedules started at 22 h light and decreased to 10 h light by 7 weeks of age, but the artificial LED lighting was maintained at a much higher lux of 100 because the pullets were destined for an outdoor environment. Chicks and pullets were vaccinated to meet both regulatory requirements and standard recommendations for the region (Newcastle disease, Marek’s disease, fowl pox, fowl cholera, egg drop syndrome, Mycoplasma gallisepticum, Mycoplasma synoviae, infectious bronchitis, infectious laryngotracheitis, and avian encephalomyelitis).

Table 1.

The differing composition of key ingredients in the diets as the birds developed. All feed was supplied by a commercial provider.

Composition % DM	Chick starter (0–7 wks)	Pullet grower (7–10 wks)	Pullet grower (10–16 wks)	Pre-lay (16–21 wks)	Lay (21–65 wks)
Dry matter	90.86	90.39	90.44	90.74	91.26
Protein	21.17	18.24	16.31	17.01	19.82
Fat	3.86	4.09	4.15	4.10	4.19
Fibre	2.21	3.72	3.69	3.14	2.90
Calcium	1.0	1.0	1.42	2.53	4.22
Av. phosphorus	0.61	0.43	0.45	0.48	0.46

Fig. 1.

Examples of the three rearing pen treatments (control, novelty, structural) when the pullets were 2 weeks old.

2.2.1. Pullet video recording and observations

Hikvision Network cameras (Model DS-2CD2232-I5 4 mm, Hikvision Australia, Mt Waverly, VIC) were installed to capture the indoor rearing pens during light hours at 2, 4, 11 and 14 weeks of age. Fig. 1 shows part of one pen per treatment when the pullets were 2 weeks old. Pullet video was observed by a single trained observer per two behaviours across two days at the four age points. Within each of the nine pens, the video was observed using scan sampling from 08:00 until 17:30 pausing every 30 min to count the number of birds that were perching (two feet latched onto a surface elevated off the floor), as well as the number of birds that were exhibiting wing-flapping (wings are outstretched and moving in an up and down motion; could occur while the bird was standing still, running, or jumping up onto an elevated surface). For the perching observations, the birds were counted on different areas of the pen that were elevated. Elevated surfaces were different among treatments as birds could perch on some of the enrichments provided in the novelty treatment and there were specifically provided metal perching apparatuses in the structural treatment. In the observation weeks, the novelty enrichments provided were as follows (placed on the ground unless stated as hanging): week 2 – broom, overturned plastic buckets, children's rubber balls; week 4 – cardboard egg flats, overturned plastic tubs and containers, cinder blocks; week 11 – hanging strings, plastic children's sand rollers, soda bottles, plastic watering cans, traffic cones; week 14 – flexible plastic piping, hanging straw broom, hanging plastic chain and milk bottles. The birds also perched on areas not intended for perching, but they were unable to be prevented from doing so. These areas included a thin sheet metal guard (approximately 15 cm height) in front of the external shed doors for keeping litter in (not present in the three centre pens but balanced across all treatments), the rims of the circular feeders, and the water line (feeders and water line equally present in all pens). For the wing-flapping observations, the video was played for a period of 30 s at each time point to capture the motion.

2.2.2. Pullet bone collection and breaking

At 15 weeks of age, 90 pullets were removed across the three rearing treatments (30 per treatment, 10 per pen replicate) for dissection across the same day. The birds were weighed using poultry-specific electronic hanging scales (BAT1; VEIT Electronics, Moravany, Czech Republic), killed via cervical dislocation and both the left tibia and humerus bones were excised, placed immediately into a cooler bin then stored with flesh at −20 °C until further processing (organs were also removed during these dissections as per (Campbell et al., 2020a). For breaking, the bones were placed at room temperature overnight to thaw and then defleshed by hand using a scalpel. Bones were wrapped in gauze-soaked saline to prevent drying, placed into fridge storage overnight and were then all broken across the following day. The bones were transported from the fridge to the bone breaking facility (both on the university campus) in a cooler bin where they were stored across the day of breaking. Both the tibias and humeri were each subjected to testing for breaking strength (N) with an Instron® electromechanical universal testing machine (Instron® Mechanical Testing Systems, 825 University Ave, Norwood, MA, United States) set up with a 300 KN load cell and 50 mm 3-point flexure test for the tibias, and a 30 mm 3-point flexure test for the humeri. The mechanical force was applied to the midpoint of the bones placed in consistent orientations.

2.3. Layer housing

At 16 weeks of age, 1386 pullets were weighed, crated, and transported (8.1 km) to the layer facility where they were distributed across 9 visually isolated, identical pens (4.8 m L x 3.6 m W; 154 hens/pen, approximate density: 9 birds/m²), in a single shed. The pullets were socially remixed within their treatment replicates but housed separately by rearing treatment. Each pen had two suspended round feeders, water nipples in a single line, perches, nest boxes and rice hulls as the floor substrate. The perches were four-tiered triangular racks (301 cm L with each wooden square perch (slightly rounded on top) separated by 30 cm from the ground and then from each other), and the large two-tiered nest boxes had wooden square (flat) perches at each tier that were 57 cm and 97 cm off the ground. Resources were provided to meet the Australian Model Code of Practice for the Welfare of Animals – Domestic Poultry (Primary Industries Standing Committee, 2002) except the perching space (perch rack and nest box perches combined) was 10 cm per bird due to logistical space restrictions within the pen. Birds also perched on the tops of the feeders and waterlines which was not included in the perching space calculations. Water and commercial mash (pre-lay, then lay) was provided ad libitum (Table 1). By 30 weeks of age, the LED lighting schedule had gradually reached 16 h L: 8 h dark with an average pen intensity of 10.0 (± 0.84 SE) lux (Lutron Light Meter, LX-112,850; Lutron Electronic Enterprise CO., Ltd, Taipei, Taiwan) at birds’ eye height measured from 3 pen locations (front, middle, back) when the pop-holes were closed. This lux was the highest that could be achieved with the shed lighting system. The shed was fan-ventilated with no other environmental control. Each indoor pen was connected to a separate outdoor range area accessible via two pop-hole openings (18 cm W x 36 cm H) for daytime range access from 09:15 until after sunset. The pop-hole openings were up to approximately 30 cm height off the ground, depending on the depth and movement of the litter inside the pen. The pop-holes first opened at 25 weeks of age (May 2018) with approximately 9 hours of available daily ranging time across winter and approximately 11 hours of available ranging time from October 2018 onwards. The range area was a straight 31 m run of predominantly grassed/dirt area with no trees or shelters. All food, water, shelter, nest box, and perch resources were located inside the shed.

2.3.1. Layer video recording and observations

The same video cameras as per pullet rearing were installed to capture the indoor pens at the layer facility at 26, 31, 41, 50, 60, and 64 weeks of age. Approximately one week of video was recorded at each age point with two days selected for observations based on a full set of recordings across all the pens with no missing video resulting from technical failures. The selected days per age week were not consecutive but were the same for each pen. A trained observer then counted the number of hens perching using scan sampling point counts every 30 mins from 09:30 (pop-holes opened at 09:15) until 17:30 (just before sunset) or until 19:30 from 50 weeks onwards following daylight-saving time change. In total there were 2052 observations of perching across the flock cycle (17 observations x 2 days x 6 age points x 9 pens = 1836 + an additional 216 observations following daylight saving). Point counts of birds perching inside included birds on the perch rack, nest box perches, water line, and feeder rims (Fig. 2). The observer who watched the hen video was different from the observer who watched the pullet video, this observer was able to be blinded from rearing treatment given all layer pens looked the same.

Fig. 2.

One of the indoor pens at the free-range laying hen facility where hens are seen perching on the perch rack, nest box perches, feeder rims and the water line.

2.3.2. Layer radio-frequency identification (RFID) tracking (16–64 weeks of age), and bone breaking

All birds wore an adjustable leg band (Roxan Developments Ltd, Selkirk, Scotland) that contained a glued microchip (Trovan® Unique ID 100 (FDX-A): operating frequency 128 kHz) to track their movement in and out of the range pop-holes via radio-frequency identification (RFID) systems. The RFID systems were designed and supported by Microchips Australia Pty Ltd (Keysborough, VIC, Australia) with equipment developed and manufactured by Dorset Identification B.V. (Aalten, The Netherlands) using Trovan® technology. The system recorded the date and time of each tagged bird passing through and in which direction (onto the range, or into the pen) with a precision of 0.024 s (maximum detection velocity 9.3 m/s). Individual ranging data were collected daily from 25 until 65 weeks of age. The RFID data from 56 until 64 weeks of age (54 days of data) were used to select a sample of birds from each pen to test bone breaking strength in hens that exhibited the low and high extremes of time outside on the range. At 64 weeks of age, a total of 93 hens were selected across all of the 9 pens that were categorized as ‘indoor’ hens, accessing the outdoors on one or zero of the 54 days, and 104 ‘high outdoor’ hens, accessing the range for 54 of 54 days for 5 h 12 mins to 9 h. There was variation in ranging between pens, thus the ‘high outdoor’ birds were selected as the highest for their pen with the balancing of hen selection across pens within treatments where possible.

These selected hens were transported (pen by pen) to a separate post-mortem facility located 5.5 km from the free-range facility. Hens had been weighed 4 to 5 days prior using poultry specific hanging scales as part of separate datasets on body weight and external welfare scoring (Bari et al., 2020b, Bari et al., 2020a). Upon arrival at the post-mortem facility, hens were killed using CO₂ then opened up immediately for post-mortem examination as detailed in Bari et al. (2020a). The left tibia and humerus bones were excised and stored with flesh at −20 °C until further processing. Later, the bones were placed at room temperature overnight for thawing and then defleshed by hand using a scalpel. They were wrapped in saline-soaked gauze, placed in the fridge overnight, then broken the following day. As for the pullet samples, the bones were transported from the fridge to the bone breaking facility (both on the university campus) in a cooler bin where they were stored across the day of breaking. The same Instron® machine with a 300 kN load cell used for the pullet bones was used to test for breaking load. A 30 mm 3-point flexure test was applied for the tibias and humeri with the mechanical force applied to the midpoint of the bones placed in consistent orientations.

2.4. Data and statistical analyses

All analyses were conducted in JMP 17.2.0 (SAS Institute, Cary, NC, USA) with α = 0.05. Data transformations were applied as necessary with the raw data presented in the figures.

The number of pullets perching inside each pen across the 4 age points (2, 4, 11, 14 weeks) were compiled (n = 1440 data points: 20 daily observations x 2 days x 4 age points x 3 treatments x 3 pens). These were summed across the different observed elevated areas in the pens and converted to the proportion of pullets per pen, accounting for any mortality across rearing. The number of pullets observed wing-flapping were also converted to proportional values. The proportional values for perching and wing-flapping were averaged across the 20 daily scan observations to create a final dataset of 72 values per behaviour (2 days x 3 treatments x 3 pens x 4 age points). A general linear mixed model was applied to the logit-transformed proportional data with treatment, age, and their interaction as fixed effects, and pen nested within treatment as a random effect. Where significant differences were present, post-hoc Student’s t-tests were applied to the least squares means. The raw data on the different elevated surfaces that the pullets perched on in the different treatment pens are visually presented.

The breaking load of the humerus and tibia for each pullet (n = 90) was analysed using two separate general linear mixed models with treatment as a fixed effect, pen nested within treatment as a random effect and accounting for body weight in the model. Where significant differences were present, post-hoc Student’s t-tests were applied to the least squares means.

The number of adult hens perching inside each pen across the 6 age points (26, 31, 41, 50, 60, and 64 weeks of age) were converted to the proportion of hens per pen, accounting for any mortality. The proportional values were averaged across the 17–21 daily scan points, so the final dataset was comprised of 108 data points (3 treatments x 3 pens x 6 age points x 2 observation days). A general linear mixed model was applied to the logit-transformed proportional data with treatment, age, and their interaction as fixed effects, and pen nested within treatment as a random effect. Where significant differences were present, post-hoc Student’s t-tests were applied to the least squares means.

The breaking load of the humerus and tibia for each adult hen (n = 197) was analysed using two separate general linear mixed models with treatment and range use, including their interaction as fixed effects and pen nested within treatment as a random effect while accounting for body weight in the model. Where significant differences were present, post-hoc Student’s t-tests were applied to the least squares means.

3. Results

There was a significant effect of rearing treatment (F_(2,6) = 33.54, P = 0.0006) and age (F_(3,54) = 67.61, P < 0.0001) but not their interaction (F_(6,54) = 1.94, P = 0.09) for the number of pullets perching. The most pullets were observed perching in the structural treatment, followed by novelty, and the least in the control (Fig. 3). The least perching was observed when the chicks were 2 weeks old, with equal proportions of birds perching observed at the remaining observed age points (Fig. 3). In all pens, pullets were observed perching on the water lines and feeder rims across rearing, as well as the metal door guard but this mostly occurred at 4 weeks of age (Fig. 4). The pullets in the enriched pens also perched on the varying novel objects, particularly at the younger ages when they were smaller and could sit on top of the enrichments (Fig. 4). There were no significant differences across rearing treatments (F_(2,6) = 0.05, P = 0.95), age (F_(3,54) = 1.74, P = 0.17), nor their interaction (F_(6,54) = 0.41, P = 0.87) for the proportion of birds that were observed wing-flapping (Fig. 3).

Fig. 3.

Perching and wing-flapping in pullets at four age points (2, 4, 11, 14 weeks of age) across three rearing enrichment treatments (control, novelty, structural). Means (±SEM) of the raw data are presented with analyses conducted on transformed data. The structural hens perched the most but there were no treatment effects for wing-flapping.

Fig. 4.

The mean (±SEM) proportion of pullets perching on the different types of elevated surfaces in the pens at four age points (2, 4, 11, 14 weeks of age) across three rearing enrichment treatments (control, novelty, structural). The enrichments were only present in the ‘novelty’ treatment, and the structures only present in the ‘structural treatment’. The feeders, water, and metal door guard (only present in 6 of the 9 pens) were unintended perching surfaces but the pullets could not be prevented from using them. Raw data are presented.

There was no significant effect of rearing treatment on the breaking load of the pullets’ tibias (F_(2,6.0) = 0.08, P = 0.93, Fig. 5). There was, however, a significant effect of treatment on the breaking load of the pullets’ humeri (F_(2,5.65) = 5.70, P = 0.04) with the novelty hens showing the lowest breaking load (Fig. 5).

Fig. 5.

The mean (±SEM) breaking load (N) of pullet humerus and tibia bones across three rearing treatments (control, novelty, structural) in pullets at 15 weeks of age. Dissimilar connecting letters indicate significant treatment differences for the humeri only (P < 0.05). Raw data are presented. Note: different settings were used on the breaking machine for the humerus and tibia bones where the humerus bones had a 30 mm 3-point flexure test applied, and the tibias a 50 mm 3-point flexure test. These differences would result in a higher breaking load for the humeri.

There was a significant effect of rearing treatment (F_(2,6) = 6.90, P = 0.03) and age (F_(5,84) = 74.09, P < 0.0001) but not their interaction (F_(10,83) = 1.48, P = 0.16) on the number of adult hens perching inside. More hens from the novelty treatment were observed perching (Fig. 6). Across all treatments, the most perching was observed around mid-lay (31 and 41 weeks) and the least at the end of lay (64 weeks) (Fig. 6).

Fig. 6.

The proportion of adult laying hens observed perching at six age points (26, 31, 41, 50, 60, 64 weeks of age) across three rearing enrichment treatments (control, novelty, structural). Mean (±SEM) raw values presented with analyses conducted on transformed data. Across all ages combined, analyses showed the novelty hens perched the most (P = 0.03).

There was a significant effect of rearing treatment on the breaking load of the tibias in the adult hens (F_(2,191) = 17.49, P < 0.0001) with the novelty hens showing the highest breaking load (Fig. 7). There was no significant effect of range use (F_(1,191) = 0.66, P = 0.42), nor the interaction between treatment and range use (F_(2,191) = 0.69, P = 0.50). There was no significant effect of treatment (F_(2,5.67) = 0.23, P = 0.80), range use (F_(1,188.9) = 0.02, P = 0.89), nor their interaction (F_(2,188.2) = 0.15, P = 0.86) on the breaking load of the humerus bones (Fig. 7).

Fig. 7.

The breaking load (N) of laying hen humerus and tibia bones across three rearing treatments (control, novelty, structural) and two range use patterns (indoor, high outdoor) in hens at 65 weeks of age. Mean (± SEM) raw values are presented. Across both range use patterns combined, the analyses showed the novelty hens had the highest breaking load in the tibia bones only (P < 0.0001). There were no effects of range use on humerus or tibia bones.

4. Discussion

This study assessed the effects of rearing enrichment for pullets housed in experimental floor pens on their perching and wing-flapping behaviour, and bone breaking load at the end of rearing. As predicted, pullets perched more when they were given perching structures, but this did not affect the breaking load of the pullet tibias. The pullets from the novelty treatment had the lowest breaking load of their humeri, but wing-flapping behaviour was observed equally across all treatments. In the laying facility, the novelty hens were observed perching the most across the flock cycle and also showed the highest tibial breaking load at the end of lay. Differing range use patterns were not correlated with the breaking load of the tibias or humeri. This study indicates that enriched rearing environments can have long-term impacts on hen behaviour and subsequent skeletal health, but that more time spent outside on the range does not impact bone strength.

The increased perching in the structural rearing treatment pens was expected given the custom-designed H-shaped metal structures were intended to facilitate perching (as well as spatial navigation) behaviour. However, pullets could also perch on some of the other types of enrichments provided in the novelty treatment, particularly when they were younger where multiple birds could sit on top of the enrichments. Elevated flat surfaces were utilised such as those provided by cinder blocks and overturned plastic containers. As the pullets aged, the novel items became more directed toward stimulating pecking behaviours. Furthermore, birds in all treatments found elevated surfaces to perch on (feeders, water line), despite the control treatment being intended to be floor rearing only. This highlights the motivation of pullets to access elevated surfaces, and thus, the importance of height complexity to the birds. It may be that the degree of perching was sufficient for all pullets to develop their bones. Loading may have also come from other forms of exercise in the pens such as running behaviour as was observed in chicks across all pens in (at least) the first six weeks of age (Campbell, Belson, Dyall, Lea, & Lee, 2022). In contrast, no behavioural differences were observed in wing-flapping behaviour, yet the novelty pullets had the lowest breaking load in their humeri. It is unclear why this difference was observed. There may have been behavioural variation that occurred at other age points during rearing not observed, or other behaviours such as jumping and balancing that utilised the wings but were not captured in the wing-flapping ethogram description. It is possible that the same individuals were being observed across the scans if certain birds exhibited wing-flapping more frequently. Overall, the novelty pullets may have engaged in more exploratory pecking behaviours as a result of the enrichments provided to them such as strings, hanging chains, and other novel objects that did not utilise the wings.

In contrast with predictions and the patterns documented during rearing, the novelty hens were observed perching more than the structural hens in the laying facility. This aligns with observations of nest box use in the same hens as reported on in Bari et al. (2020c). The novelty hens used the elevated nest boxes the most, thus exhibiting lower floor laying across the flock cycle (Bari et al., 2020c). However, there were no treatment differences in egg production rates nor egg weights across the flock cycle (Bari et al., 2020c). The novelty hens were also observed perching the most in the first week following transfer to the laying hen facility in observations that were directed at capturing adaptation to the new environment (Campbell et al., 2020a). This was a pattern which the current study found continued throughout the flock cycle. It is possible the novelty hens showed better adaptation to the new environment as a result of the frequent environmental changes they were exposed to during rearing. The structural hens may have also become accustomed to the size, material, and shape of the metal perches and flat metal surfaces experienced during rearing, which differed from the wooden perches in the layer pens. Previous studies have shown that the behavioural effects of rearing on layer facility system use have varying degrees of persistence. Some effects dissipate across the first month as all hens adjust to their new surroundings (Brantsæter et al., 2016; Pullin et al., 2024), whereas other early life differences last longer into the lay cycle (Ali et al., 2019; Pullin et al., 2024). The structural hens did show the most ranging from around mid-lay onwards (Campbell et al., 2020b). It is possible the structural treatment had a greater impact on spatial navigation capabilities that resulted in increased outdoor access rather than sustained perching differences. The structures were designed with opaque sides intended to improve spatial navigation abilities of the chicks to locate occluded resources (Freire et al., 2004). With all the food/water/shelter/roosting resources located inside the shed, any hens that went outside would need to navigate back to access these.

The increased perching in the novelty hens aligns with these birds showing the highest tibial breaking load. However, there were no effects of rearing treatments on keel bone damage (Bari et al., 2020a) which can also be impacted by perch use (DePaoli et al., 2024). In computed tomography (CT) scanning of the bird carcasses as reported on in Bari et al. (2020a), there was a trend for an effect of rearing treatments on bone mass. The novelty hens had higher bone mass than the control hens, with structural treatment birds showing no difference from either. The findings from the current study and previously reported results on the same birds confirm that in addition to rearing effects, exercise in the laying phase is important for maintaining bone health as the hens age (Casey-Trott et al., 2017b; Regmi et al., 2016). If there had been more accessible perching space than the approximate 10 cm /bird, including more space at the higher and perhaps more preferred levels (Campbell et al., 2016c), it is possible that more perching may have been observed across all pens regardless of rearing environment.

Range use differences in individual hens has been shown to be correlated with varying health and welfare parameters, including in the birds specifically used in this study (Bari et al., 2020a). The outdoor hens showed lower body weight, better feather coverage, fewer comb wounds, and reduced toenail length compared to the indoor hens (Bari et al., 2020a). No differences were found in keel bone damage between the ranging groups (Bari et al., 2020a). CT scanning of bird carcasses showed the high outdoor hens had lower body fat and muscle than indoor hens but there were no differences in bone mass (Bari et al., 2020a). The finding of no differences in tibial bone breaking load (nor of humeral breaking load) aligns with Kolakshyapati and colleagues (2019) and Sibanda and colleagues (2020b) who also found no impact of individual range usage on tibial health using a wider range of bone health measures. It is unclear if other measures on the bones in the current study such as mineral density or cortical thickness would have revealed differences. While previous studies show free-range systems improve bone health relative to cage (conventional and enriched) systems (Yilmaz Dikmen et al., 2016; Sharma et al., 2021), it may be that hens inside are showing similar degrees of mechanical loading exercise through standing, jumping, and perching. Thus, time outside does not confer additional benefits to the skeleton, aligned with the findings on bone mass and keel bone damage of the same birds, as reported in Bari et al. (2020a). At a flock level, more ranging time has been shown to be negatively correlated with time spent on the upper tier in an aviary free-range system (Sibanda et al., 2020c). Hens that remained inside may have spent more time up on the perches relative to those out ranging. The RFID system in the current study did not allow individual tracking of egg laying. Previous research has shown differences in production rates between groups of hens differentiated by their range use patterns (Sibanda et al., 2020d) and negative relationships between the heritability of range use and egg laying traits at the individual bird level (Icken et al., 2008). Monitoring range use, egg production/quality, and skeletal health of individual birds may provide a more complete understanding of the interplay between these factors.

In conclusion, this study showed perching during rearing was affected by the types of enrichments provided in experimental floor pens, but tibial bone breaking load was not increased as predicted. The rearing enrichment impacts continued throughout the flock cycle but were not in line with predictions of more perching in those birds provided perching structures during rearing. Neither tibial nor humeral bone strength was correlated with individual range usage, thus indoor system usage may provide the same degree of exercise to birds that do not range. Further research should continue to explore the behavioural and physical effects of different types of enrichments that can be provided in floor-based laying hen rearing systems.

Funding

Funding: This research was funded by Poultry Hub Australia, grant number 2017–20.

Ethical statement

All research was approved by the University of New England Animal Ethics Committee (AEC17-092).

CRediT authorship contribution statement

Dana L.M. Campbell: Writing – review & editing, Writing – original draft, Visualization, Supervision, Resources, Project administration, Methodology, Investigation, Funding acquisition, Formal analysis, Data curation, Conceptualization. Md Saiful Bari: Writing – review & editing, Methodology, Investigation, Data curation. Andrew M. Cohen-Barnhouse: Writing – review & editing, Supervision, Resources, Methodology, Investigation, Conceptualization. Prafulla Regmi: Writing – review & editing, Resources, Methodology, Investigation, Conceptualization.

Declaration of competing interest

The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Dana Campbell reports financial support was provided by Poultry Hub Australia. If there are other authors, they declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.