Assessment of the Effects of Stocking Density on Laying Hens Raised in Colony Cages: Part II—Egg Production, Egg Quality, and Welfare Parameters

Abstract

Stocking density is one of the major concerns in all production systems, which is why lawmakers, retailers, and consumers are increasingly concerned about this issue and its relation to animal welfare. The aim of this study was to identify if stocking density had an effect on white egg layer production, egg quality, bird health, and welfare parameters. For this study, five stocking densities were evaluated in colony cages: 1342 cm²/hen, 897 cm²/hen, 671 cm²/hen, 535 cm²/hen, and 445 cm²/hen. Egg production and physical egg quality were measured. Hen health and welfare parameters including corticosterone levels, H/L, oxidative stress pathways, jejunum tight junction protein expression, bone health, gut histology, body condition, and cytokine expression were evaluated. The results from this indicated that higher stocking densities resulted in lower production but larger eggs, while feed efficiency remained unaffected. Furthermore, physical egg quality parameters also remained unaffected. When stress and welfare parameters were analyzed, this research identified that decreasing stocking density did not affect jejunum oxidative stress pathways, pro-inflammatory cytokine expression, bone health, or intestinal health. At the final sampling period (69 weeks), the highest stocking density demonstrated higher corticosterone concentrations and IL-10 expression compared to the lowest stocking density. Furthermore, feather scores were found to be poorer as density increased. Finally, the highest density had higher Hansen’s test scores compared to other densities, which indicates a greater fear response. In conclusion, it appears that decreasing stocking density may provide some benefits in production and welfare to commercial egg layers, particularly at the end of the laying cycle.

1. Introduction

Intensification of laying hen farming has led to increased egg numbers, decreased prices, and, overall, has caused eggs to become an extremely affordable source of protein, accessible to all income classes. Within the last several years, many government officials and retailers have moved away from intensified production practices in favor of cage-free and other extensive practices. This move is mainly attributed to the potential welfare benefits that these systems may impart [1,2]. There is also evidence that the general consumer has very little knowledge of modern production systems and their effect on the animals and, therefore, many consumers view extensive systems as better for the animal [3]. Interestingly enough, even though many consumers claim to support these ideals, many studies have shown an unwillingness from the consumer to pay for higher-priced eggs from systems like cage-free and free-range [1,4,5]. Regardless of consumers’ willingness to pay, there is still a general perception by consumers that current stocking density measurements for laying hens may cause undue stress and may not be appropriate for proper hen welfare.

Research on stocking density utilizing modern white layer strains in colony cage systems seems to be lacking. In the past, it has been shown that as stocking density increases, production parameters decrease [6,7,8]. However, it is also known that hen genotype can have an impact on how hens respond to higher stocking densities [6,9]. The majority of stocking density research has been performed utilizing traditional battery cages with small numbers of hens, and very little stocking density research has been carried out using colony cage systems (sometimes referred to as California Cages). The major difference between battery cages and colony cages is their size, as colony cages are built to house many more birds than battery cages. Furthermore, in 2017, the United Egg Producers (UEP) published guidelines recommending maximum stocking densities for white and brown egg layers at 67 in² per hen for white egg layers to 86 in² per hen for brown egg layers (432 cm² and 555 cm²), stating that destinies beyond these guidelines would significantly reduce welfare, decrease production, and increase mortality [10]. Several research reports on stocking density show that densities below the UEP guidelines produced poor welfare in birds; however, these practices are no longer performed in the United States and the European Union [11,12,13].

In the USA, the UEP recommends 67 in² (445 cm²) per white-egg-laying hen in cages [10]. Conversely, the European Union (EU) has mandated that hens need more room to prevent major welfare issues and that hens be allowed enrichments such as a nesting area, perches, and a scratch area within the cage [13,14]. As noted above, many interest groups and consumers believe that high stocking densities can potentially be detrimental for hens, as they do not have room to perform natural behaviors [1,2,15,16,17]. Therefore, there is a need to compare the current stocking density with lower densities to determine if the current stocking density does result in poorer welfare and health for commercial laying hens.

Evaluating laying hen welfare is often carried out simply by visual inspection on the farm; however, welfare can go beyond what lies on the surface [18]. High stress can be an indicator of poor welfare and is typically determined in hens by analyzing corticosterone and heterophil/lymphocyte levels. Fearfulness tests can also be used to indirectly quantify stress and phycological welfare of hens. When hens are more stressed they tend to elicit a greater response to external stimuli [9,19]. Many studies also analyze bone health, as the calcium demand on laying hens can cause poor bone quality and increase the potential for breakages (which are a known welfare issue for laying hens), which can be exacerbated by stress [20,21]. Oxidative stress pathways have also been shown to be affected by stressors in poultry [7,22]. Visual inspection can also be a powerful tool to determine welfare [23,24,25]. Identifying injuries, such as keel bone damage and comb wounds, in conjunction with feather coverage, can be used to identify potential welfare and health issues in commercial laying hens.

Adjusting stocking density has been shown to affect hen stress and health parameters. Many studies have identified that greatly increasing stocking density also increases blood corticosterone and heterophil and lymphocyte ratios [14,25]. Researchers have also found that increasing stocking density decreases feather coverage and feather quality in laying hens and have attributed the decrease to an increase in injurious feather pecking and cage abrasion [26,27,28,29]. For bone health, researchers have identified that stocking density can affect the humerus and femur strength; however, stocking density was not found to affect tibia strength, which is one of the most common bones to be measured for poultry bone strength [30]. Interestingly, it appears that there is a lack of research pertaining to stocking density’s effect on oxidative stress pathways, gut health, and inflammatory cytokine expression, which can be key components of evaluating poultry welfare and health. The objective of this study was to evaluate hen production in conjunction with health and welfare across several stocking densities and determine if current stocking density practices negatively affect hens. It is hypothesized that decreasing hen stocking density will improve production, as well as the health and welfare parameters of commercial white-egg-laying hens.

2. Materials and Methods

This study was performed at the North Carolina Department of Agriculture and Consumer Services, Piedmont Research Station in Salisbury, NC. The laying cycle of this study began in November 2020 and ended in October 2021. The study was approved by the North Carolina State University’s Institutional Animal Care and Use Committee and utilized a conventional Tri-Decked layer colony cage system measuring 26 in × 48 in per cage (also known as California Cages). These cages did not include any enrichments. Chicks were brooded in a similar Tri decked cage system in the pullet house and moved to the layer house at 16 weeks of age. The lighting schedule for the laying cycle was 14L:10D. Hens were fed ad libitum and the diet is presented in

Table 1. Ingredient composition and calculated nutrient analysis for the diet fed to all hens.

Ingredient	Diet (%)
Corn	51.84
Soybean Meal	32.24
Calcium Carbonate	9.42
Dicalcium Phosphate	1.81
Salt	0.38
DL-Methionine	0.18
Soybean Oil	3.73
Santoquin	0.05
Choline Chloride	0.05
NCSU Trace Mineral Premix 1	0.20
NCSU Vitamin Premix 2	0.05
NCSU Selenium Premix 3	0.05
Calculated Values
Crude Protein %	19.50
Metabolizable Energy kcal/kg	1328.0
Calcium %	4.14
Available Phosphorus %	0.45
Total Lysine %	1.10
Total Sulfur Amino Acids %	0.825

¹ Mineral premix supplied the following per kilogram of feed: 120 mg of Zn as ZnSO₄H₂O, 120 mg of Mn as MnSO₄H₂O, 80 mg of Fe as FeSO₄H₂O, 10 mg of Cu as CuSO₄, 2.5 mg of I as Ca(IO₃)₂, and 1.0 mg of Co as CoSO₄. ² Vitamin premix supplied the following per kilogram of feed: vitamin A, 26,400 IU; cholecalciferol, 8000 IU; niacin, 220 mg; pantothenic acid, 44 mg; riboflavin, 26.4 mg; pyridoxine, 15.8 mg; menadione, 8 mg; folic acid, 4.4 mg; thiamin, 8 mg; biotin, 0.506 mg; vitamin B12, 0.08 mg; and ethoxyquin, 200 mg. The vitamin E premix provided the necessary amount of vitamin E as DL-α-tocopheryl acetate. ³ Selenium premix provided 0.3 ppm Se from sodium selenite.

. A total of 720 shaver white hens (Hendrix-Genetics, CK Boxmeer, The Netherlands) were randomly assigned to 6 replicates of 5 different density treatments. These treatments are as follows: 1342 cm² per hen (6 hens per cage), 897 cm² per hen (9 hens per cage), 671 cm² per hen (12 hens per cage), 535 cm² per hen (15 hens per cage), and 445 cm² per hen (18 hens per cage). The laying cycle began at 17 weeks of age and ended at 65 weeks of age.

2.1. Production Parameters

All production data were measured using 28-day intervals (henceforth referred to as periods) starting at week 17. Eggs were collected and recorded daily for egg production. Egg production was measured in two ways. Hen-day egg production, which is the percent production based on live hens, and hen-housed production, which is the percent production based on the number of hens placed at 17 weeks of age. Feed weigh backs were performed at the end of each 28-day interval and feed consumption was measured as grams of feed per bird per day. Feed efficiency was calculated as the total grams of eggs produced within the 28-day period divided by the total feed consumed within the 28-day period. During the third week of each 28-day interval, all eggs within a 24 h period were collected. These eggs were used for egg weights USDA (United States Department of Agriculture), egg grades, and USDA egg size analysis. The USDA egg grades and sizes are used for marketing purposes in the United States and the qualifiers can be found in the USDA egg grading manual [31]. USDA egg grade and USDA egg size analysis was performed by a trained grader according to the USDA grading handbook [31]. Mortality was calculated over the entire study and had to be transformed for statistical analysis as the mortality data were not normally distributed. Mortality included hens that were found dead and those that were culled due to injury or sickness. If a hen was observed to have a debilitating injury, such as a major open wound/broken leg or wing, or was observed to be lethargic due to sickness, then these hens were humanly euthanized.

2.2. Egg Quality Parameters

Physical egg quality parameters were measured at 23, 31, 39, 47, 55, and 63 weeks of age. All egg quality parameters were constructed with six eggs per replicate per period. Shell strength and elasticity were measured utilizing a texture analyzer (TA-HDplus, Stable Micro Systems, Surrey, UK) equipped with a 250 kg load cell, which measured force in grams. Vitelline membrane strength and elasticity were measured using a texture analyzer (TA.XTplus, Stable Micro Systems, Surrey, UK) equipped with a 5 kg load cell and a 1 mm blunt probe. Haugh unit, albumen height, shell color, and yolk color were determined utilizing the TSS QCD System (Technical Services and Supplies, Dunnington, York, UK), and yolk color was calculated utilizing the DSM yolk color fan [32,33]. Shell color was determined based on light reflectivity. An 83.3% reflectivity is considered pure white, while 0 is considered pure black. Whole, yolk, and albumen solids were performed by taking yolk and albumen from six eggs in each replicate and, along with the whole eggs, placed in bags and stomached for 30 s. The mixtures were weighed and then dried in a drying oven at 50 °C for 48 h or until dry. The dry pans were then weighed, and egg solids were calculated by dry sample weight divided by liquid sample weight. The egg components percentage was measured by weighing the whole egg, separating the yolk from the albumen, rolling on a paper towel to separate all albumen, and then weighed. Shells were washed, dried, and then weighed. Albumen weight was taken by subtracting shell and yolk weight from whole egg weight. Finally, shell thickness was calculated by utilizing digital calipers, and each egg was measured twice at the equator and then averaged.

2.3. Blood Parameters

At 23, 39, 47, and 63 weeks of age, a single blood sample (3 mL) was taken from the brachial vein, using a BD PrecisionGlide hypodermic needle (22 ga, 1 1/2 in) (Becton Dickinson, Franklin Lakes, NJ, USA), of three birds per replicate. The same birds were bled each time. If a bled bird died, then another bird was chosen to replace it. This blood was placed in a 7 mL tube containing sodium heparin to prevent clotting. For heterophil/lymphocyte ratios (H/L), 3 μL of whole blood was smeared on microscope slides and then stained using Wright’s Giemsa stain (Fisher HealthCare, Pittsburg, PA, USA). A total of 100 leukocytes, specifically heterophils and lymphocytes, were counted utilizing a 10× magnification microscope following a schematic diagram (Discover Echo, Revolve, San Diego, CA, USA). The H/L ratio was then calculated by dividing the number of heterophils by the number of lymphocytes counted. Whole blood was also placed into capillary, hematocrit tubes, centrifuged, and then measured utilizing a micro-capillary reader (Damon/IEC Division, IEC CAT No 2201, Needham, MA, USA). The blood was then spun in a centrifuge (Eppendorf, Hamburg, Germany) at 3000 RPM for 15 min to separate plasma from whole blood and stored at −20 °C until analysis. Plasma corticosterone was determined utilizing a commercial corticosterone ELISA kit (Invitrogen by Thermofisher, Waltham, MA, USA) according to the manufacturer’s recommendations (Cayman Chemical Company, item: 501320, Ann Arbor MI, USA). Concentrations were determined by a standard curve as picograms of corticosterone per milliliter of plasma.

2.4. Bone Parameters

At the end of the study (69 weeks of age), tibias were removed from two hens per replicate for bone health analysis. This analysis included bone quality index, bone breaking strength, bone width, and bone length. The bone quality index (BQI) was determined by ultrasound bone sonometer machine (Nanjing Kejin Industries Co. Ltd., Qixia District, Nanjing City, China) at the proximal end, the middle of the shaft, and the distal end of each tibia and then averaged for one average BQI per bone. Bone length was then measured using digital calipers and expressed in mm (VWR International, Radnor, PA, USA). Bone width was measured utilizing the same digital calipers at the distal end, the middle of the shaft, and the proximal end and averaged for an average width per bone. Finally, bone-breaking strength was conducted utilizing a texture analyzer with a 250 kg load cell measuring force in Newtons for peak force and Newtons per mm² for bending moment (TA-HDplus, Stable Micro Systems, Surrey, UK).

2.5. Welfare Assessments

Fearfulness tests were conducted at weeks 31, 47, and 63. Fearfulness was recorded utilizing a modified Hansen’s test, as well as a novel object latency to feed test [34,35,36]. A Hansen’s test was performed by waving a pencil in front of a cage and recording the reactions of the hens on a scale of zero to four, with zero being no reaction and four being an extreme reaction. The Hansen’s test was performed by the same trained individual for each cage at each time point. The latency-to-feed test was performed by taking two of the novel objects presented in Figure 1, placing them in the feeder evenly spaced, and then timing how long it took for hens to return to the feeder. Objects were removed after a maximum time of fifteen minutes if the hens had not returned to the feeder. The object was built to mimic the eyes of a predator.

Welfare tests were performed in accordance with the Poultry Extension Collaborative newsletter’s recommendations [18]. These tests were performed at 39, 47, 55, and 63 weeks of age. Keel bone, toes, footpad, comb, beak, and feather coverage were visually inspected by trained individuals and scored according to the newsletter. Keel bones were palpated and scored a 0 for no damage or a 1 for the presence of damage. Damage included keelbone deviations and fractures. Toes were inspected for injury (including bumblefoot) and broken bones and scored a 0 for no damage or a 1 for the presence of damage such as wounds or breakages. Footpads were examined for dermatitis. A score of 0 indicates no damage, a score of 1 indicates some necrosis and moderate swelling, and a score of 2 indicates extreme swelling. Combs were visually inspected for pecking wounds and damage. A score of 0 indicates no evidence of peck wounds, a score of 1 indicates 3 or fewer small lesions, and a score of 2 indicates at least one wound greater than 2 cm in diameter or more than 3 wounds. Finally, the feather condition was analyzed. A score of 1 indicates no or slight wear such as single missing feathers, a score of 2 indicates moderate wear such as damaged feathers (worn or deformed) or one or more featherless areas that do not exceed 5 cm in diameter, and a score of 3 indicates at least one featherless area greater or equal to 5 cm in diameter.

2.6. Histology and Gut Health

At the end of the study (69 weeks of age), birds were euthanized by cervical dislocation, and intestinal cuttings (about 3 cm) were taken from the midpoint of the jejunum of two hens per replicate (a total of 12 per treatment) and fixed in 10% neutral formalin for storage and transport. These samples were then dehydrated in a graded ethanol series then embedded in paraffin. For each hen, one 4 µm thick section was stained with hematoxylin and eosin. Villus height (tip to bottom), villus width (at the tip measured 1/8th of the way down from the lumen side of the villi and at the bottom measured 1/8th of the way up from the base), external muscle layer thickness (muscularis mucosa, submucosa, and smooth muscle layer), and crypt depth were examined. A 40× lens with a light microscope (AmScope T340B-DK-Led Siedentopf Trinocular, Irvine, CA, USA) and camera (AmScope FMA050, Irvine, CA, USA) was used. Ten villi were measured per hen and averaged as the morphological value. These villi were measured with the Amscope image software (AmScope 3.7, Irvine, CA, USA).

2.7. Gene Expression

Intestinal cuttings (about 1 cm long) were taken from 2 hens per replicate (12 per treatment) and pooled by treatment in RNAlater^® and stored at −20 °C before processing at the end of the study (69 weeks of age). Gene expression analysis was performed utilizing real-time PCR analysis. RNA was isolated using an RNAeasy kit (QIAGEN, Hilden, Germany) according to the manufacturer’s instructions. RNA yields were quantified using a NanoDrop2000 (Thermo Fisher Scientific, Waltham, MA, USA) to ensure proper RNA dilutions. RNA was then transcribed to cDNA for QPCR using a high-capacity cDNA synthesis kit (Applied Biosystems, Waltham, MA, USA). QPCR was performed on each sample gene combination in triplicate. Each well contained 2.5 ng of the sample, 500 mM of forward and reverse primers specific to one of the genes (

Table 2. Genes and genetic sequences used for gene expression.

Gene	Primer	Directional Sequence	Sequence
Beta Actin	b-Actin	Forward	GTCCACCTTCCAGCAGATGT
		Reverse	ATAAAGCCATGCCAATCTCG
Claudin	Claudin1	Forward	CACACCCGTTAACACCAGATTT
		Reverse	GAGGGGGCATTTTTGGGGTA
Occludin	OCLN	Forward	GCTCCCGGCTGCCATTTTAAG
		Reverse	GGAGCGTCGTCCACGTAGTA
Zona Occludin	TJPI	Forward	GCAGTCGTTCACGATCTCCT
		Reverse	TCTCTGCTTCGAAGACTGCC
Heat Shock Protein	HSP70	Forward	GCGGAGCGAGTGGCTGACTG
		Reverse	CGGTTCCCCTGGTCGTTGGC
Interleukin 1beta	IL-1b	Forward	CTGCCTGCAGAAGAAGCCT
		Reverse	TGTCAGCAAAGTCCCTGCTC
Interleukin 10	IL10	Forward	AGGAGACGTTCGAGAAGATGGA
		Reverse	TCAGCAGGTACTCCTCGATGT
TNF-alpha	LITAF	Forward	CTGTGGGGCGTGCAGTG
		Reverse	ATGAAGGTGGTGCAGATGGG
Mucin	MUC2	Forward	GTGAATGGCACTACGAGCCT
		Reverse	CTGGGGTAGCAACCTTCCAG
Superoxide Dismutase	SOD1	Forward	AAATGGGTGTACCAGCGCA
		Reverse	ACTCCTCCCTTTGCAGTCAC
Catalase	CAT	Forward	TCAGGAGATGTGCAGCGTTT
		Reverse	GTGCGCCATAGTCAGGATGA
Glutathione Peroxidase	GPX3	Forward	ACCCTGCAGTACCTCGAACT
		Reverse	CCCAAATTGGTTGGAGGGGA

), and 2× power SYBR green master mix (Applied Biosystems, Waltham, MA, USA). PCR was performed on the Applied Biosystems StepOnePlus real-time PCR system. All results were normalized to Beta Actin expression and the reciprocal was taken in order to more easily interpret the data. The formula was as follows: 1/(target gene cycle threshold/beta-actin cycle threshold). All data are presented in reciprocal cycle thresholds or CT⁻¹.

2.8. Statistical Analysis

Statistical analysis was performed utilizing Rstudio 4.2.2 with packages tidyverse, agricolae, AICcmodavg, and janitor [37,38,39,40,41]. The statistical model for production and egg quality was treatment, age, and the interaction of treatment and age. Mortality was transformed using an arcsine square root function to normalize the data as the data were not normally distributed. For health and welfare parameters, the data were analyzed each period with the model only containing treatment. The statistical unit used was the cage replicate except for gene expression where intestinal cuttings were randomly selected from the 12 hens taken for each treatment. Cage level was investigated as a blocking factor; however, it was determined through effect p-value and AICs that the blocking factor was not needed. Tukey’s HSD was performed for multiple comparisons and a p-value of <0.05 was considered to be statistically significant.

3. Results

3.1. Production Data

Table 3. The effect of stocking density and age on commercial white egg layer egg production, feed consumption and feed efficiency.

	Hen-Day Prod. (%)	Hen-Housed Prod. (%)	Feed Consumption (g/bird/day)	Feed Efficiency (egg g/feed g)	Egg Weight (g)	Mortality (%)
Density (cm2)
1342	92.1 A	91.1 A	103.9	0.510	57.48 AB	2.78
897	91.2 AB	90.7 AB	100.7	0.522	57.52 AB	0.93
671	90.6 B	89.3 BC	101.1	0.514	57.60 AB	2.78
535	91.4 AB	90.2 AB	102.0	0.514	57.21 B	2.22
445	90.3 B	88.3 C	102.6	0.512	57.96 A	4.63
SEM	2.06	2.04	1.32	0.012	0.516	1.31
p-value	<0.001	< 0.001	0.096	0.156	0.003	0.533
Age (weeks)
17–20	34.6 F	34.6 E	77.9 E	0.203 H	45.67 F
21–24	92.8 E	92.7 CD	99.1 D	0.496 G	52.94 E
25–28	100.5 A	100.3 A	106.3 BC	0.526 EF	55.60 D
29–32	95.7 BCD	95.3 BC	106.7 B	0.520 FG	58.29 C
33–36	99.8 A	99.2 A	115.2 A	0.519 FG	59.62 C
37–40	96.5 BC	95.5 BC	104.4 BCD	0.545 CDEF	58.98 BC
41–44	97.5 B	96.0 B	104.2 BCD	0.553 BCDE	59.05 BC
45–48	97.1 BC	95.4 BC	102.4 BCD	0.567 ABC	59.75 B
49–52	97.2 BC	95.3 BC	101.8 BCD	0.574 AB	59.69 B
53–56	95.1 CD	92.8 CD	100.2 CD	0.581 A	61.05 A
57–60	92.6 E	90.2 D	99.0 D	0.560 ABCD	59.77 B
61–64	93.9 DE	91.5 D	107.3 B	0.537 DEF	61.25 A
SEM	0.451	0.598	1.09	0.005	0.206
p-Value	<0.001	<0.001	<0.001	<0.001	<0.001
Interaction p-value	0.724	0.953	0.819	0.759	0.444

^A,B,C,D,E,F Signifies statistical significance (p < 0.05) between groups.

presents the production and mortality data for both age and density, although mortality was only measured as whole study mortality. Both hen-day and hen-housed production followed similar trends. This study demonstrated that the lowest density of 1342 cm² per hen had higher (p < 0.001) hen-day egg production than 671 cm² and 445 cm² per hen by 1.5% and 1.8%, respectively. Similarly, hen-housed production followed similar trends as hen-day production. The 1342 cm², 897 cm², and 535 cm² densities per hen demonstrated statistically higher (p < 0.001) hen-housed egg production compared to the 445 cm² per hen by 2.8%, 2.4%, and 1.5%, respectively, while the 1342 cm² density treatment also had higher hen-housed production than the 671 cm² treatment by 1.7%. Both hen-day and hen-housed production by age followed expected patterns, with both reaching peak production at 25–28 weeks of age and declining after. Density was not found to affect feed consumption; however, age was found to affect feed consumption following expected trends with feed consumption increasing with hen age and with peak production and declining shortly after. Comparatively, density was not found to affect feed efficiency, while age was found to have an effect with feed efficiency increasing as the hen aged. The 535 cm² density was found to produce lighter (p < 0.01) eggs than the 445 cm² density by 0.75 g. Eggs were found to increase in weight as hens aged. Finally, stocking density was not found to affect mortality. There was no interaction effect between stocking density and age for any production parameter measured.

3.2. USDA Egg Size and Quality

Table 4. The effect of stocking density and age on commercial white egg layer USDA marketable egg sizes and egg grades.

	Grade A (%)	Grade B (%)	Loss (%)	XL (%)	L (%)	M (%)	S (%)
Density (cm2)
1342	92.4 B	0.35	7.27 A	48.68	39.40	3.16	8.62
897	94.7 AB	0.19	5.16 AB	49.78	37.66	3.36	9.21
671	95.5 A	0.20	4.23 B	48.62	37.83	3.92	9.62
535	96.0 A	0.26	3.75 B	47.23	38.73	3.72	10.2
445	96.3 A	0.43	3.26 B	48.34	39.19	3.57	8.68
SEM	0.656	0.16	0.637	4.62	4.14	0.741	3.08
p-value	<0.001	0.790	<0.001	0.564	0.781	0.887	0.121
Age (weeks)
17–20	99.1 A	0.17 AB	0.78 C	1.34 F	0.66 G	4.74 BC	93.3 A
21–24	97.6 AB	0.48 AB	1.94 BC	1.14 F	67.83 A	16.54 A	14.4 B
25–28	98.8 A	0.11 B	1.06 C	1.77 F	89.89 A	6.83 B	1.50 C
29–32	95.7 ABC	0.56 AB	3.73 ABC	8.24 F	89.57 A	2.08 CD	0.11 C
33–36	94.9 ABCD	0.11 B	4.94 ABC	16.9 E	81.60 A	1.44 CD	0.00 C
37–40	93.1 CD	0.22 AB	6.66 A	49.3 D	48.42 C	2.00 CD	0.28 C
41–44	96.4 ABC	0.11 B	1.94 BC	66.8 C	30.14 D	2.90 BCD	0.29 C
45–48	92.5CD	0.11 B	7.36 A	86.8 A	11.75 EF	1.20 D	0.27 C
49–52	93.1 CD	0.23 AB	6.63 A	78.8 B	19.24 E	1.66 CD	0.33 C
53–56	93.7 BCD	0.00 B	6.26 AB	94.2 A	4.96 FG	0.54 D	0.00 C
57–60	91.1 D	1.32 A	7.59 A	90.4 A	7.25 FG	2.02 CD	0.37 C
61–64	93.7 BCD	0.00 B	6.34 A	87.2 A	11.42 EF	0.60 D	0.37 C
SEM	0.957	0.191	0.934	1.44	1.64	0.734	0.48
p-Value	<0.001	0.016	<0.001	<0.001	<0.001	<0.001	<0.001
Interaction p-value	0.102	0.607	0.214	0.917	0.660	0.886	<0.001

^{A,B,C,D,E,F,G} Signifies statistical significance (p < 0.05) between groups.

also presents the USDA egg quality and size distribution data for the study. Descriptions of the egg grades can be found in the USDA egg grading manual [31]. An amount of 1342 cm² per hen was found to have fewer (p < 0.001) grade A eggs than 671 cm², 535 cm², and 445 cm² per hen by 3.1%, 3.6%, and 3.9%, respectively. Furthermore, as hens aged, they produced fewer grade A eggs. Following the inverse trend to grade A eggs, the 1342 cm² treatment produced a higher (p < 0.001) percentage of grade loss eggs than the 671 cm², 535 cm², and 445 cm² treatments by 3.04%, 3.52%, and 4.01%, respectively. Age also had an effect on the percentage of grade loss eggs produced, which increased as the hens aged. Density was not found to have an effect on grade B percentage; however, age did seem to have an effect on grade B percentage, with the highest number of grade B eggs being produced at the end of the study. Density did not have a significant effect on USDA egg size distribution; however, age did affect % XL, % L, % M, and % S. As hens aged, XL % increased, whereas most L eggs were produced during and close to peak production. The most S and M eggs were produced at and before peak production. Hens were found to produce more XL eggs at the end of the study, more L eggs at and right after peak production, more M eggs right at peak production, and more S eggs before peak production. There was no interaction between density and age for any USDA parameters measured except for the percentage of small eggs.

3.3. Physical Egg Quality

Eggshell and vitelline membrane strength parameters are presented in

Table 5. The effect of stocking density and age on commercial white egg layer vitelline membrane (VM) and shell quality.

	VM Strength (N/mm2)	VM Elasticity (mm)	Shell Strength (N/mm2)	Shell Elasticity (mm)	Shell Thickness (mm)
Density (cm2)
1342	2.15	1.77	5.11	0.279	0.374
897	2.19	1.81	5.19	0.294	0.377
671	2.09	1.68	5.18	0.283	0.375
535	2.17	1.77	5.06	0.281	0.376
445	2.19	1.79	4.98	0.280	0.371
SEM	0.043	0.057	0.098	0.011	0.002
p-value	0.642	0.978	0.138	0.730	0.382
Age (weeks)
23	2.41 A	2.12 A	5.92 A	0.294	--
31	2.05 CD	1.60 BC	5.36 B	0.258	0.377
39	2.16 BC	1.83 B	4.46 D	0.225	0.379
47	2.15 BC	1.73 B	5.12 BC	0.366	0.372
55	2.23 AB	1.83 B	4.97 C	0.340	0.373
63	1.94 D	1.49 C	4.80 CD	0.219	0.373
SEM	0.039	0.051	0.069	0.005	0.002
p-Value	<0.001	<0.001	<0.001	0.882	0.223
Interaction p-value	0.465	0.311	0.551	0.755	0.319

^A,B,C,D Signifies statistical significance (p < 0.05) between groups.

. This study found that density did not have an effect on the vitelline membrane and eggshell strength and elasticity. Furthermore, stocking density also did not have an effect on eggshell thickness. Age was found to have an effect on vitelline membrane strength and elasticity, as well as eggshell strength. Both vitelline and eggshell strength were found to be the strongest at the start of the study and declined in strength as the hens aged. Vitelline membranes were found to be the most elastic at the beginning of the study and decreased in elasticity as hens aged. There was no difference found in shell elasticity and shell thickness as the hen aged.

Table 6. The effect of stocking density and age on commercial white egg layer internal egg quality and shell color.

	Shell Reflectivity	Albumen Height (mm)	Egg Weight (g)	Haugh Unit	Yolk Color
Density (cm2)
1342	80.7	7.95	58.3	89.2	5.89
897	80.3	8.08	58.5	89.9	5.90
671	80.5	8.14	58.5	90.2	5.87
535	80.6	8.06	58.1	89.9	6.00
445	80.5	8.04	58.7	89.6	5.87
SEM	0.873	0.094	0.564	0.582	0.135
p-value	0.953	0.552	0.809	0.563	0.860
Age (weeks)
23	84.66 A	8.59 A	52.36 C	94.2 A	7.38 A
31	84.07 AB	8.03 BC	57.30 B	90.0 BC	5.69 BC
39	80.75 C	8.03 BC	59.51 A	89.5 BC	6.01 B
47	82.21 ABC	7.77 CD	59.65 A	87.9 CD	5.57 CD
55	81.83 BC	7.50 D	60.64 A	86.1 D	5.23 D
63	69.64 D	8.44 AB	61.06 A	91.1 B	5.50 CD
SEM	0.202	0.077	0.287	0.433	0.064
p-Value	<0.001	0.001	<0.001	<0.001	<0.001
Interaction p-value	0.842	0.140	0.883	0.112	0.356

^A,B,C,D Signifies statistical significance (p<0.05) between groups.

presents eggshell reflectivity, albumen height, egg weight from the egg quality measurements, Haugh Unit, and yolk color measurements. This study did not detect any differences in eggshell reflectivity, albumen height, egg weight Haugh Unit, or yolk color between densities. Eggshells were found to be whiter; albumens were taller, yolks were darker, and eggs weighed lighter at the beginning of the study. As the hens aged, the shells became darker, albumens fell, yolks became lighter, and eggs became heavier. Moreover, eggs from younger hens were found to have more desirable Haugh units, which decreased as the hen aged.

Table 7. The effect of stocking density and age on commercial white egg layer egg solids and components.

	Shell (%)	Albumen (%)	Yolk (%)	Whole Egg Solids (g)	Yolk Solids (g)	Albumen Solids (g)
Density (cm2)
1342	9.68	63.6	26.8	25.1	49.2	12.4
897	9.72	63.5	26.8	26.0	49.3	12.3
671	9.62	63.5	26.9	23.8	49.3	12.4
535	9.75	63.4	26.9	24.4	49.2	12.1
445	9.52	63.8	26.7	25.0	49.3	12.2
SEM	0.062	0.157	0.155	0.723	0.312	0.137
p-value	0.102	0.423	0.921	0.512	0.757	0.167
Age (weeks)
31	9.78 AB	64.4 A	25.8 C	24.7	48.3	13.0 A
39	9.88 A	63.5 B	25.6 B	25.2	49.5	12.7 AB
47	9.51 C	53.3 B	27.2 A	25.5	50.3	12.0 C
55	9.57 BC	63.4 B	27.1 AB	25.0	49.7	12.3 BC
63	9.54 C	63.2 B	27.3 A	23.9	48.5	11.4 B
SEM	0.058	0.134	0.117	0.707	0.255	0.086
p-Value	<0.001	<0.001	<0.001	0.479	0.612	<0.001
Interaction p-value	0.293	0.080	0.291	0.665	0.390	0.463

^A,B,C Signifies statistical significance (p < 0.05) between groups.

presents the percentage of the shell, albumen, and yolk of the whole egg, as well as the dry matter of the whole egg, the yolk, and the albumen. This study did not detect any differences between stocking densities for any of the aforementioned quality parameters. Age did have an effect on the shell, albumen, and yolk percentages. The shell was a greater percentage of the whole egg at the beginning of the study and decreased as the hens aged. At 31 weeks of age, the albumen was found to be a higher percentage of the egg than the other ages. The yolk percentage of the egg was also found to increase as the hens aged. This study also did not detect any change in whole egg or yolk dry matter as the hens aged; however, this study found that hens had a greater amount of albumen solids when the hens were younger than when they were older. Finally, there was no interaction effect detected between age and stocking density for any of the parameters listed above.

3.4. Stress Parameters

Table 8. The effect of stocking density on white egg layer stress and fear parameters.

	1342 cm2	897 cm2	671 cm2	535 cm2	445 cm2	SEM	p-Value
Week 23
CORT (pg/mL)	470.7	657.2	428.6	562.5	562.5	130.5	0.818
H/L Ratio	0.232	0.231	0.327	0.252	0.247	0.043	0.700
Week 39
CORT (pg/mL)	370.6	387.1	423.0	672.1	458.5	85.1	0.117
H/L Ratio	0.047	0.080	0.067	0.016	0.0.56	0.023	0.400
Hematocrit	36.3	33.4	37.1	36.8	35.5	2.83	0.833
Hansen’s Test	1.67 AB	1.50 B	2.40 AB	2.29 AB	3.17 A	0.448	0.013
Latency to Feed (s)	243.0 A	219.3 AB	181.6 AB	130.1 B	119.3 B	32.8	0.002
Week 47
CORT (pg/mL)	443.7	439.9	350.5	786.7	742.5	154.5	0.090
H/L Ratio	0.213	0.157	0.131	0.209	0.132	0.055	0.594
Hematocrit	30.8	30.1	34.8	30.0	32.4	2.33	0.654
Hansen’s Test	1.33 B	1.83 AB	1.83 AB	1.83 AB	2.83 A	0.322	0.008
Latency to Feed (s)	353.3	319.7	213.5	298.2	459.3	74.3	0.452
Week 63
CORT (pg/mL)	315.9 B	472.9 AB	357.9 AB	542.1 AB	734.6 A	95.2	0.008
H/L Ratio	0.038 B	0.041 AB	0.062 AB	0.042 AB	0.064 A	0.007	0.031
Hematocrit	27.7 AB	27.1 B	27.7 AB	28.9 AB	30.7 A	0.896	0.022
Hansen’s Test	1.67 AB	1.33 B	1.33 B	1.83 AB	2.67 A	0.251	0.010
Latency to Feed (s)	233.7	141.3	57.7	264	407.3	72.6	0.102
Catalase (CT-1)	0.883	0.890	0.894	0.862	0.868	0.014	0.192
HSP 70 (CT-1)	0.866	0.583	0.848	0.813	0.876	0.012	0.773
Glutathione Peroxidase (CT-1)	0.847	0.845	0.88	0.812	0.854	0.019	0.777
Superoxide Dismutase (CT-1)	0.852	0.865	0.886	0.824	0.876	0.012	0.972

^A,B Signifies statistical significance (p < 0.05) between groups.

contains both corticosterone (CORT) and heterophil lymphocyte ratios (H/L ratios) for each time point collected. This study did not identify any differences in corticosterone, hematocrit, or H/L ratios during weeks 23, 39, or 47 between densities. At week 63, hens kept at 445 cm² had higher (p = 0.008) basal corticosterone levels than hens under 1342 cm² by 418.7 pg/mL. At week 63, hens under 445 cm² also had higher (p = 0.031) H/L ratios than hens under 1342 cm² by 0.062 heterophils per lymphocyte. Hens under 445 cm² per bird also experienced higher (p = 0.022) hematocrit compared to hens under 897 cm² during week 63 by 3.7. This study did not find any difference in genetic expression in the jejunum of catalase, glutathione peroxidase, superoxide dismutase, or HSP-70 at week 63.

3.5. Fearfulness

Table 8 also contains data from the fearfulness tests, modified Hansen’s test (MHT), and latency to feed (LTF) performed in this study. At week 39, hens reared under 445 cm² elicited a stronger (p = 0.0134) reaction to Hansen’s test than hens reared under 897 cm² by 1.6. Hens raised under 1342 cm² took longer (p = 0.002) to return to the feed during the LTF compared to hens under 535 cm² and 445 cm² by 112.9 s and 123.7 s, respectively. During week 47, hens raised under 445 cm² elicited a greater (p = 0.0079) response to Hansen’s test than hens under 1342 cm² by 1.5. No difference between densities was found for the LTF test at week 47. During week 63, hens with the highest stocking density elicited a higher (p = 0.01) response to Hansen’s test than hens under 897 cm² and 671 cm² by 1.34 for both densities. This study found no differences in latency to feed during week 63.

3.6. Bone and Intestinal Health

The effect of stocking density on bone quality is presented in

Table 9. The effect of stocking density on bone health, villus health, and jejunim gene expression inspection of commercial white egg layers.

	1342 cm2	897 cm2	671 cm2	535 cm2	445 cm2	SEM	p-Value
Bone Health
Bone width (mm)	6.82	6.79	7.01	6.96	6.95	0.101	0.216
Bone length (mm)	113.7	113.4	114.8	114.1	114.2	0.829	0.551
Bone quality index	84.6	83.3	81.1	83.0	80.8	1.83	0.178
Bending moment (N/mm)	0.105	0.107	0.109	0.116	0.119	0.007	0.104
Peak force (N)	14.7	15.2	14.2	16.1	16.3	0.673	0.080
Villus Health
Villus height (µm)	672.6 AB	818.6 A	746.4 AB	706.6 AB	533.5 B	42.2	0.028
Villus tip width (µm)	122.0	111.5	108.2	109	135.9	7.86	0.347
Villus bottom width (µm)	152.2	131.3	130.8	144.7	165.5	8.85	0.183
Villus area (µm2)	94,011	97,696	89,427	89,288	83,517	7483	0.228
Crypt depth (µm)	119.4	123.0	138.1	104.8	112.9	6.52	0.177
Villus/crypt ratio	6.03	7.76	6.34	7.13	5.36	0.537	0.289
Muscularis (µm)	149.6	148.4	148.6	159.9	135.7	11.48	0.658
Gene Expression
Claudin 1 (CT-1)	0.650	0.687	0.688	0.645	0.686	0.008	0.417
Occludin (CT-1)	0.620	0.651	0.656	0.644	0.655	0.010	0.094
Zona Occludin (CT-1)	0.790	0.811	0.808	0.772	0.799	0.009	0.498
IL-1b (CT-1)	0.641	0.679	0.707	0.605	0.684	0.013	0.920
IL-10 (CT-1)	0.624 B	0.666 AB	0.675 AB	0.639 AB	0.690 A	0.125	0.022
Tnf-a (CT-1)	0.769	0.793	0.814	0.765	0.803	0.010	0.365
MUC2 (CT-1)	0.879	0.907	0.919	0.863	0.901	0.013	0.989

^A,B Signifies statistical significance (p < 0.05) between groups.

. Overall, stocking density did not influence bone width, bone length, bone quality index, bending moment, or peak breaking force of chicken tibia bones. Table 9 also presents information on the villus from the jejunum of these hens. Overall, tip width, bottom width, villus area, crypt depth, villus/crypt ratio, and muscularis thickness were not significantly different between density treatments. There was a difference in villus height where hens in the highest stocking density exhibited shorter (p = 0.028) villi than hens under 897 cm² by 285.1 nm. Furthermore, levels of Claudin-1, Occludin, Zona Occludin, and MUC-2 were not found to be different between density treatments. Table 9 also presents the expression of cytokines in the jejunum. This study did not find any differences in IL-1b and TNF-a between density treatments. This study did, however, find that hens with the highest stocking density had greater (p = 0.022) expression of IL-10 than the lowest stocking density by 0.068.

3.7. Welfare Parameters

All welfare parameters from visual inspection and palpations are presented in

Table 10. The effect of stocking density on welfare scores from visual inspection of commercial white egg layers.

	1342 cm2	897 cm2	671 cm2	535 cm2	445 cm2	SEM	p-Value
Week 31
Beak	1.00	1.00	1.00	1.00	1.00	0.00	0.161
Keel	0.00	0.00	0.00	0.06	0.06	0.02	0.129
Toe	0.00	0.06	0.00	0.00	0.00	0.01	0.489
Foot	0.06	0.11	0.22	0.17	0.17	0.09	0.351
Comb	0.11	0.06	0.06	0.06	0.17	0.07	0.629
Feathers	1.00	1.00	1.00	1.00	1.00	0.01	0.161
Week 39
Beak	1.00	0.80	0.90	0.90	0.10	0.94	0.505
Keel	0.06	0.11	0.06	0.00	0.06	0.05	0.505
Toe	0.06	0.00	0.00	0.00	0.00	0.01	0.161
Foot	0.28	0.28	0.17	0.17	0.33	0.12	1.000
Comb	0.06	0.06	0.06	0.11	0.17	0.07	0.221
Feathers	1.10 AB	1.06 B	1.10 AB	1.50 A	1.30 AB	0.11	0.035
Week 47
Beak	0.78	0.78	0.78	0.89	0.78	0.09	0.709
Keel	0.11	0.11	0.11	0.11	0.22	0.09	0.368
Toe	0.11	0.00	0.06	0.06	0.17	0.06	0.452
Foot	0.50	0.33	0.17	0.28	0.28	0.11	0.206
Comb	0.11	0.06	0.11	0.11	0.06	0.07	0.809
Feathers	1.10 B	1.00 B	1.20 B	1.40 AB	1.83 A	0.098	0.001
Week 55
Beak	0.89	0.83	0.83	0.94	0.10	0.05	0.097
Keel	0.11	0.17	0.11	0.28	0.28	0.10	0.187
Toe	0.06	0.10	0.06	0.00	0.06	0.06	0.585
Foot	0.44	0.17	0.22	0.17	0.28	0.12	0.417
Comb	0.17	0.17	0.17	0.28	0.06	0.076	0.660
Feathers	1.00 C	1.22 BC	1.33 BC	1.67 AB	2.28 A	0.13	0.001
Week 63
Beak	0.61	0.67	0.53	0.56	0.94	0.15	0.248
Keel	0.39	0.31	0.27	0.39	0.61	0.16	0.296
Toe	0.22	0.19	0.33	0.17	0.50	0.14	0.255
Foot	0.11	0.22	0.07	0.06	0.28	0.11	0.678
Comb	0.17	0.11	0.27	0.11	0.22	0.09	0.691
Feathers	1.00 C	1.06 C	1.33 BC	1.78 B	2.56 A	0.10	0.001

^A,B,C Signifies statistical significance (p < 0.05) between groups.

. During week 31, there were no differences found in any of the welfare parameters across stocking densities. At week 39, no difference in beak, keel, toe, foot, or comb scores was found. Feather score, however, was found to be poorer (p = 0.035) from hens in 535 cm² compared with the scores from hens in 897 cm² by 0.44. During week 47, no differences in beak, keel, toe, foot, or comb scores were found across densities. Feather score was found to be highest (p < 0.001) in the highest stocking density compared with the scores from hens in 1342 cm², 897 cm², and 671 cm² by 0.73, 0.83, and 0.63, respectively. At week 55, no differences were found in beak, keel, toe, foot, or comb scores between densities. Hens with the lowest density (1342 cm²), however, were found to have better (p < 0.001) scores than hens with 535 cm² and 445 cm² densities by 0.67 and 1.28, respectively. Hens with the highest density also had higher feather scores than hens under 897 cm² and 671 cm² densities by 1.50 and 1.23, respectively. Finally, during week 63, this study found no differences in beak, keel, toe, foot, and comb scores. The highest stocking density was found to have higher (p < 0.001) feather scores than all other densities by between 1.56 and 0.88. Hens raised under 535 cm² also had higher feather scores than hens in 1342 cm² and 897 cm² densities by 0.78 and 0.72, respectively. When birds were assessed, no broken keels were identified at any time point and no bumble foot was discovered at any time point.

4. Discussion

Stocking density is an important issue in poultry production. The current study aimed to determine how stocking density affected hen production, health, and welfare. The method of changing stocking density in this study was reducing group size. While the density change and the group size change may both be attributed to the differences seen, this method was chosen because it is the most applicable to the industry. If producers decide to increase space per hen by decreasing stocking density, they will simply remove a bird or two from the cage over installing different size cages.

4.1. Production and Egg Quality

Many studies have identified that dramatically increasing stocking density will decrease hen-day egg production across several production systems, which holds true with the current study, specifically between the highest and lowest stocking densities [7,27,36,42]. Kang et al. [14] identified that brown egg layers in higher stocking densities exhibited higher levels of heterophils (and similar levels of lymphocytes), demonstrating that brown egg layers in higher stocking densities exhibit higher levels of physiological stress than hens in lower stocking densities. Higher levels of stress could be a reason why hens with the highest stocking density had lower egg production. Furthermore, when evaluating hen-housed egg production, the trend still holds true although the difference between the highest and lowest densities is greater for the hen-housed production parameter than the hen-day production parameter. While mortality was not significantly different between treatments the highest-density treatment had the highest numerical mortality. Higher numeric mortality coupled with lower hen-day egg production led to the greater hen-housed production difference between high and low densities. From these results, it seems that decreasing stocking density can improve egg production among the hens placed. The mortality results of this study correspond with other laying hen stocking density studies, even those performed with brown egg layers, or in other housing environments although a number of studies did not report mortality data [27,36,43].

Another interesting result from this study is the absence of significant differences between treatments when looking at feed consumption and feed efficiency. Many studies in the past have reported depressed feed consumption and poorer feed efficiency at higher stocking densities and reduced feeding space; however, the present study disagrees with these findings [14,43,44]. Studies utilizing modern genetic strains (within the past 5 years) exhibit no differences in feed efficiency or feed consumption between density treatments except for those treatments that provide less space than UEP recommended guidelines [10,30,42].

When analyzed in conjunction with egg weight, this study showed an inverse correlation between hen-day egg production and egg weight, which could explain why feed efficiency was not statistically significant. Results from past studies on the effect of stocking density on egg weight are conflicting. Several studies indicate that stocking density does not affect egg weight, although several of these studies also reported no change in egg production [14,30,44,45]. The depression in egg production of the highest stocking density in this study was perhaps the cause of the increase in egg mass in that treatment, as it is known that lower egg production directly correlated to heavier egg weight [46]. It appears that, currently, white-egg-laying hens are bred to minimize stress and maximize economic production at the UEP recommended stocking density; however, when increasing the density further, according to several studies cited previously, production falls and social stress increases [45].

USDA egg sizes and grades are novel measurements to this study and important for marketing table eggs in the American market. This study revealed that hens in the lowest stocking density produced more loss grade eggs and fewer grade A eggs than hens in other stocking densities. In a study by Anderson et al. [36], hens in lower stocking densities were found to spend more time moving in the cage. If the hens are more mobile, they could cause damage to freshly laid eggs, or cause body checks to eggs that have not been laid yet. Furthermore, as more hens are placed in a cage, more obstacles (in the form of birds) are in the way of the eggs rolling to the collection trays, which will slow the eggs down and potentially alleviate some cracking from the eggs, as they cannot gather the same speed as eggs in lower density cages.

Rather unexpectedly, this study did not detect any differences in physical egg quality parameters measured when compared between the experimental densities. From the USDA egg grade results, we expected to at least see differences in shell quality; however, the absence of these differences in shell quality reinforces the theory that egg losses are possibly due to eggs being damaged as they are rolling to the collection tray. Other studies have yielded similar results, particularly with white egg layers in densities that meet or exceed the UEP guidelines, revealing no differences in physical egg quality between densities [29,47]. Even egg quality from brown egg layers remained mostly consistent across stocking densities according to previous studies [20,26,28]. Therefore, it appears that density alone does not affect egg quality in modern hens.

Production can be used as an indication of stress (as highly stressed hens typically have poorer production), but it is not the only way to assess welfare, nor should it be. Many strains of chicken have been bred to be able to produce better in different environments, temperatures, or even high-stress conditions [46,48,49]. Poorer egg production can be an indication of heightened stress or poorer welfare in high-density birds; however, to fully capture the impact that stocking density has on hen welfare, health, and stress, other parameters were evaluated to provide a more holistic picture. The production results do indicate that by reducing stocking density to a point, the individual birds become more efficient with their production.

4.2. Hen Welfare

From the results of this study, it seems that physiological stress does not reveal itself until the hens have reached a certain age and only between the two extremes, as corticosterone and H/L ratios were significantly higher in the highest stocking density compared to the lowest stocking density at week 63. As a note, starting with the youngest hens and continuing to the oldest hens, while the data were not significant until the final sampling date, the p-values of CORT steadily fell until significant at the final week. This can potentially indicate that stress slowly builds in the highest density, revealing itself at the end of the study. These observations are in line with many other studies that have identified increased stocking density as a stressor of laying hens [14,25]. Results from Roy et al. [25] demonstrate similar results, as at the beginning of the study, there was no difference in CORT levels, but, by the end of the study, the highest density was recorded at the highest CORT levels. Similarly, the present study found that H/L ratios were not significant until the end of the study, with the highest stocking density recording the highest H/L ratio. H/L ratios are utilized as another measure of stress, typically chronic stress, in conjunction with CORT. A higher H/L ratio means a greater presence of heterophils and during a stress event, the body produces more heterophils to protect against bacterial infection from a potential injury as an evolutionary defense mechanism [50,51]. Another indicator of stress is the expression of heat shock proteins. While heat shock proteins are typically expressed as a defense against thermal stress, these can be expressed in response to other stressors and are considered by some to be part of the general stress response [52,53,54]. While this study collected HSP-70 samples from the jejunum only, HSP-70 has been found expressed in a wide variety of biological systems to protect proteins from denaturing in transport during times of physiological stress [52,55]. Our research did not find any difference in HSP-70 between densities at the end of the study, possibly indicating that HSP-70 may not play a role in modulating stress due to increased stocking density. However, this study only determined HSP-70 levels in the jejunum. As HSP-70 is also produced in other organ systems, evaluating HSP-70 levels in these systems can be worthwhile research.

Another indication of stress is the upregulation of oxidative stress pathway proteins. Three of the most important proteins involved in cellular protection against oxidative stress are catalase, glutathione peroxidase, and superoxide dismutase. These molecules protect body cells against reactions with free radicals [56,57,58,59]. These free radicals can be produced by stressors such as emotional, psychological, disease, and environmental stress, some of which can be direct consequences of increasing stocking density [59,60,61,62]. Catalase is an enzyme that converts the reactive species of H₂O₂, produced during stress and normal biological functions, into water and oxygen, thereby protecting cellular components from the reactive hydrogen species [63]. Glutathione peroxidase also works to reduce peroxides and other organic hydroperoxides into water and oxygen, similar to catalase, but by using reduced glutathione as an electron donor [64]. Finally, superoxide dismutase limits levels of superoxide (O₂⁻), which can cause great harm within biological systems. From the results of this study, no differences were discovered in levels of these proteins indicating that stocking density does not place a strain on the oxidative stress pathways in the jejunum. However, further analysis needs to be performed, looking into different time points as well as expressions from different organs. These genes are expressed in many organs throughout the body system and oxidative stressors could have changed at different time points and then normalized by the end of the study.

Levels of fearfulness can also indicate stress and emotional state. Research indicates that fear responses become greater when stress and anxiety levels are elevated although some researchers have found that certain stressors were reduced in higher fear hens [36,65,66]. More docile hens are preferred as docile hens are easier for farm staff to work with and some research has indicated that fear and poor welfare are highly correlated [67,68,69]. In the present study, the MHT was used to identify a reaction to a sudden stress event, and the latency to feed test was used in replacement of the tonic immobility test, identifying how long hens would remain in a fearful state. In the beginning, hens elicited a greater latency to feed in the lower stocking densities, compared with hens in the highest two densities. Conversely, it was found that hens with the highest stocking density elicited the strongest response to the MHT than the lower stocking densities. The MHTs were consistent across the study in that the highest scorers were always the highest density; however, the lowest scores shifted between the three lowest densities across the study. This could be due to the scorer’s experience because, as the study progressed, the scorer became more experienced in scoring the tests. Compared to other research, the latency-to-feed test seems to be a novel contribution to laying hen density research and was chosen as a less time-intensive measurement. In broilers, researchers have not found a difference in latency to feed between densities; however, many of these tests did not utilize an object specifically to induce fear [70,71]. The object created for the present study was made to mimic the eyes of a predator and induce a response in the hens. Other researchers have discovered a direct correlation between laying hen stocking density and tonic immobility indicating greater fear in higher densities using this method [72]. While it is unknown why latency to feed was found to be higher in younger hens, and the lowest MHT scorers were not consistent, it is clear from this study that hens in elevated stocking densities will elicit a stronger reaction from a sudden stressor than lower densities. Further research is needed to broaden the understanding of how stocking density affects fearfulness responses, such as tonic immobility, open field test, or even hen vocalization [73]. Utilizing behavior observation software could also provide more accurate and objective results for the MHT as well.

Visual inspection by trained personnel is one of the easiest and quickest ways to determine welfare in the field and is quite possibly the most widely used method. Starting with the beak, excessive trimming or poor beak health can prevent hens from eating feed [74]. Next, keel bone fractures are one of the most prevalent major injuries in laying hens, particularly in extensive environments, as hens will land on the keel when participating in vertical movements [75]. While this study found keel deviations in some birds, no broken keels were found, as the hen’s vertical movement was limited. Also, both keel bone and toe fractures can provide an indication of the hens that are deficient in calcium [76]. Observers will examine the hen’s footpad to determine if the conditions that they are being kept in are sanitary and clean, as dermatitis presents itself on the hen’s feet and can cause pain and distress [77]. Consequently, no bumblefoot was discovered during this study, indicating that the densities observed do not cause bumblefoot in white laying hens in the cages utilized. When examining the comb, an observer checks for damage caused by other hens, which can indicate high levels of aggression or frustration within the flock [78]. Finally, evaluating feather coverage can inform levels of feather pecking or lower levels of preening [78]. A lack of feathers can also cause skin lesions as feathers protect the hen’s skin [79]. In the present study, stocking density did not affect beak, keel, toe, foot, or comb scores at any measurement period. Furthermore, while keel bones were evaluated for deviations and breakages, no breakages were observed, only keel bone deviations. It was found that stocking density affected feather coverage, with the highest stocking densities consistently having the poorest feather scores and the lowest stocking densities having better scores. In general, feather coverage on laying hens decreases as the hen ages [80]. Therefore, it is important to ensure that this feather loss is not exacerbated by other factors. Several other studies found similar results in feather quality indicating that poorer feather coverage in high stocking densities is a common problem in the laying hen industry [27,28,29]. The lack of feather coverage in higher stocking densities may be indicative of greater incidence of feather pecking, as well as the hens rubbing against each other or the sides of the cage due to the decreased personal space during movements. Research on other housing systems indicates that while hens in cages are more protected from certain physical injuries (such as keel bone fractures, broken legs, and bumblefoot), the same hens often experience a greater level of feather loss, which is a result of aggressive feather pecking (particularly in high densities) manifested from hens frustration associated with movement restriction [81]. In the present study, hens in higher stocking densities had much poorer feather quality further indicating that movement restriction leads to greater aggression in hens. It appears from the results of this study that decreasing stocking density can improve feather coverage and quality in laying hens. However, this study did not identify any other areas where decreased stocking density improved.

4.3. Hen Health

Bone health is a critical component of laying hen welfare, particularly as housing environments become more extensive. Egg-laying hens have an incredible calcium and phosphorus demand on their bodies through the production of eggs [51]. If not managed correctly, this calcium demand can cause their bones to become more fragile and prone to fractures and breaking [82]. Furthermore, evidence exists connecting poor bone health to elevated stress and poorer welfare in laying hens [21,69]. Therefore, monitoring and maintaining bone health is critical for the proper welfare of laying hens. From the results of the present study, it appears that stocking density does not affect bone quality, particularly breaking strength, or bone density. However, this study utilized the tibia to conduct bone quality analysis, while other studies that utilized other bones found that higher stocking densities caused a weaker humerus and femur and left the tibia unchanged [30]. Jalal et al. [45] found no difference in bone ash between stocking densities, further ratifying the results of the present study. Kang et al. [14] discovered that brown egg layers reared in higher densities had lower bone mineral densities than those at lower densities; however, this study was performed with brown egg layers and in a cage-free environment. Perhaps the added freedom and structural stress from vertical movement in Kang’s study caused the weaker bones. In accordance with others, the present study found that stocking density did not cause a change in tibial quality although further research should be performed diversifying the bones utilized.

Gut health is also critically important in poultry production. In this study, several gut health parameters were analyzed to create a picture of overall gut health at the end of the study. Villus dimensions are a good indication of nutritional absorbance ability as greater surface area correlates to a higher number of absorptive cells [83,84]. Many researchers have shown that nutritional stress can cause villi in the gut to become atrophied and shrunken; however, it has also been found that other stressors, such as heat stress, can have these same effects [85,86,87,88]. While our study did find that hens under the highest stocking density had shorter villus length than hens in a lower density, these differences evaporated when comparing the area of the villi, indicating that increasing stocking density does not have a negative impact on the morphological function of the villi in the jejunum. It is also understood that many of the secretory cells in the intestines are located in the crypt and when the depth of the crypt is shortened, then the amount of these cells is lessened [83,84]. These goblet cells play an important role in the defense of the brush border against pathogens and nutrient absorption by secreting a protective layer of mucins, encoded by genes such as Mucin-2, which covers the brush villi of the intestinal tract [89,90,91]. Shallower crypts could indicate a potential for less protection and less goblet cells and lesser expression of Mucin-2 has been shown to alter inflammatory and metabolic pathways in mice and cause tumors in the intestinal tract [92]. The present study did not find that stocking density caused a change in crypt depth or Mucin-2 expression, thereby indicating that current stocking density levels may not adversely affect the morphological function of the intestinal tract for the laying cycle. Further research is needed as this study only explored these parameters after a typical cycle and did not explore a second cycle post molt, which, while uncommon, is still a practice that is used in the industry.

Assessing the genetic expression of certain proteins in the intestines can also give insight into the gut health of the chicken. In this study, the expression of the genes Claudin-1, Occludin, ZO-1, IL-1b, IL-10, and TNF-a were analyzed. Claudin-1 is a major contributor to tight junctions within the intestines and plays a role in their structural integrity [93,94]. A reduction in these tight junction proteins can cause a degradation of mucosal barrier function [95]. Occludin is an enzyme that is also located in tight junctions and functions to regulate their formation and perform maintenance on tight junctions [93,96,97]. Reducing the presence of Occludin has been shown to cause less complex tight junctions, disrupt the barrier function of the intestinal wall, and induce chronic inflammation, as well as many other ailments [98,99]. ZO-1, also known as Zona Occludin, plays a role in the structure of tight junctions, acting as a scaffolding protein and regulating the cytoskeleton in cells [100,101]. Research indicates that a reduction in ZO-1 will increase the permeability of the intestines, thereby causing a greater risk of disease [102]. This study did not find any differences in tight junction protein expression at the end of the study. This indicates that current stocking densities probably do not cause a reduction in intestinal structure or permeability. More research should be performed, specifically evaluating higher stocking densities to identify if extreme stocking densities will cause a reduction in structural gut health.

Cytokines modulate inflammatory and immunological functions in the body. IL-1b acts as a pro-inflammatory cytokine that plays a role in response to diseases and stress states and is typically upregulated during stress and disease [103]. The present study did not identify any change in the expression of IL-1b in the jejunum of laying hens, indicating that current stocking density levels do not illicit a stress response that causes an increase in IL-1b production. Conversely, IL-10 acts as an anti-inflammatory cytokine, repressing the action of the immune system when not needed [104]. However, higher levels of IL-10 can also hinder an animal to respond to disease [104]. The present study did find that hens in the highest stocking density had higher levels of IL-10 than hens in the lowest stocking density. An increase in IL-10 is typically seen as a positive, as it means that the immune system is not under stress; however, when IL-10 is increased, immune function is also decreased. Due to the depressed IL-10 levels, the immune system may not be able to deal with a disease challenge as well as in the lower densities. Past research indicates that increased IL-10 expression can be indicative of a depressed immune system [104]. Finally, the last cytokine that was measured in this study is tumor necrosis factor alpha (TNF-a). The function of TNF-a is to trigger a host of other inflammatory cytokines in response to a disease or other stressor [105,106,107]. The current study did not identify any changes in TNF-a expression, potentially indicating that the current stocking density used does not cause a negative inflammatory response in laying hens.

5. Conclusions

In summary, this study found that decreasing stocking density within the colony cage did improve egg production but did cause hens to lay smaller eggs. While the lowest stocking density had higher production, eggs from this density were categorized as a loss at a higher proportion than other treatments, possibly due to increased movements of the hens. Feed consumption, feed efficiency, mortality, and all physical egg quality parameters remained unaffected between treatments. This study also revealed that altering stocking density may affect some welfare and stress parameters. While corticosterone levels remained unaffected through most of the study, blood corticosterone concentrations were higher at the highest density at the end of the study. This may indicate that stress may be slowly increasing across the lifespan of the animal in the higher densities. It was also found that hens within higher stocking densities elicit a higher reactive fearfulness response (in the form of a Hansen’s test). This may be due to a higher level of anxiety or due to more interactions between birds extending the response. Oxidative stress pathways in the jejunum, however, were unaffected by the change in stocking density. While no bone parameters were affected by stocking density, this study found that villus height in the jejunum was lower in the highest density. Furthermore, no genes measured were found to be affected by stocking density except for IL-10, which was highest in the highest-density birds. Finally, while this study did not detect differences in most body quality measurements, feather coverage was affected with feather coverage decreasing as stocking density increased, perhaps due to an increase in cage abrasion. This paper aimed to evaluate the welfare and health of hens housed in various stocking densities, but it was in no way meant to be an exhaustive list of tests. Further research could be performed to further strengthen the understanding of fearfulness such as tonic immobility, vocalization, or the open field test. Furthermore, while a molt is rare in the industry, it is still a practice that is carried out. Research into the combined effects of molt and stocking density changes would provide more understanding of a major acute stressor on hens in different stocking densities. In conclusion, It appears raising hens in lower stocking densities can bring a handful of benefits to the egg layer production system. However, it appears that increasing density past 897 cm² does not provide extra benefits to production or welfare.