From November 2021 to June 2023, data was collected on 23 dairy sheep and 20 dairy goat farms in Switzerland. Sheep farms were visited three times whereas goat farms were visited two times. The first visit (V1) took place in the first 3 weeks after birth, the second visit (V2) 30 ± 2 (mean ± standard deviation) days after V1 and the third visit (V3; sheep farms only) 95 ± 5 days after V1 (Figure 1). Animals were furthermore tracked via the official Swiss national animal identification, registration and movement database (Animal Tracing Database; ATD) (32), until the age of 12 months for assessment of animal mortality and slaughter ages (see Section 2.3). All farmers were recruited by phone or letter. Participation was voluntary, and written consent was given by all participating farmers. Inclusion criteria were active dairy production and lambing times between September and April.

If the animals were sold to a fattening farm between V1 and V3, the subsequent data collection was conducted on the respective fattening farm. Fattening farms were categorized as farms to which lambs or goat kids were transported with the objective of fattening. The transfer had to be registered in the ATD. One dairy sheep farm, three dairy goat farms and one goat kid fattening farm had to be excluded from the study owing to insufficient data regarding animal-based indicators. This exclusion resulted in data from 22 dairy sheep farms, 17 dairy goat farms, eight lamb fattening farms and one goat kid fattening farm. The animal study was reviewed and approved by the Committee of Animal Experiments of the Canton of Zurich (Ethics Reference No: TG08_2021) and was in accordance with the Swiss guidelines for animal welfare.

An interview with the farmer was conducted either before or during V1, usually on site by the author HV. The interview included questions on general farm data (e.g., herd size, breeds, affiliation to a label; see Table 1) and data on birth management, colostrum phase, the rearing system, suckling systems, weaning age, fattening system and usual age at slaughter. If the answers contained ranges of values such as a range in milk feed allowance per animal, the average value was calculated. If any management changes relevant to the sampled animals were reported by the farmer during the subsequent farm visits, the information was adjusted accordingly.

On each dairy farm, 10% of the annually expected offspring (with a minimum number of 10 and a maximum number of 34 lambs or goat kids per farm) were selected at V1 for the animal-based data collection. On two sheep farms, two distinct sample groups were chosen to reflect two different management practices and thus more than 10% of the expected offspring were selected. Animals intended for herd replacement were excluded from the sample when this information was available. This selection resulted in an initial sample of 543 lambs and 247 goat kids. The number of animals declined across the subsequent visits (V2 and V3) owing to factors such as death, slaughter or missing data. The sample sizes analyzed were 539 lambs and 235 goat kids at V1, 476 lambs and 212 goat kids at V2, and 379 lambs at V3. No goat kids were assessed at V3 because most had been slaughtered by that time. The numbers can also be found in a table format in the Supplementary Table S1.

Statistical analysis and data visualization were conducted using R [Version 4.4.1; R Core Team, 2024 (34)]. Descriptive results of the animal-based indicators are provided in the Supplementary Table S2.

Fixed effects were defined and categorized based on the interview and individually assigned to reflect the individual animals’ experiences as follows: For “contact dam V1,” it was categorized if a lamb or goat kid was either “still with its dam” or “without its dam” at V1.

“Rearing system” was categorized based on the duration lambs or goat kids remained with their dams during the whole rearing period. If a lamb or goat kid was separated from the dam after the time allotted for colostrum ingestion (colostrum phase), it was assigned to “artificial” (ART) rearing. If a lamb or goat kid stayed with its dam until weaning, it was assigned to “mother-bound” (MB) rearing. In cases where the young initially stayed with its dam but was later separated during the suckling period for artificial rearing, it was assigned to “temporarily mother-bound” (TMB) rearing.

Because changes in milk feed allowance were common throughout the suckling period, the data collected from the interview was divided into two phases for analysis: the “beginning of the suckling period” (immediately after the colostrum phase) and the “end of the suckling period” (just before weaning). Suckling period refers to the full time that lambs were offered milk feed, regardless of whether they still were with their dam or not.

The two fixed effects “milk feed allowance beginning of suckling period” and “milk feed allowance end of suckling period” were defined based on the following criteria: the milk feed allowance was defined as “ad libitum” if milk feed was always available without restrictions on quantity or if it was provided via an automated system that lacked individual animal recognition and specific feeding times or if lambs and goat kids were suckled by their dams. Milk feed allowance was defined “temporarily ad libitum” if there were no restrictions on the quantity consumed per feeding but temporary limitations on access to milk throughout the day. This restriction was achieved either by removal of the artificial feeding device after all young stopped showing immediate interest or by separation of dams and young at night (as practiced by one dairy sheep and two dairy goat farms). If restrictions were imposed on both the frequency of access to milk feed and the amount offered, the milk feed allowance was defined as “restrictive.”

The fixed effect “weaned from milk V2” described if an animal was already weaned from milk (“yes” or “no”) at V2. “Place of weaning” described whether an animal was weaned on the birth farm or if weaning took place after transfer to a fattening farm. “Place of fattening” described whether a lamb or goat kid either remained on its birth farm for the full duration of fattening or if it was transferred and fattened on a fattening farm. The fixed effect “concentrate feed” categorized whether lambs and goat kids were offered any concentrate (“yes” or “no”) based on the definition of Suisse Bilanz (35) after V2.

As outcome variables, gamma globulin status (in g/dl at V1), health scores (healthy/morbid at V1, V2, V3), ADG (in g/day in the periods V1–V2 and V2–V3) and mortality in lambs (died/did not die) were analyzed with mixed models with “farm” as random effect. The model for gamma globulin status and health V1 combined both species. For all other models, the models were set up species specific because the management of lambs and goat kids differed substantially between species after V1. No models were calculated for goat kids for health at V3 or ADG V2–V3 because most goat kids had been slaughtered after V2. Data from one dairy goat farm (n = 12 goat kids) were excluded from the statistical analysis for health V2 and ADG V1–V2 owing to its unique management practices regarding the rearing compared with the other farms.

The mixed models analyzing gamma globulin status and ADG were calculated using the “lmer” function of the “lme4” package (36), whereas models for the health score were run using the “glmer” function of the “glme” package (37). Model assumptions were checked visually through graphical analysis of residuals (normality and homoscedasticidy). Each global model was tested against its null model to assess the overall explanatory value of the models (with p ≤ 0.05 as level of significance). If this global test was significant, dummy variables with sum contrasts were used for the fixed effects included in the models to evaluate the individual factor’s statistical influence on the outcome variable. The p-values were obtained by comparing the full model with models reduced by single factors, using the “pbkrtest” package (38). Model estimates and confidence intervals [upper/lower] were calculated with the “effect” package (39). The models that were run were:

On average, gamma globulin serum levels were lower in lambs (estimated mean [lower/ upper confidence interval]: 1.0 [0.77, 1.14] g/dl) than in goat kids (1.3 [1.14, 1.56] g/dl; p = 0.010; Table 5). The values of the samples ranged from 0.1 to 2.6 g/dL in lambs and from 0.1 to 3.5 g/dL in goat kids. The gamma globulin serum levels were higher in younger lambs and goat kids than in older ones (p = 0.005; Figure 3). Neither the health score at V1 nor the rearing system at V1 showed statistical influence on the gamma globulin serum levels.

At V1, 17.6% of lambs and 11.5% of goat kids showed signs of at least one health issue (Table 6), with umbilical disorders being most prevalent in lambs and fecal soiling or diarrhea in goat kids. At V2, 30.5% of lambs and 16.5% of goat kids were affected with the main issues being fecal soiling or diarrhea and orf-like skin lesions in lambs and orf-like skin lesions in goat kids. At V3, 22.2% of the lambs had health findings, of which respiratory quality was most predominant. Morbidity varied greatly across farms in both species. In lambs, farm morbidity ranged from 0 to 43.3% at V1, 0–83.3% at V2 and 0–53.3% at V3. In goat kids, farm morbidity ranged from 0 to 75.0% at V1 and from 0 to 100% at V2. At V2, MB goat kids had a better health status than ART goat kids (p = 0.004; Table 5). No other statistical effects of the models’ fixed factors were found on the health status at V1, V2 and V3 in lambs or on the health status at V1 in goat kids (Table 5).

The ADG of the lambs was 238 ± 86 g/day (range: 10 to 476 g/day) in the first phase of rearing (V1–V2) and 196 ± 63 g/day (range: 29 to 377 g/day) in the second phase of rearing (V2–V3). During the first phase of rearing, lambs that were fed milk feed ad libitum or temporarily ad libitum at the end of the suckling period had a higher ADG (268 [250/285] g/day and 240 [205/274] g/day) than restrictively fed lambs (213 [196/230] g/day; p = 0.041; Table 5; Figure 4). Lambs that were scored as healthy at V2 had a higher ADG than morbid ones (250 [239/261] vs. 218 [205/232] g/day; p < 0.001). Furthermore, lambs weaned on the dairy farms had a higher ADG (255 [243/267] g/day) than those weaned on fattening farms (198 [179/217] g/day; p = 0.003). No effect was found for the rearing system or the milk feed allowance at the start of the suckling period. During the second phase of rearing (V2–V3), TMB lambs had a higher ADG (248 [219/277] g/day) than ART (201 [179/223] g/day) and MB lambs (164 [130/199] g/day; p = 0.002; Figure 4). Lambs that were sold before weaning and then weaned on the fattening farm had a higher ADG during V2–V3 than those fattened on dairy farms or sold after weaning (235 [210/259], 211 [193/229], 154 [125/182] g/day, respectively; p = 0.002). No effect was found for the milk feed allowance at the end of the suckling period, the health at V3 and the provision of concentrate feed (Table 5). Over the entire rearing period (V1–V3), ad libitum-fed lambs had an on average higher ADG of 223 ± 50 g/day than lambs fed milk restrictively (207 ± 44 g/day) or temporarily ad libitum (193 ± 39 g/day). For the rearing system, TMB lambs had the comparatively highest ADG (216 ± 53 g day), followed by MB (216 ± 47 g/day) and ART (198 ± 36 g/day).

The ADG V1–V2 of the goat kids was 227 ± 62 g/day (range: 39 to 386 g/day). No difference was found in the ADG V1–V2 regarding the rearing system, the milk feed allowance at the beginning of the suckling period, the milk feed allowance at the end of the suckling period and the health status at V1 and V2 (Table 5).

Depending on the time the lambs and goat kids were kept with their dams, we grouped them into three rearing systems, namely, MB, TMB and ART. Systems using ART and MB rearing (with and without separation from the dam during certain times of day) have been described before (25, 41, 42) and are common rearing systems in Europe. In contrast, the only mention of TMB as rearing system outside of Switzerland (43) was described in the EFSA-report on welfare regarding livestock in Europe (44). In our study, around 25% of the lambs were reared TMB. Thus, TMB rearing seems to be a relevant system on Swiss small ruminant farms allowing prolonged contact of dam and young and the complete sale of the milk produced after separation of the young from its mother.

We found a large variety within these rearing systems regarding the duration of time lambs were kept with their mothers, milk feed provision (milk feed type, way of feeding, daily milk allowance, frequency of daily feeding), types of solid feed offered as well as age and weight at weaning and weaning strategies. Similar differences in artificially reared goat kids were reported in a survey conducted in 16 countries across three continents (28). With little research available on small ruminants in Central Europe, no definite best practice has been described in the literature for dairy farming of small ruminants. The situation of the dairy market, animal health issues or seasonal demands can lead to temporary changes in the rearing system and/or implementation of two or more rearing systems on the same farm in our study. Especially for goat kids, demand and prices for meat vary greatly between seasons and the carcass weight (45). Apparently, farmers of our study constantly evaluate the situation and adapt accordingly.

Another interesting aspect in our study was found in the fattening and marketing strategies of lambs compared to goat kids. Lambs from dairy sheep farms were frequently sold to fattening farms, and were commonly either slaughtered unweaned as suckling lambs or, more often, raised to a live weight of 40–50 kg. Their meat, which is in demand throughout the year (6), was mainly marketed via butcheries and bulk buyers. In contrast, goat kids were almost exclusively slaughtered at less than 12 weeks of age and often marketed directly, probably as a result of the lower, more seasonal demand for goat meat which peaks around Easter (6). Self-marketing makes farmers less dependent on the target goat kid slaughter weight of about 6.0 to 7.9 kg, which is enforced by bulk buyers with deviation being penalized (45). Unlike in countries such as France, where goat kids are commonly transported to specialized fattening farms within days of birth (24), this was the case only on one dairy goat farm in our study. However, this should be interpreted with caution. Dairy farms that fatten their lambs or goat kids on-site may be overrepresented because dairy farms relying in external fattening farms could not be included in this study if the associated fattening farm was not willing to participate.

We found lower gamma globulin serum levels in lambs (1.0 [0.77/1.14] g/dl) than in goat kids (1.3 [1.14/1.56] g/dl), which aligns with previous findings showing lower IgG in lambs (0.8 g/dL [(46)]) than in goat kids (1.2 g/dL [(17)]). The differences may be due to biological differences between the two species. While those previous studies measured IgG, the international standard for immune status, we measured gamma globulin due to financial constrictions. However, we consider the results comparable, as gamma globulin serum levels have been shown to closely correlate with IgG in equines (47, 48) as well as in lambs (49). Measurement of gamma globulin serum levels was shown to mostly produce slightly more conservative estimates as compared with IgG (47).

Adequate colostrum supply is crucial for both lambs (15) and goat kids (50) because immunoglobulin transfer occurs after birth (51). Thresholds of 1.5 g/dL of IgG for lambs and 1.2 g/dL of IgG for goat kids have been widely used to define failure of passive transfer (17, 42, 52, 53). The mean values of our results were above the threshold for goat kids and below the thresholds for lambs. However, most previous studies were conducted with meat breeds. Colostrum from dairy breeds tends to have. Lower IgG concentration (54). Thus, young from dairy breeds might generally have lower blood serum levels after the same amount of colostrum intake than those of meat breeds.

Gamma globulin serum levels varied widely in both species, ranging from 0.1 to 2.6 g/dL in lambs and 0.1 to 3.5 g/dL in goat kids and reflecting earlier findings [lambs: 0.1 to 3.8 g/dL (49), goat kids: 0.6 to 2.3 g/dL (55)]. This variance likely results from individual differences in colostrum quality (54), the amount ingested and absorbed by the lambs and goat kids (22) as well as a potentially the frequency of colostrum intake (56). No statistical effect of contact to the dam (MB vs. ART) at V1 was found in our study, likely because most farmers allowed their lambs or goat kids to suckle their dams for up to 5 days postpartum for ingestion of colostrum. Given the high variance in individual colostrum IgG, it may be advisable for farmers to assess the colostrum quality, for example by refractometry, and intervene if values are considered too low.

Gamma globulin serum levels in both lambs and goat kids were higher in younger animals and lower in older animals. This inverse relationship of gamma globulin serum levels and age is consistent with findings by Constant et al. (59), who observed a decline in IgG levels in goat kids during the first 3 weeks of life after an initial peak at 12 to 24 h. This decline occurs because ruminants do not produce their own immunoglobulins until later in life and can only absorb them from external sources during the first few days (15). Thus, the gamma globulin ingested from colostrum is gradually utilized and degraded over time (60).

We also expected lower gamma globulin serum levels in sick animals as has been found in other studies in lambs (19) and goat kids (18) and can be explained either by usage of gamma globulins during illness or increased susceptibility to illness due to low levels. However, we could not detect an effect of the health score at V1 on gamma globulin serum levels, which aligns with another study in which goat kids showed no correlation between health and immunoglobulin levels (22). Likely other factors such as good hygiene and low infection pressure may be able to mitigate any direct connection between gamma globulin serum levels and health. Another factor to consider is that potentially animals with very low health and gamma globulin died before our evaluation and a survival bias might be present. To fully evaluate the various influences, further research is required.

The overall average morbidity rates were 17.6% (V1), 30.5% (V2) and 22.2% (V3) in lambs and 11.5% (V1) and 16.5% (V2) in goat kids across the different visits. However, due to the nature of our study it was not feasible to assess if animals got sick between visits. In previous studies, a wide range of morbidity rates in lambs has been reported. Morbidity rates ranged from 1.6 to 86.6% in neonatal lambs (61), and a cumulative morbidity of 12.9% in lambs up to 3 months of age (62) and a preweaning morbidity of 27.3% in lambs (63) were found. Our results fall into that reported range. Our assessed morbidity rates in goat kids seem comparable or comparatively low to previous research [morbidity rates of 42.2% were described in goat kids from 0 to 30 days of age, dropping to 31.8% in goat kids aged 31 to 90 days and 26.0% in goat kids aged 91 to 180 days (64)]. The variance may be explained by the fact that some diseases such as orf or diarrhea can spread rapidly across the whole flock and thus affect many animals on a single farm while not affecting any animal on another farm.

The most common health issues in early life (V1) were umbilical disorders in lambs and fecal soiling or diarrhea in goat kids in our study. At V2, orf-like crusts and, in lambs, fecal soiling or diarrhea were most predominant. At V3, a reduction in respiratory quality was the main finding in lambs. Diarrhea and pneumonia were also reported by others as the main health issues of sick animals during the first 3 months of life (62, 65), whereas respiratory disease was found to increase with age and to be a significant cause of illness and death in feedlot lambs (23, 66). A good health score at V2 was positively related to ADG in lambs between V1 and V2, similar to the study by Lacasta et al. (23) where lambs with lung infections had a lower ADG than healthy lambs. Thus, promoting good health is important for farmers not just due to ethics and animal welfare but also out of economic considerations.

The overall mortality rate as calculated from the ATD was 11.9% in lambs and 6.1% in goat kids over the full course of our study. Previously found mortality rates ranged from 2% to up to 68.0% in lambs of various ages (12, 13, 61) and from 1.6% to up to 51% in goat kids (12, 14). Therefore, the mortality rates in lambs found in our study were in the expected range whereas mortality in goat kids was comparatively low. However, as opposed to our study, the previously mentioned studies included stillborn lambs and goat kids.

We found no statistical effects of the rearing system (in both lambs and goat kids), milk feed allowance, or place of fattening (in lambs only) on health scores at any of the visits, with the exception of the rearing system in goat kids at V2, where kids raised in the MB system showed better health scores than those in the ART system. Likely other factors such as infection pressure and general barn hygiene that were not considered in our analysis had an influence on the health outcomes. A potential influence of the place of fattening was expected because of stress and increased infection risk from transport and mixing of young animals from different farms (66–68). However, mortality was substantially higher in lambs transported to fattening farms before weaning (18.5%) than in those that remained on their birth farms (10.8%) or were transported to fattening farms after weaning (5.4%), suggesting a survival bias with sick or weak lambs possibly having died prior to health assessments, potentially masking any negative effects. However, this has to be interpreted with caution as 25% of the lambs sold to a fattening farm were fattened on only one fattening farm.

The ADG in lambs during the first phase of rearing (V1–V2: 238 ± 86 g/day) was in the range of other studies with lambs until 30 days of age [223 to 240 g/day (69, 70)], whereas the ADG of lambs during the second phase of rearing (V2–V3: 196 ± 63 g/day) was low compared with another study (245 g/day) in which lambs were fattened from 2 months old until a slaughter weight of 18 kg at about 143 days of age (71). This relatively low average is likely due to the fact that some farms fattened their lambs extensively. We found no statistical effect of the rearing system during the first phase of rearing, whereas TMB lambs had a higher ADG than MB and ART lambs during the second phase of rearing. This effect is difficult to explain. Results from previous studies regarding the effect of the rearing system are controversial. Some found a higher ADG in ART than in MB lambs (69), whereas others found a higher ADG in MB than in ART lambs (72). Hernández-Castellano et al. (27) found the highest ADG in MB lambs until the start of the weaning period (at 10 kg), but differences in ADG disappeared across treatment groups during weaning. Apparently, aspects other than the rearing system itself such as breed, season, number of milk feeding bouts allowed or composition of the milk feed might possibly influence the ADG. For example, the composition of cow’s milk differs greatly from that of sheep’s milk (e.g., in energy value) and might therefore cause a reduced ADG when offered as milk feed in lambs (73).

Consistent with previous findings (74), a high milk allowance (ad libitum or temporarily ad libitum) toward the end of suckling had a positive effect on the ADG in lambs during the first phase of rearing but no effect during the second phase. Santos et al. (31), assessing lambs after weaning at 15 kg of weight, also found a higher ADG in ad libitum fed lambs than in those with restricted milk allowance.

Lambs that were transported to fattening farms before weaning had a lower ADG in the first rearing phase than those that remained on the dairy farms in our study. However, between V2 and V3, these lambs had an ADG comparable to that of lambs remaining on their dairy farms and much higher than that of lambs sold to fattening farms after weaning. The lower ADG in lambs that were transported to fattening farms before weaning is likely due to the previously discussed stress and increased infection risk from transport and mixing of young animals from different farms (66–68). Their high ADG in the second phase of rearing may again be due to the survival bias in lambs not transported to a fattening farm before weaning as discussed in the previous chapter. In comparison, the rather low ADG of the lambs being transferred to the fattening farms after weaning might then be the output of the more recent stress of the transfer. As Becker et al. (67) suggested in calves, certain management adjustments could help to alleviate the stress and health challenges faced on fattening farms, but the implications of this form of fattening and the optimal time of transfer to a fattening farm of these lambs should be investigated in more depth.

The ADG of 227 ± 62 g/day (range: 39 to 386 g/day) of the goat kids in our study was comparatively higher than in other studies [136 to 158 g/day (75, 76)]. Even though the ADG varied considerably between farms, we found no statistical effect of the rearing system, the milk feed allowance or the health status on the ADG. As in lambs, results from other studies are controversial. Some studies also found no differences in the ADG between goat kids in different rearing systems (75, 77), whereas others found a higher ADG in MB than in artificially fed goat kids (29). An explanation for our findings could be differences between breeds, quality of milk replacer or the fact that most of the goat kids in this study were fed with generous amounts of milk feed (ad libitum or temporarily ad libitum) toward the end of their suckling period, mitigating possible differences in milk feed amount between MB and artificially fed goat kids that might have been present in other studies.

To conclude on the rearing systems for lambs and goat kids on dairy farms, we must keep in mind that welfare encompasses not only good health and the absence of pain and suffering but also the ability to live reasonably natural lives and to experience positive affective states (78). In natural conditions, lambs and goat kids strongly bond to their mothers (16) and are typically suckled by their dams for 100 to 180 days until natural weaning (79). Lambs suckle their dams up to 36 times per day in the first few days of life and reduce the frequency to about 14 times per day at 6 to 7 weeks of age (80). This clearly differs from the situation the lambs and goat kids experienced on the dairy farms of our study, especially if lambs and goat kids were raised artificially. For example, if a restriction on milk feed amount or time was applied, the number of feeding bouts in our study was considerably low compared with natural conditions. The inability to behave naturally can lead to mutual suckling, and such behavioral disorder is considered a sign of impaired welfare (80). Although not further assessed, we observed mutual suckling on several occasions in ART lambs. Also, separation from the dam (81) or transport at a young age (80) can cause distress apparently without directly statistically affecting the ADG or health (but at least mortality in our study). Therefore, further studies with focus on behavioral indicators such as mutual suckling, playing time and vocalization (72, 81) are needed to comprehensively evaluate animal welfare in the different rearing and fattening systems in these animals.